Labor Market

Highlights The risk asset friendly outcomes in the French and South Korean elections are the latest examples of fading geopolitical risk, and we expect that to continue over the remainder of 2017. Although it has been well over a year since the last 10% pullback, the U.S. equity market is not "due" for a correction. For many investors, the drop in commodity prices has replaced geopolitics as the most likely cause of the next equity market correction. What is Dr. Copper's diagnosis? We re-examine our Yield and Protector portfolios to find out which assets will hold up best if there is a correction. Many investors cite the monthly report on average hourly earnings as evidence that the Fed has it wrong on the economy and the labor market. We disagree. Feature U.S. stock prices remain within striking distance of their all-time highs and many investors continue to worry about the next correction. The risk asset friendly outcomes in the French and South Korean elections are the latest examples of fading geopolitical risk, and we expect that to continue over the remainder of 2017. The market has all but ignored the recent political turmoil in Washington. For many investors, the drop in commodity prices has replaced geopolitics as the most likely cause of the next equity market correction, while others note that it's been more than 15 months since the last 10%+ correction and that we are "due" for one. But is Dr. Copper still a reliable indicator of equity market tops? And if a correction is at hand, which assets would hold up best on the way down? We also review yet another disconnect between the Fed and the market: average hourly earnings. Geopolitical Risk Continues To Fade As A Market Concern Emmanuel Macron's victory was a resounding one as French voters rejected Le Pen's anti-Europe message in last week's election. Removing the possibility of a French President that is dedicated to exiting the eurozone is obviously positive for European stocks and investor risk appetite the world over. Next up are the two rounds of legislative elections in June. Polls are sparse, but they support the view that Macron's En Marche and the center-right Les Republicains will capture the vast majority of seats in the legislature. A Macron presidency supported by Les Republicains in the National Assembly would be a bullish outcome for investors, according to our geopolitical strategists. On the international stage - where the president has few constraints - France will be led by a committed Europhile willing to push Germany towards a more proactive policy. On the domestic stage - where the National Assembly dominates - Macron's cautiously pro-growth agenda will be pushed further to the right by Les Republicains. Such an election outcome would make possible the passage of genuine structural reforms that would suppress wage growth and make French exports more competitive. The presidential election result in South Korea last week was exactly what the market expected, and should help to reduce tensions on the Korean peninsula. For now, the situation in Washington around President Trump's firing of FBI Director Comey has not had a major impact on markets. If the Democrats win the House of Representatives in 2018, our geopolitical team believes that impeachment proceedings will begin against Trump. On one hand, this means that polarization in the U.S. is about to reach record-high levels. On the other, it should motivate the GOP to get tax reform done before it is too late. Bottom Line: Investors may be shocked into pricing greater odds of Euro Area dissolution when Italy comes back into focus, but that is a risk for 2018. We expect market-friendly policies emerging from Washington this year, although the Comey affair highlights that the road will be anything but smooth. Corrections And Pullbacks In Context Geopolitical risk appear to have faded for now, but with U.S. equities at or close to all-time highs, talk of a correction is hard to avoid. We continue to favor stocks over bonds this year and suggest that any sell-off in equities will be bought not sold. A hard landing in China, major disappointment on the Trump legislative agenda, a prolonged spell of weakness in the U.S. economic data1, and an overly aggressive Fed in 2017 may all serve as catalysts for a pullback. Above average PE ratios and measures of market volatility that are at cycle lows have only added to the chorus of those saying we are "due" for a correction. History suggests otherwise. From the end of WWII through 2009, the S&P 500 has experienced, on average, two 10% corrections and 10 corrections of 5% of more during equity bull markets. Since the start of the current bull market in March 2009 we've had 22 pullbacks of 5% or more and six corrections of more than 10% (using market closing prices) Table 1. This suggests that the market has seen its fair share of pullbacks and corrections since 2009, and isn't really "due". Chart 1 takes a different approach, but reaches the same conclusion. At 15 months (325 days) since the end of the last 10% correction, the current bull market is right of the middle of the pack of all bull markets since 1932. Table 1Six S&P 500 Corrections Of 10% Or More Since March 2009: We're Not "Due" Still Awaiting The Next Pullback Still Awaiting The Next Pullback Chart 1Current Equity Bull Market Is Not Long In The Tooth Still Awaiting The Next Pullback Still Awaiting The Next Pullback Our view remains that any pullback in U.S. equities will be bought, not sold, and we favor stocks over bonds in 2017. There are few notable imbalances in the U.S. or global economies and we see an acceleration in both over the remainder of 2017. The Fed will raise rates gradually this year, and there is general agreement between the Fed and the market on the pace of hikes at least for 2017. The Fed and the market remain far apart on hikes in 2018. Our view of the economy and labor market suggests that the market will ultimately move toward the Fed's view. The U.S. corporate earnings outlook remains solid, after a very good Q1 earnings season and favorable guidance for Q2 2017 and beyond. Bottom Line: Equity pullbacks - even during bull markets - are normal and healthy. We do not believe that the market is especially "overdue" for a pullback, but when the inevitable pullback or correction occurs, we expect that investors will take the opportunity to add to equity positions and not turn the pullback into a bear market. Dr. Copper? Chart 2Metals Prices Are Rolling Over...##BR##But Is It A Signal? Metals Prices Are Rolling Over... But Is It A Signal? Metals Prices Are Rolling Over... But Is It A Signal? The recent setback in the commodity pits has added to investor angst regarding global growth momentum. The LMEX base metals index is up almost 20% on a year-ago basis, but has fallen by 8% since February (Chart 2). From their respective peaks earlier this year, zinc and copper are down about 10%, nickel has dropped by 22% and iron ore has lost almost half of its value. Is the venerable "Dr. Copper" sending an important warning about world growth? Some of our global leading economic indicators have edged lower this year, as we have discussed in recent Weekly Reports. Nonetheless, the decline in base metals prices likely has more to do with other factors, such as an unwinding of the surge in speculative demand that immediately followed the U.S. election last autumn. Speculators may be disappointed by the lack of progress on Republican promises to cut taxes and boost infrastructure spending. The main story for base metals demand and prices, however, is the Chinese real estate sector. China accounts for roughly 50% of world consumption for each of the major metals. The Chinese authorities are trying to cool the property market and transition to a more consumer spending-oriented economy, thereby reducing the dependence on exports, capital spending and real estate as growth drivers. Fiscal policy tightened last year and new regulations were introduced to limit housing speculation. The effect of policy tightening can be seen in our Credit and Fiscal Spending Impulse indicator, which has been softening since mid-2016 (Chart 3). The economy held up well last year, but the policy adjustment resulted in a peaking of the PMI at year-end. Growth in housing starts also appears to be rolling over (annual growth is shown on a 12-month moving-average basis in Chart 4 because of the extreme volatility in the series). Both the PMI and housing starts are correlated with commodity prices. Chart 3China is The Main Story##BR##For Base Metals Demand China is The Main Story For Base Metals Demand China is The Main Story For Base Metals Demand Chart 4Direct Fiscal Spending And Infrastructure##BR##Have Picked Up Recently Direct Fiscal Spending And Infrastructure Have Picked Up Recently Direct Fiscal Spending And Infrastructure Have Picked Up Recently The good news is that BCA's China Investment Strategy service does not expect a major downshift in Chinese real GDP growth this year, which means that commodity import demand should rebound: Chart 5Dr. Copper Is Not Signaling##BR##A Slowdown in Global Growth Dr. Copper Is Not Signaling A Slowdown in Global Growth Dr. Copper Is Not Signaling A Slowdown in Global Growth There is no incentive for the authorities to crunch the economy given that consumer price inflation is still low and the surge in producer price inflation appears to have peaked. Monetary conditions have tightened a little in recent months, but overall conditions are not restrictive. Moreover, both direct fiscal spending and infrastructure investment have picked up noticeably in recent months (Chart 4). Export growth will continue to accelerate based on our model (not shown). The upturn in the profit cycle and firming output prices should boost capital spending. Robust demand will ensure that housing construction will continue to grow at a healthy pace. Households' home-buying intentions jumped to an all-time high last quarter. Tighter housing policies in major cities will prevent a massive boom, but this will not short-circuit the recovery in housing construction. This all adds up to a fairly benign outlook for base metals. Our commodity strategists do not see the conditions for a major bull or bear phase on a 6-12 month horizon. Within commodity portfolios, they recommend a benchmark allocation to base metals, an underweight in agricultural products and an overweight in oil. We intend to update our view on oil prices in the May 22, 2017 edition of this report. Bottom Line: From a broader perspective, our key message is that "Dr. Copper" is not signaling that global growth will soften significantly this year. Chart 5 highlights that the LMEX base metals index has a high positive correlation with the U.S. stock-to-bond total return ratio on a daily change basis. However, in terms of trends and turning points, base metals are far from a reliable indicator for the stock-to-bond ratio. Where To Hide In A Stock Market Correction Over the past several years, BCA's U.S. Investment Strategy service has periodically recommended that investors add a variety of investments as portfolio "insurance" to help guard against the possibility of a material correction in equities. More recently, we have highlighted two specific forms of insurance: our yield and protector portfolios. We last discussed the protector portfolio in the October 17, 2016 and November 7, 2016 Weekly Reports2, and in today's report we revisit the issue by comparing both portfolios to a more common form of insurance: shifting from cyclical to defensive stocks within an equity allocation. Charts 6, 7, and 8 show a breakdown of the relative performance of S&P 500 defensives along with our yield and protector portfolios. Panels 2 and 3 of Charts 6, 7 and 8 present the rolling 1-year beta and alpha of each strategy vs. the S&P 500. Here, we present alpha as the difference between the actual year-over-year excess return of the portfolio (vs. short-term Treasury bills) and what would have been expected given the portfolio's beta. This measure is sometimes referred to as "Jensen's alpha". Chart 6A Modestly Low-Beta Option A Modestly Low-Beta Option A Modestly Low-Beta Option Chart 7A Lower Beta Than Defensives A Lower Beta Than Defensives A Lower Beta Than Defensives Chart 8A Negative Beta, And Positive Alpha A Negative Beta, And Positive Alpha A Negative Beta, And Positive Alpha There are several noteworthy observations from the charts: Based on the historical beta of the three portfolios vs. the S&P 500, defensive stocks are the most correlated with the overall equity market. Our protector portfolio has a negative correlation to the broad market, and our yield portfolio is somewhere in between, with a positive but relatively low beta. This is consistent with the equity composition of the three portfolios (shown in Table 2); with our protector portfolio composed entirely of non-equity assets. Table 2A Breakdown Of Three##BR##Portfolio Insurance Options Still Awaiting The Next Pullback Still Awaiting The Next Pullback After accounting for their lower beta, all three portfolios have tended to outperform the S&P in risk-adjusted terms since the onset of the global economic recovery. But this outperformance has been more significant for our yield and protector portfolios: the top panel of Charts 7 and 8 highlight that both portfolios have generated essentially the same return as equities have since the end of the recession (since the relative profile has been flat), despite exhibiting considerably less volatility than stocks. All three portfolios have experienced a relative decline vs. the S&P 500 since the election, but this has largely occurred due to passive rather than active underperformance. In other words, they have underperformed due to a failure to keep up with the S&P 500 rather than because of losses in absolute terms. There are two important conclusions from Charts 6, 7 and 8 for U.S. multi-asset investors. First, the lower beta of our yield and protector portfolios compared with S&P defensives means that the former represent a better insurance bet against a sell-off in the equity market than the latter. Second, the persistently positive volatility-adjusted returns for our insurance portfolios highlights an investor preference for these assets over the past few years, which is likely to persist over the coming 6-12 months. But investors should also recognize that this preference could eventually be subject to a reversal if the long-term economic outlook significantly improves, an event that could be catalyzed either by organic economic developments or policy decisions by the Trump administration. For now, our investment bias towards equities over government bonds makes us less inclined to favor a low beta position within a balanced portfolio. But our analysis suggests that clients who anticipate the need for portfolio insurance over the coming year should favor our yield and protector portfolios over a defensive sector allocation within an equity portfolio, and we are likely to recommend an allocation to these portfolios for all clients were we to see any material progression towards the sell-off triggers that we identified earlier in the report. Bottom Line: Investors seeking some protection against a potential equity market sell-off should favor our yield and protector portfolios over defensive sector positioning. We do not currently recommend these portfolios for all clients, but we are likely to do so if our key sell-off trigger "red lines" are breached. What's Up With Wage Growth? On the surface, the April jobs report-released in early May seemed to send mixed signals to investors and the Fed about the health of the labor market3. Our view remains that the economy is growing fast enough to tighten the labor market, push up wages and ultimately inflation, which will lead the Fed to raise rates twice more in 2017. But even though the economy is very close to full employment and the output gap has nearly closed, patience is required. Although it's a close call, the next hike is likely to come next month. Markets remain somewhat skeptical of this view, and have only priced in 39 bps of tightening by the end of the year, and have not yet fully priced in a June rate hike. The lack of wage growth (up just 2.5% year-over-year in April according to average hourly earnings (AHE)) remains a key source of the market's skepticism about the pace and timing of Fed rate hikes. Many investors cite the monthly report on average hourly earnings as evidence that the Fed has it wrong on the economy and the labor market. Does the Fed see something the market does not? Or is it the other way around? Markets tend to focus on data that are timely. That requirement certainly fits the AHE. The monthly wage measure is the most timely data point on labor compensation. While timeliness is an important factor when assessing the health of the labor market, it is also critically important to watch what the Fed watches. Investors should note that the AHE data is only one of at least four measures of labor compensation the Fed mentions in its Semi Annual Monetary Report to Congress. Since Fed Chair Yellen took office in 2014, the Fed has specifically referenced (and charted together) three measures of labor compensation in the report: Average hourly earnings Employment Cost Index and Compensation per Hour in the nonfarm business sector, and Chart 9The Fed Tracks All Four Of##BR##These Compensation Measures The Fed Tracks All Four Of These Compensation Measures The Fed Tracks All Four Of These Compensation Measures The Atlanta Fed's Wage Tracker was mentioned in the June 2016 Monetary Policy Report, and the Fed added it to the chart of the other three metrics in the most recent report, released in February 2017. As Chart 9 shows, all have moved higher in recent years, although it is clear that AHE has lagged the others. Given the attention it receives in the financial news media on and just after "Employment Friday" each month, it may surprise investors to learn that neither AHE nor wages were directly mentioned in any of the FOMC statements since Yellen took charge. However, wage growth (or lack thereof) has been a topic of discussion at all but a few of the 13 post FOMC press conferences Yellen has held. When asked about wages, she is careful to note that the Fed watches a wide range of indicators of labor compensation, but has lamented the lack of progress on wages. In her most recent press conference, Yellen noted that "I would describe some measures of wage growth as having moved up some. Some measures haven't moved up, but there's some evidence that wage growth is gradually moving up, which is also suggestive of a strengthening labor market." Average hourly earnings are routinely mentioned in the FOMC minutes, but only alongside mentions of the other metrics noted above. On balance, average hourly earnings are viewed by the Fed - and therefore should be viewed by the market - as one of several indicators of the health of the labor market, but not the only indicator. Chart 10 shows that only a third of industries have seen an acceleration in wage increases over the past year, which supports the market's view that the economy is not growing quickly enough to push up wages and inflation. A recent report by the Kansas City Fed4 takes a different view. Using a bottom-up approach, the author points out that only a few industries (mostly in the goods producing sector of the economy) have accounted for much of the rise in wages, notably manufacturing, construction and wholesale trade. Financial services, retail trade, professional and business services and leisure and hospitality - all service sector industries - have been the laggards. The study done by the economists at the Kansas City Fed shows that although earnings growth has lagged in those more service-oriented industries since 2015, hours worked have seen faster growth than in the mainly goods producing sector (chart not shown). This suggests to the author - and we concur - that labor demand has been strong in the past few years in areas that have not seen much wage growth. As the labor market continues to tighten, wages in these industries may accelerate, but patience may be required. Chart 11 shows that it takes two to three years after a bottom in the output gap for a decisive turn higher in ECI or AHE. While this cycle has seen a more shallow recovery - especially in AHE - both have moved higher since the output gap bottomed out in 2009/2010. Chart 10Only 33% Of Industries Have Seen##BR##Wage Acceleration Over The Past 12 Months Only 33% Of Industries Have Seen Wage Acceleration Over The Past 12 Months Only 33% Of Industries Have Seen Wage Acceleration Over The Past 12 Months Chart 11Measures Of Labor Compensation Move##BR##Higher After Output Gap Bottoms Out Measures Of Labor Compensation Move Higher After Output Gap Bottoms Out Measures Of Labor Compensation Move Higher After Output Gap Bottoms Out Bottom Line: Investors are always wise to watch what the Fed watches. The evolution of wage growth will be critical to FOMC policymakers, because a clear acceleration will confirm that the economy is truly at full employment and, thus, at risk of overheating. We do not expect a surge in wages, but a steady upward trend will keep the Fed on a gradual tightening path. John Canally, CFA, Senior Vice President U.S. Investment Strategy johnc@bcaresearch.com Mark McClellan, Senior Vice President The Bank Credit Analyst markm@bcaresearch.com Jonathan LaBerge, CFA, Vice President Special Reports jonathanl@bcaresearch.com 1 Please see U.S. Investment Strategy Weekly Report "Growth, Inflation And The Fed", dated May 8, 2017, available at usis.bcaresearch.com. 2 Please see U.S. Investment Strategy Weekly Reports "Portfolio Insurance: What, How, When?", dated October 17, 2016 and "Policy, Polls, Probability", dated November 7, 2016, both available at usis.bcaresearch.com. 3 Please see U.S. Investment Strategy Weekly Report "Growth, Inflation And The Fed" dated May 8, 2017, available at usis.bcaresearch.com. 4 See "Wage Leaders and Laggards; Decomposing the Growth in Average Hourly Earnings" The Macro Bulletin, February 15, 2017; Federal Reserve Bank of Kansas City.

Highlights The U.S. unemployment rate stands 0.1 points below the FOMC's year-end projection and 0.4 points below its estimate of NAIRU. If the unemployment rate keeps falling, it will have nowhere to go but up - and the U.S. has never been able to avoid a recession whenever the unemployment rate has risen by more than one-third of a percentage point. So far the FOMC has failed in its efforts to tighten monetary policy. U.S. financial conditions have actually eased sharply since the Fed resumed hiking rates in December. The Fed will turn more hawkish over the coming months. Stay short the January 2018 fed funds futures contract and position for a stronger dollar. What happens in the euro area has become increasingly irrelevant for what happens to EUR/USD. Even if the ECB raises rates somewhat more rapidly than expected, this will be largely counterbalanced by hawkish actions by the Fed. Investors should stay cyclically overweight global equities, but be prepared to pare back exposure next summer. Feature Beware Of Full Employment Chart 1Recoveries Usually Lose Steam##br## WhenThe Unemployment Rate Falls Below NAIRU Recoveries Usually Lose Steam WhenThe Unemployment Rate Falls Below NAIRU Recoveries Usually Lose Steam WhenThe Unemployment Rate Falls Below NAIRU After eclipsing 10% in 2009, the U.S. unemployment rate fell to 4.4% in April, 0.1 points below the median end-2017 dot in the Fed's Summary of Economic Projections, and 0.4 points below the FOMC's estimate of NAIRU.1 The fact that most Americans who want to work are able to find jobs is obviously a good thing. However, today's increasingly tight labor market does have a dark side: As Chart 1 illustrates, recoveries have tended to run out of steam whenever the unemployment rate has fallen below its full employment level. Two points about the unemployment rate are worth keeping in mind: The unemployment rate has rarely been stable over time; usually, it is either rising or falling. The former tends to occur very quickly, while the latter is more drawn out. The unemployment rate displays momentum over short horizons, but is "mean-reverting" over the long haul (Chart 2).2 Since there is a limit to how low the unemployment rate can go, periods when it is below its full employment level typically do not last long. This is confirmed by Chart 3, which shows that there is a clear positive correlation between the degree of labor market slack and the onset of the next recession: High slack means that a recession is usually far away, whereas low slack means that a downturn is approaching. And it doesn't take much of an increase in the unemployment rate to sow the seeds for another recession - the U.S. has never escaped a recession in the postwar period whenever the three-month moving average of the unemployment rate has risen by a mere one-third of a percentage point (Chart 4). Chart 2The Unemployment Rate Is Mean-Reverting Over The Long Haul, But Displays Momentum In The Short Term The Fed's Dilemma The Fed's Dilemma Chart 3The Degree Of Labor Market Slack And The Onset Of The Next Recession: A Clear Positive Correlation The Fed's Dilemma The Fed's Dilemma Chart 4What Goes Down Must Come Up? What Goes Down Must Come Up? What Goes Down Must Come Up? Rising unemployment tends to generate all sorts of vicious cycles. When someone loses their job, they spend less. The resulting decline in aggregate demand forces firms to lay off workers, leading to even less spending throughout the economy. A weaker economy also makes it more difficult for borrowers to pay back loans, causing them to pare back spending. Falling asset prices only serve to exacerbate this problem. Threading The Needle Today's low unemployment rate puts the Federal Reserve in a bind. On the one hand, if the Fed raises rates too quickly, this could precipitate exactly the sort of downturn that it is trying to avoid. On the other hand, if the Fed fails to raise rates quickly enough, this could cause the economy to overheat. This, in turn, may force the Fed to raise rates aggressively - something that would destabilize both the economy and financial markets. The hope is that the Fed succeeds in threading the needle to ensure that the economy achieves a soft landing. There are some reasons to be optimistic about such an outcome, but also several reasons to be pessimistic. On the optimistic side, inflation expectations remain well anchored. This means that an overheated economy is unlikely to produce a powerful price-cost spiral such as the one that broke out in the 1970s. This limits the risk that the Fed will be forced to raise rates dramatically. The real economy is also not suffering from the sort of clear-cut imbalances that plagued the late innings of the last two business cycles - a massive capex overhang in the late 1990s, and an even larger housing overhang in the years leading up to the Global Financial Crisis. Private debt levels have also fallen as a share of GDP for most of the recovery, unlike in past cycles (Chart 5). On the pessimistic side, uncertainty about the level of the neutral rate - the interest rate consistent with full employment and stable inflation - will make it difficult for the Fed to calibrate monetary policy in a way that ensures a soft landing. It typically takes 12-to-18 months for changes in monetary conditions to fully make their way through the economy. Thus, if the Fed does end up either too far behind or too far ahead of the curve in normalizing monetary policy, it may not realize this until it's too late. Structurally slower potential GDP growth could also complicate matters. The Congressional Budget Office estimates that real potential GDP growth will average only 1.8% over the next 10 years, compared to 3.1% between 1980 and 2007 (Chart 6). Today's equity valuations are arguably pricing in faster GDP growth. Should growth settle below 2% - a rate that has often been associated with stall speed - risk assets could suffer, complicating the Fed's efforts in achieving a soft landing. Chart 5The Economy Is Not Showing ##br##Clear-Cut Signs Of Imbalances The Economy Is Not Showing Clear-Cut Signs Of Imbalances The Economy Is Not Showing Clear-Cut Signs Of Imbalances Chart 6Potential GDP Growth Is Not ##br##What It Used To Be Potential GDP Growth Is Not What It Used To Be Potential GDP Growth Is Not What It Used To Be The Fed's Choice Given the choice between erring on the side of raising rates too slowly or too quickly, the Fed has opted for the former. This is a quantitative statement, not a qualitative one. Chart 7 shows that U.S. financial conditions have eased considerably since the Fed resumed raising rates last December, thanks to a weaker dollar, tighter credit spreads, and a soaring stock market. If the whole point of hiking rates is to tighten financial conditions, then the Fed has not done enough. Worries that the headline unemployment rate may understate the true amount of labor market slack partly explain the Fed's angst in raising rates as quickly as it has in past cycles. While the headline rate has fallen back to its 2007 low, the broader U-6 unemployment rate - which incorporates people who are out of the labor market but claim to want a job, as well as those who are working part-time for economic reasons - is still 0.7 points above it. Likewise, the employment-to-population ratio for prime-age workers (ages 25-to-54) is 1.7 points below its pre-recession levels. The "quits rate" - a good measure of labor market confidence - also remains a notch below its pre-recession peak. Perhaps most glaringly, the median duration of unemployment has only fallen back to 10.2 weeks, which is still close to the high of the previous cycle (Chart 8). Chart 7Financial Conditions Have Been Easing Financial Conditions Have Been Easing Financial Conditions Have Been Easing Chart 8Headline Unemployment Rate ##br##Back To 2007 Levels, But Other ##br##Measures Still Point To Slack Headline Unemployment Rate Back To 2007 Levels, But Other Measures Still Point To Slack Headline Unemployment Rate Back To 2007 Levels, But Other Measures Still Point To Slack Each of these factoids has a counterargument: The elevated share of involuntary part-time workers may be partly due to the effects of Obamacare, which has made it burdensome for companies to add full-time workers to the payrolls;3 the low quits rate and the high median length of unemployment may reflect the aging of the population as well as lower gross job creation (Chart 9); and automation, globalization, and low-skilled immigration may have depressed real wages for less-educated workers, causing them to abandon the labor market (Chart 10). Nevertheless, with core inflation still below the Fed's 2% target, it is not hard to see why the Fed has elected to take a "go slow" approach so far. Chart 9The Labor Market Has Become Less Dynamic The Labor Market Has Become Less Dynamic The Labor Market Has Become Less Dynamic Chart 10Less-Educated Men Are Fleeing The Labor Market The Fed's Dilemma The Fed's Dilemma The Hawks Spread Their Wings That may be changing, however. The growth in nominal unit labor costs has already surpassed 2% and is close to the peaks reached in 2000 and 2007 (Chart 11). Most other measures of wage growth remain in a clear uptrend (Chart 12). If GDP growth accelerates over the remainder of the year, as we expect, the Fed will pursue a more aggressive tightening path than what the market is currently discounting. Chart 11Unit Labor Cost Inflation Close To Past Peaks Unit Labor Cost Inflation Close To Past Peaks Unit Labor Cost Inflation Close To Past Peaks Chart 12Most Measures Of Wage Growth Are In An Uptrend Most Measures Of Wage Growth Are In An Uptrend Most Measures Of Wage Growth Are In An Uptrend Recent communications from the Fed have revealed an increasingly hawkish bias. The latest Fed statement downplayed the slowdown in Q1 as "transitory." This follows Chair Yellen's comment that "waiting too long to remove accommodation would be unwise, potentially requiring the FOMC to eventually raise rates rapidly, which could risk disrupting financial markets and pushing the economy into recession."4 Investment Conclusions Higher U.S. rate expectations should give the dollar a boost (Chart 13). We do not agree with the often-heard argument that the actions of foreign central banks will materially weaken the dollar. Consider the case of the ECB. There has been much speculation that the ECB will phase out some of its emergency measures. That may well happen, but even if it does, a full-fledged hiking cycle is nowhere on the horizon. According to a recent ECB study, the rate of labor underutilization still stands at 18% in the euro area, 3.5 points higher than in 2008 (Chart 14).5 Stripping out Germany, the rate of underutilization would be seven points higher (Chart 15). It is still too early for Mario Draghi to begin removing monetary accommodation in a concerted manner. Chart 13Higher U.S. Rate Expectations ##br##Should Give The Dollar A Boost Higher U.S. Rate Expectations Should Give The Dollar A Boost Higher U.S. Rate Expectations Should Give The Dollar A Boost Chart 14Labor Market Slack In The Euro Area Remains High... The Fed's Dilemma The Fed's Dilemma Chart 15...Especially Outside Of Germany The Fed's Dilemma The Fed's Dilemma Moreover, anything the ECB does which inadvertently leads to a stronger euro will likely be matched by offsetting hawkish actions by the Fed. Remember that the Fed needs to tighten financial conditions in order to prevent the unemployment rate from falling so much that it has nowhere to go but back up. A weaker dollar runs contrary to that strategy. The argument above can be applied more broadly. The euro rallied in the lead-up to the French election on the now-realized hope that Emmanuel Macron would prevail. Put aside the fact that Macron's platform calls for cutting the budget deficit from 3.2% of GDP this year to 1% of GDP in 2022 - something which, all things equal, would lead to less monetary tightening and a correspondingly weaker euro. Even if Macron's victory somehow did manage to allow the ECB to raise rates earlier than it would have otherwise, it is hard to believe that this would not influence the pace of Fed rate hikes. U.S. financial conditions could tighten through some combination of higher rates and/or a stronger dollar. The only way the Fed could engineer a tightening in financial conditions while the trade-weighted dollar still weakened would be to jack up interest rates by an inordinate amount. However, this outcome would require that other central banks raise rates even more. That's not going to happen. Stay short EUR/USD. We think the euro will reach parity against the dollar later this year. Where does this leave equities? So long as global growth remains solid and corporate earnings are in an uptrend, the path of least resistance for stocks is up. However, the risk is that the Fed overplays its hand and ultimately tightens monetary policy too much. This could lead to a broad-based global slowdown towards the end of 2018. Investors should stay cyclically overweight global equities, but be prepared to pare back exposure next summer. Peter Berezin, Chief Global Strategist Global Investment Strategy peterb@bcaresearch.com 1 The Non-Accelerating Inflation Rate of Unemployment (NAIRU) is the unemployment rate consistent with stable inflation. 2 An Ordinary Least Squares (OLS) regression using monthly data between 1960 and 2017 shows that the change in the unemployment rate over the coming three months is positively associated with a change in the unemployment rate over the prior three months, and negatively associated with the level of the unemployment gap. 3 See, for example: Marcus Dillender, Carolyn Heinrich, and Susan Houseman, "Effects of the Affordable Care Act on Part-Time Employment: Early Evidence," Upjohn Institute Working Paper, 2016. 4 Janet Yellen, "Semiannual Monetary Policy Report To The Congress," February 14, 2017. 5 Please see ECB, "Focus: Assessing Labour Market Slack," Economic Bulletin Issue 3, 2017. Strategy & Market Trends Tactical Trades Strategic Recommendations Closed Trades

Highlights ECB policy is set to become less dovish relative to other central banks. Stay long the euro; stay underweight German bunds within a global bond portfolio; and overweight euro area Financials within a global Financials portfolio. Female labour participation is surging. The state of the euro area labour market is not nearly as bad as many pessimists would have you believe. Play the mega-trend of rising female labour participation with a structural overweight in the Personal Products sector. Allowing for euro break-up risk, European equities are fairly valued - rather than cheap - versus U.S. equities. Prefer to gain exposure via a 50:50 combination of Germany (DAX) and Sweden (OMX). Feature "Domestic sources of risk to euro area growth have diminished while global, geo-global sources of risk have increased." - Mario Draghi The Cleanest Dirty Shirt Since the end of 2014, an unspectacular 1.9% growth rate1 has been enough to make the euro area the world's top-performing major economy - bettering the U.S., U.K. and Japan (Chart I-2). Chart of the WeekThe Percentage Of The French Population In Employment Is At An All-Time High The Percentage Of The French Population In Employment Is At An All-Time High The Percentage Of The French Population In Employment Is At An All-Time High Chart I-2The Euro Area Is The Top-Performing Economy The Euro Area Is The Top-Performing Economy The Euro Area Is The Top-Performing Economy The euro area economy has achieved this outperformance with exceptionally low volatility. For eight consecutive quarters, growth2 has remained within a very tight 1.2-2.2% band, less than half of the equivalent volatility in the U.S., U.K. and Japan. And growth is now "solid and broad", meaning that it includes all countries. The ECB's dispersion index of value-added growth in different countries stands at a historical minimum. We expect the euro area to remain the cleanest dirty shirt. As Draghi points out, the ECB is less worried about domestic risks and more worried about global risks. Specifically: "Markets are in the course of reassessment of U.S. fiscal policy" - Trumponomics will not be nearly as stimulative as first thought. "How the U.K. economy does post-Brexit has a channel of economic consequences for the euro area." "Possible negative surprises in some emerging market economies" - notably China. If any of the global risks do flare up, the ECB will sit pat, but other central banks will have to become more dovish relative to current expectations. If the risks do not flare up, the ECB will start to reduce its own extreme dovishness - at least with words, if not actions. Either way, ECB policy is set to become less dovish relative to other central banks. And the investment implications are: stay long the euro; stay underweight German bunds within a global bond portfolio; and overweight euro area Financials within a global Financials portfolio. Female Labour Participation Is Surging Chart I-3Rising Participation Boosts Employment Rising Participation Boosts Employment Rising Participation Boosts Employment As Emanuel Macron prepares to become the twenty fifth President of the French Republic, he can take heart from a statistic which may surprise you: The percentage of the French population in employment has never been this high. (Chart of the Week). How can this be when the French unemployment rate is still hovering around 10%? The answer is: as millions of formerly inactive French citizens have entered the labour market, it has lifted the percentage of the population with jobs to an all-time high (Chart I-3). But the flip side of rising participation is that it has kept the unemployment rate elevated - because some citizens who were formerly 'uncounted inactive' are now 'counted unemployed'. Remember that to count as unemployed, a person has to be in the labour market available for work. Some argue that French citizens have simply flooded into the labour market to claim generous and long-lasting unemployment benefits. This argument might hold during downturns, but it cannot explain the 25-year uptrend which also includes economic booms. Unpalatable as it might be to the pessimists, we are left with a more optimistic explanation. France has raised activity levels in the working age population with policies that encourage much greater female participation in the labour market. The important lesson is that when labour participation is rising or falling, we must interpret the headline unemployment rate with extreme care.3 If a country's unemployment rate is high because labour participation has increased - as in France - the labour market is not quite as bad as the high unemployment rate might suggest.4 Conversely, if a country's unemployment rate is low because labour participation has decreased - as in the U.S. (Chart I-4) - the labour market is not quite as good as the low unemployment rate might suggest. Counted unemployment has just been replaced with uncounted inactivity. We propose that the percentage of the working age population in employment is the truer measure of labour utilisation. With surging female participation boosting employment in France and most other European countries (Chart I-5), the state of the euro area labour market is not nearly as bad as many pessimists would have you believe. Chart I-4Participation Down In The U.S.,##br## But Up In Europe... Participation Down In The U.S., But Up In Europe... Participation Down In The U.S., But Up In Europe... Chart I-5...Led By ##br##Women ...Led By Women ...Led By Women Play the mega-trend of rising female labour participation with a structural overweight in the Personal Products sector. Political Risk Is Correctly Priced Many people saw the Brexit and Trump victories as the leading edge of a wave of economic nationalism. However, subsequent election results in the Netherlands, Austria, Finland, Bulgaria and now France have seen economic nationalists consistently underperforming their expectations. In hindsight, the Brexit and Trump victories were idiosyncratic. Both the Remain and Clinton campaigns were lacking in personality or a strong emotional message, and this proved to be their undoing. Nowadays, many voters care about personalities more than policies; emotional appeal matters more than rational appeal. Behavioural psychologist and Nobel Laureate Daniel Kahneman calls the emotional way of thinking "System 1", and the colder rational way of thinking "System 2". Crucially, in a tight contest, both the Brexit and Trump campaigns resonated with the emotional System 1 with passionate pleas such as "Take Back Control" and "Make America Great Again". By contrast, the Remain and Clinton campaigns tried to appeal mainly to the rational System 2. But as Kahneman explains, when rational System 2 competes with emotional System 1, emotional System 1 almost always wins. Chart I-6Euro Break-Up Probability = 5% A Year Euro Break-Up Probability = 5% A Year Euro Break-Up Probability = 5% A Year In more recent elections, candidates and parties opposing the nationalists - including Emanuel Macron - have used a good balance of System 1 and System 2 arguments, thereby helping to prevent shock outcomes. This is also likely to be case in the two round French legislative elections on June 11 and 18 which we do not expect to impact financial markets significantly. Does this mean that political risk is over in Europe? No. Until the euro area turns into a permanent and irreversible political union, there has to be a probability of euro break-up. To value euro area assets, investors must ask: what is this break-up probability? The sovereign bond market says it is 5% a year (Chart I-6). This shows up in a discount on German bund yields, because after a euro break-up a new deutschmark would rise; and a symmetrical premium on Italian BTP yields, because a new lira would fall. For the aggregate euro area bond, the risk largely cancels out because intra-euro currency redenomination would be zero sum. But European equities must trade at a discount for this tail-event. At the peak of the euro debt crisis in 2011, the Eurostoxx600 underperformed the S&P500 by 25% in one year. In an outright break-up, the underperformance would almost certainly be worse, let's conservatively say 30-40%. So assuming the tail-event probability is 5% a year, European equities must compensate with a valuation discount which allows a 1.5-2.0%5 excess annual return over U.S. equities. Today, the valuation discount on European equities relative to U.S. equities implies an excess annual return of 1.8%.6 This makes European equities cheap versus U.S. equities only if the annual probability of euro break-up is less than 5%. Our assessment is that a 5% annual risk is about right. Therefore, European equities are fairly valued - rather than cheap - versus U.S. equities. But to avoid the undesirable sector skews in the Eurostoxx600, a much better way to gain long-term exposure to European equities is via a 50:50 combination of Germany (DAX) and Sweden (OMX) (Chart I-7). Chart I-7Prefer A DAX/OMX Combo To The Eurostoxx50 Or Eurstoxx600 Prefer A DAX/OMX Combo To The Eurostoxx50 Or Eurstoxx600 Prefer A DAX/OMX Combo To The Eurostoxx50 Or Eurstoxx600 Dhaval Joshi, Senior Vice President European Investment Strategy dhaval@bcaresearch.com 1 At an annualized rate. 2 At an annualized rate. 3 Geek's note: the unemployment rate can be expressed as: 100*(participation rate - employment to population rate) / (participation rate). Hence, all else being equal, a rising participation rate will raise the unemployment rate and a falling participation rate will depress the unemployment rate. 4 This lesson applies equally to any studies of labour market slack such as this one: https://www.ecb.europa.eu/pub/pdf/other/ebbox201703_03.en.pdf that do not take into account the dynamics of participation rates. 5 5% multiplied by 30-40% equals 1.5-2.0% 6 Through the next ten years. Please see the European Investment Strategy Weekly Report titled "Markets Suspended In Disbelief" dated April 13, 2017 available at eis.bcaresearch.com Fractal Trading Model The rally in the CAC40 after the French election is technically extended. The recommended technical trade is to short the CAC40 versus the Eurostoxx600. For any investment, excessive trend following and groupthink can reach a natural point of instability, at which point the established trend is highly likely to break down with or without an external catalyst. An early warning sign is the investment's fractal dimension approaching its natural lower bound. Encouragingly, this trigger has consistently identified countertrend moves of various magnitudes across all asset classes. Chart I-8 Short CAC40 / Long EUROSTOXX600 Short CAC40 / Long EUROSTOXX600 The post-June 9, 2016 fractal trading model rules are: When the fractal dimension approaches the lower limit after an investment has been in an established trend it is a potential trigger for a liquidity-triggered trend reversal. Therefore, open a countertrend position. The profit target is a one-third reversal of the preceding 13-week move. Apply a symmetrical stop-loss. Close the position at the profit target or stop-loss. Otherwise close the position after 13 weeks. Use the position size multiple to control risk. The position size will be smaller for more risky positions. * For more details please see the European Investment Strategy Special Report "Fractals, Liquidity & A Trading Model," dated December 11, 2014, available at eis.bcaresearch.com Fractal Trading Model Recommendations Equities Bond & Interest Rates Currency & Other Positions Closed Fractal Trades Trades Closed Trades Asset Performance Currency & Bond Equity Sector Country Equity Indicators Bond Yields Chart II-1Indicators To Watch - Bond Yields Indicators To Watch - Bond Yields Indicators To Watch - Bond Yields Chart II-2Indicators To Watch - Bond Yields Indicators To Watch - Bond Yields Indicators To Watch - Bond Yields Chart II-3Indicators To Watch - Bond Yields Indicators To Watch - Bond Yields Indicators To Watch - Bond Yields Chart II-4Indicators To Watch - Bond Yields Indicators To Watch - Bond Yields Indicators To Watch - Bond Yields Interest Rate Chart II-5Indicators To Watch##br## - Interest Rate Expectations Indicators To Watch - Interest Rate Expectations Indicators To Watch - Interest Rate Expectations Chart II-6Indicators To Watch ##br##- Interest Rate Expectations Indicators To Watch - Interest Rate Expectations Indicators To Watch - Interest Rate Expectations Chart II-7Indicators To Watch##br## - Interest Rate Expectations Indicators To Watch - Interest Rate Expectations Indicators To Watch - Interest Rate Expectations Chart II-8Indicators To Watch##br## - Interest Rate Expectations Indicators To Watch - Interest Rate Expectations Indicators To Watch - Interest Rate Expectations

Highlights The Economic Surprise Index has declined and may continue to roll over until expectations wash out. But that shouldn't derail risk assets or the Fed. The GDP data is a mix of art and science. For investors focused on what the quarterly GDP release reveals about the state of the economy, it is important to remember that the advance release involves more of the former. The FOMC called the weakness in Q1 "transitory". The U.S. economy can grow fast enough over the final three quarters of the year to meet the Fed's 2.0% growth target. The recent readings on inflation and the labor market remain consistent with 2 more rate hikes this year, starting in June. We expect the stock-to-bond ratio to hit new highs by the end of the year even without a big move in equity prices. Feature U.S. equities have now returned to their early March highs despite the ongoing weakness in economic surprises. The latest high profile negative surprises were in the Q1 GDP report, and the March reading on core PCE inflation. Have equity prices disconnected from the underlying economic fundamentals or is something else at play? More importantly, how does the Fed view the recent weakness in economic data? The outlook for inflation, the Fed, and growth supports the relative performance of stocks vs bonds, even assuming modest returns to the former. What To Expect After A Weak Q1 The Q1 GDP report was weak. It was the latest in a string of U.S. economic reports stretching back to mid-March that have disappointed relative to (raised) expectations. In February,1 we highlighted the risk that the "current period of economic surprise could last for another month or two..." before inevitably giving way to elevated expectations and finally disappointment. On average since 2010, elevated levels of economic surprise have lasted roughly two months, with the latest period lasted about 11 weeks (Chart 1). So now what? Chart 1Economic Surprise Index Has Rolled Over Since Early to Mid March Economic Surprise Index Has Rolled Over Since Early to Mid March Economic Surprise Index Has Rolled Over Since Early to Mid March Each day that passes, economic expectations move lower, adjusting the bar down for the next batch of economic reports. The starting point was set relatively high just after last fall's election and early this year, as investors anticipated quick action from the Trump Administration and Congress on tax cuts, tax reform and infrastructure. More recently however, some of the key data have not only failed to match raised expectations, but have begun to roll over. Since 2010, periods of disappointing economic reports have persisted, on average, for 4 months (Chart 1). We are nearly 2 months in, implying that expectations will be washed out soon. With a solid backdrop for corporate earnings, and ebbing geopolitical risk, any equity pullback based on near-term weakness in the economic data should be short-lived. Q1 real GDP growth came in at just 0.7%, well below expectations of a 1.1% increase. At the start of 2017, consensus estimates were in the 2 to 2.5% range, but we were not surprised by the weak report and markets should not have been either. In our two most recent reports,2 we highlighted the well-known seasonality issues with Q1 GDP. Markets seemed to have - correctly in our view - taken the Q1 GDP report in stride and are looking ahead to Q2 and beyond. We expect a snapback in growth in Q2 and over the rest of 2017. The Atlanta Fed's Q2 estimate (+4.2%) supports our view but the NY Fed's latest nowcast for Q2 (+1.8) suggests a more modest rebound. In addition to the potential for higher growth later in the year, there is also the chance that Q1 growth was misstated. Investors can track revisions to Q1 GDP via the Atlanta and NY Fed's Nowcasts, and should bear in mind that the GDP data is a mix of art and science. For investors focused on what the quarterly GDP release reveals about the state of the economy, it is important to remember that the advance release involves more of the former. The Bureau of Economic Analysis' (BEA) GDP data are subject to near constant revision. For example, the Q1 2007 GDP data (released in April 2007) has already been revised 10 times (Table 1). Availability to the BEA of input data that is both timely and comprehensive is at the root of this constant revision. Investors need to take this into account as they try to assess the health of the U.S. economy in "real time". In the past 8 years, Q1 GDP has been revised lower half the time between the advance estimate (1/3 of the hard data) and the second estimate (50% of the data). But as currently reported, Q1 GDP in 5 of the last 8 years is now higher than it was when first reported and in some cases these revisions have been significant in magnitude (Table 1). Which reading should investors trust? A look at the composition of those estimates may help. Table 1GDP Is A Mix Of Art And Science Growth, Inflation And The Fed Growth, Inflation And The Fed When the BEA released Q1 GDP in late April it had collected just over a third of the "hard" data that feeds into GDP (Chart 2). The rest of the data used to calculate Q1 GDP was filled in by the BEA using assumptions, or "judgmental trend," or by using data from a similar data series. By the time the second estimate is released in late May, the BEA will have just 50% of the "hard" data. Thus, a healthy dose of skepticism is warranted when evaluating the U.S. economy on the initial reports of GDP. Chart 2Advance Estimate Of GDP##br## Is More Art Than Science Growth, Inflation And The Fed Growth, Inflation And The Fed For now, U.S. equities have not been affected by the weak Q1 GDP data or the recent collapse in positive economic surprises. Our work shows that the disappointing economic data may persist for another few months. Stocks are within a few points of their all-time high set in March; which suggests that markets are less focused on the noise in the economic data, but remain intently focused on the Trump Administration passing some profit friendly legislation at some point this year. If economic disappointments persist for longer than a few more months and Congress doesn't follow through, we can't rule out a meaningful correction in U.S. equities. Nonetheless, the lack of excesses in the economy, general agreement between the Fed and the market on the path of rates for this year and rising, but still modest, inflation are likely to make any pullback in U.S. stocks a buying opportunity for investors. Bottom Line: Investors should fade the recent disappearance in positive economic surprises by staying overweight stocks vs bonds over the coming 6-12 months. FOMC: Growth Weakness Is Transitory Chart 3GDP, Inflation And Labor Market All Tracking##br## To Fed's Forecast = Gradual Rate Hikes GDP, Inflation And Labor Market All Tracking To Fed's Forecast = Gradual Rate Hikes GDP, Inflation And Labor Market All Tracking To Fed's Forecast = Gradual Rate Hikes The pace of economic growth, and more importantly how that growth impacts the labor market and inflation, remain a crucial factor in how investors assess the number of additional Fed rate hikes that can be expected this year. We continue to expect two more 25 basis point hikes in 2017, whereas the market, as of May 4, was pricing in just 38 bps. At the start of the weakness in the economic data in early March, the market had penciled in 68 bps (almost 3 rate hikes). The soft performance of the economy in Q1 was certainly a focus at last week's FOMC meeting. The FOMC's assessment was that the slowdown in growth in the economy in Q1 was "transitory." The FOMC made no material changes to its assessment of inflation or the labor market in the statement. The minutes of last week's meeting due on May 24 will provide more color. While not officially part of the Fed's dual mandate (of inflation and unemployment), economic growth obviously matters to the Fed. Growth that runs above the Fed's view of potential GDP will push the unemployment rate lower and push inflation higher. Top panel of Chart 3 shows that real GDP growth rose 1.9% from a year ago in Q1, just a tenth of a percent below the Fed's central tendency range for 2017 of 2.0 to 2.2% (Chart 3, panel 2). Despite the poor start to 2017, real GDP growth would have to average only a modest 2.5% per quarter over the rest of the year to hit the Fed's 2.0% target. Is 2.5% growth over the final three quarters achievable absent positive revisions to Q1? We think it is. Since 2010, GDP growth in the final 3 quarters of the year has averaged 2.5%. The headwinds facing the economy today are weaker than they were in the early years of the recovery. The April readings on manufacturing (54.8) and non-manufacturing (57.5) ISM imply GDP growth in the 3 to 3.5% range in Q2. The FOMC is correct to look through the temporary weakness in Q1 and continue on its gradual path of rate hikes this year to match the "modest" pace of economic growth. Investors got a few other key inputs to the FOMC's decision making process last week: The March reading on PCE inflation and the April employment report. Both readings keep the Fed on track for gradual hikes in 2017. A soft reading on core PCE inflation - the Fed's preferred measure - was also a contributor to the weakness in the economic surprise index. For now, we see few signs that suggest core inflation is headed sustainably lower. Chart 4 shows that, since 2000, core PCE inflation has closely correlated with a one year lag of real consumer spending. Even with the recent deceleration in spending, the chart suggests that the recent decline in inflation is temporary. In addition, our sense is that the Fed is more likely to tolerate a rate of inflation that is modestly below its estimate as long as growth remains strong and there is evidence that the weakness in inflation is transitory. Chart 4Core PCE Inflation Likely To Move Higher To Meet Spending Core PCE Inflation Likely To Move Higher To Meet Spending Core PCE Inflation Likely To Move Higher To Meet Spending The April labor market data was released last week as well and confirmed the FOMC's assessment of a solid labor market, but it also had a one negative surprise for markets. The 211,000 increase in jobs in April exceeded expectations (+185,000) and accelerated from the 79,000 gain in March. Over the past three months, the average monthly gain in payrolls was 174,000,well above the 100,000 to 125,000 per month pace the Fed says is needed to tighten the labor market. The drop in the unemployment rate in April to 4.4% puts the unemployment rate at pre-recession lows and more importantly, below the lower end of the Fed's 4.5% to 4.6% central tendency for this year. (Chart 3, panel 3). The negative surprise in the April jobs report came from wages. Average hourly earnings decelerated to 2.5% year-over-year in April from +2.6% in March. The consensus was looking for a 2.7% increase. Despite the lack of traction on wages, the April jobs supports the view that the economy is growing fast enough to tighten the labor market, push up wages and ultimately inflation. June remains a close call for the next Fed rate hike, but an analysis of the economy and the Fed's reaction function suggests that two rate hikes remain the most likely event this year. Our view is that the market will adjust up expectations toward the Fed's view for 2018. Bottom Line: The recent disappointment in the data is not enough to knock the Fed off course. Investors should continue to expect two additional rate hikes in 2017, with the next move coming at the June meeting. A Pro-Cyclical Asset Stance: It's Not Just About Stocks Chart 5Investors' Preference For Bonds##br## Is Understandable... Investors' Preference For Bonds Is Understandable... Investors' Preference For Bonds Is Understandable... One of the most basic ways that BCA evaluates the trend in financial markets is to look at what we call the "stock-to-bond ratio". In this publication the ratio is shown as the S&P 500 total return index divided by that of U.S. 10-year government bond. Chart 5 shows the amazing evolution of the stock-to-bond ratio over the past decade, rebased to 100 at the end of 2007 (the official beginning of the 2008-2009 recession). Panel 2 of the chart shows the component total return indexes, also rebased to 100 at the end of 2007. The chart illustrates two incredible points. First, while it is true that stocks have massively outpaced bonds since the low in March 2009, it took equity investors who bought and held at the onset of recession until late-2013 to outpace bond investors who did the same. Second, until the U.S. election in November, the stock-to-bond ratio was only 10% higher than it was in December 2007, which is a powerful testament to the ability of bonds to preserve capital over the long haul. Given these observations and the still-fresh memory of the global financial crisis, it is easy to see how some investors continue to prefer the relative safety of bonds, especially since equity multiples have risen significantly over the past year. However, Chart 6 highlights how our long stock-to-bond call is motivated by an expectation of higher stock prices and negative returns from bonds. The chart shows the likely trajectory of the 10-year Treasury yield over the coming year, under the base case scenario envisioned by our U.S. Bond Strategy service: core PCE inflation rises to 2%, and the spread between the 10-year breakeven inflation rate and core rises to 50 bps. Chart 7 illustrates the implications of this forecast for bond total returns, alongside the resulting stock-to-bond ratio. For stocks, we assume a very conservative 3% annualized nominal total return, which is the sum of a 2% dividend yield and a 1% assumed nominal price return. Chart 6...But The Bond Bull Market Is Over ...But The Bond Bull Market Is Over ...But The Bond Bull Market Is Over Chart 7A New High By Year-End A New High By Year-End A New High By Year-End The key point from Chart 7 is that the stock-to-bond ratio is likely to rise to a new high by the end of the year, even without aggressive assumptions for equity returns. We agree that bond yields will fall in the event of another risk-off event, and that 10-year Treasurys remain an important component of a diversified portfolio. But it is also important for investors to recognize that, absent these types of events, the relative performance of stocks vs. bonds is set to move higher in part because 10-year Treasurys are likely to generate a negative absolute return over the coming 6-12 months. Bottom Line: Investors should retain a pro-cyclical asset allocation stance. The outlook for the inflation, the Fed, and growth supports the relative performance of stocks vs bonds, even assuming modest returns to the former. John Canally, CFA, Senior Vice President U.S. Investment Strategy johnc@bcaresearch.com Jonathan LaBerge, CFA, Vice President Special Reports jonathanl@bcaresearch.com 1 Please see U.S. Investment Strategy Special Report "Goldilocks: For How Long?," dated February 20, 2017, available at usis.bcaresearch.com. 2 Please see U.S. Investment Strategy Special Reports "Spring Snapback" dated April 24, 2017 and "The Good And The Bad". May 1, 2017, available at usis.bcaresearch.com.

Dear Client, In addition to an abbreviated Weekly Report that you will receive later tonight, I am sending you this Special Report written by my colleague Mark McClellan of our monthly Bank Credit Analyst publication. Following up on many of the themes discussed in our latest Quarterly Strategy Outlook, Mark makes a convincing case that most of the factors that have suppressed global interest rates since the financial crisis could begin to unwind or even reverse over the coming years. Best regards, Peter Berezin, Chief Global Strategist Highlights The fundamental drivers of the low rate world are considered by many to be structural, and thus likely to keep global equilibrium bond yields quite depressed by historical standards for years to come. However, some of the factors behind ultra-low interest rates have waned, while others have reached an inflection point. The age structure of world population is transitioning from a period in which aging added to the global pool of savings to one in which aging will begin to drain that pool. Global investment needs will wane along with population aging, but the majority of the effect on equilibrium interest rates is in the past. In contrast, the demographic effects that will depress desired savings are still to come. The net impact will be bond-bearish. Moreover, the massive positive labor supply shock, following the integration of China and Eastern Europe into the world's effective labor force, is over. Indeed, this shock is heading into reverse as the global working-age population ratio falls. This may improve labor's bargaining power, sparking a shift toward using more capital in the production process and thereby placing upward pressure on global real bond yields. It is too early to declare globalization dead, but the neo-liberal trading world order that has been in place for decades is under attack. This could be inflationary if it disrupts global supply chains. Anti-globalization policies could paradoxically be positive for capital spending, at least for a few years. As for China, the fundamental drivers of its savings capacity appear to rule out a return to the days when the country was generating a substantial amount of excess savings. Technological advance will remain a headwind for real wage gains, but at least the transition to a world that is less labor-abundant will boost workers' ability to negotiate a larger share of the income pie. We are not making the case that real global bond yields are going to quickly revert to pre-Lehman averages. Global yields could even drop back to previous lows in the event of another recession. Nonetheless, from a long-term perspective, current market expectations for bond yields are too low. Investors should have a bond-bearish bias on a medium- and long-term horizon. Feature In the September 2016 The Bank Credit Analyst, we summarized the key drivers behind the major global macroeconomic disequilibria that have resulted in deflationary pressure, policy extremism, dismal productivity, and the lowest bond yields in recorded history (Chart 1). The disequilibria include income inequality, the depressed wage share of GDP, lackluster capital spending, and excessive savings. Chart 1Global Disequilibria Beware Inflection Points In The Secular Drivers Of Global Bonds Beware Inflection Points In The Secular Drivers Of Global Bonds The fundamental drivers of the low bond yield world are now well documented and understood by investors. These drivers generally are considered to be structural, and thus likely to keep global equilibrium bond yields and interest rates at historically low levels for years to come according to the consensus. Based on discussions with BCA clients, it appears that many have either "bought into" the secular stagnation thesis or, at a minimum, have adopted the view that growth headwinds preclude any meaningful rise in bond yields. However, bond investors might have been lulled into a false sense of security. Yields will not return to pre-Lehman norms anytime soon, but some of the factors behind the low-yield world have waned, while others have reached an inflection point. Most importantly, the age structure of world population is transitioning from a period in which aging added to the global pool of savings to one in which aging will begin to drain that pool. We have reached the tipping point. Equilibrium real bond yields will gradually move higher as a result. But before we discuss what is changing, it is important to review the drivers of today's macro disequilibria. Several of them predate the Great Financial Crisis, including demographic trends, technological advances, and the integration of China's massive workforce and excess savings into the global economy. Ultra-Low Rates: How Did We Get Here? (A) Demographics And Global Savings Chart 2Global Shifts In The Saving And Investment Curves Beware Inflection Points In The Secular Drivers Of Global Bonds Beware Inflection Points In The Secular Drivers Of Global Bonds The so-called Global Savings Glut has been a bullish structural force for bonds for the past couple of decades. We won't go through all of the forces behind the glut, but a key factor is population aging in the advanced economies. Ex-ante desired savings rose as baby boomers entered their high-income years. The Great Financial Crisis only served to reinforce the desire to save, given the setback in the value of boomers' retirement nest eggs.1 The corporate sector also began to save more following the crisis. Even more importantly, the surge in China's trade surplus since the 1990s had to be recycled into the global pool of savings. While China's rate of investment was very high, its propensity to save increased even faster, resulting in a swollen external surplus and a massive net outflow of capital. Other emerging economies also made the adjustment from net importers of capital to net exporters following the Asian crisis in the late 1990s. By leaning into currency appreciation, these countries built up huge foreign exchange reserves that had to be recycled abroad. In theory, savings must equal investment at the global level and real interest rates shift to ensure this equilibrium (Chart 2). China's excess savings, together with a greater desire to save in the developed countries, represented a shift in the saving schedule to the right. The result was downward pressure on global interest rates. (B) Demographics And Global Capital Spending Demographics and China's integration also affected the investment side of the equation. A slower pace of labor force growth in the developed countries resulted in a permanently lower level of capital spending relative to GDP. Slower consumer spending growth, as a result of a more moderate expansion in the working-age population, meant a reduced appetite for new factories, malls, and apartment buildings. Chart 3 shows that the growth rate of global capital spending that is required to maintain a given capital-to-output ratio has dropped substantially, due to the dramatic slowdown in the growth of the world's working-age population.2 Keep in mind that this estimate refers only to the demographic component of investment spending. Actual capital expenditure growth will not be as weak as Chart 3 suggests because firms will want to adopt new technologies for competitive or environmental reasons. Nonetheless, the point is that the structural tailwind for global capex from the post-war baby boom has disappeared. Chart 3Demographics Are A Structural Headwind For Global Capex Beware Inflection Points In The Secular Drivers Of Global Bonds Beware Inflection Points In The Secular Drivers Of Global Bonds (C) Labor Supply Shock And Global Capital Spending While the working-age population ratio peaked in the developed countries years ago, it is a different story at the global level (Chart 4). The integration of the Chinese and Eastern European workforces into the global labor pool during the 1990s and 2000s resulted in an effective doubling of global labor supply in a short period of time. Relative prices must adjust in the face of such a large boost in the supply of labor relative to capital. The sudden abundance of cheap labor depressed real wages from what they otherwise would have been, thus incentivizing firms to use more labor and less capital at the margin. The combination of slower working-age population growth in the advanced economies and a surge in the global labor force resulted in a decline in desired global capital spending. In terms of Chart 2, the leftward shift of the investment schedule reinforced the impact of the savings impulse in placing downward pressure on global interest rates. (D) Labor Supply Shock And Income Inequality The wave of cheap labor also aggravated the trend toward greater inequality in the advanced economies and the downward trend in labor's share of the income pie (Chart 5). In theory, a surge in the supply of labor is a positive "supply shock" that benefits both developed and developing countries. However, a recent report by David Autor and Gordon Hanson3 highlighted that trade agreements in the past were incremental and largely involved countries with similar income levels. The sudden entry of China to the global trade arena, involving a massive addition to the effective global stock of labor, was altogether different. The report does not argue that trade has become a "bad" thing. Rather, it points out that the adjustment costs imposed on the advanced economies were huge and long-lasting, as Chinese firms destroyed entire industries in developed countries. The lingering adjustment phase contributed to greater inequality in the major countries. Management was able to use the threat of outsourcing to gain the upper hand in wage negotiations. The result has been a rise in the share of income going to high-income earners in the Advanced Economies, at the expense of low- and middle-income earners (Chart 6). The same is true, although to a lesser extent, in the emerging world. Chart 4Working-Age Population Ratios Have Peaked Working-Age Population Ratios Have Peaked Working-Age Population Ratios Have Peaked Chart 5Labor Share Of Income Has Dropped Labor Share Of Income Has Dropped Labor Share Of Income Has Dropped Chart 6Hollowing Out Hollowing Out Hollowing Out Greater inequality, in turn, has weighed on aggregate demand and equilibrium interest rates because a larger share of total income flowed to the "rich" who tend to save more than the low- and middle-income classes. (E) The Dark Side Of Technology Advances in technology also contributed to rising inequality. In theory, new technologies hurt some workers in the short term, but benefit most workers in the long run because they raise national income. However, there is evidence that past major technological shocks were associated with a "hollowing out" or U-shaped pattern of employment. Low- and high-skilled employment increased, but the proportion of mid-skilled workers tended to shrink. Wages for both low- and mid-skilled labor did not keep up with those that were highly-skilled, leading to wider income disparity. Today, technology appears to be resulting in faster, wider and deeper degrees of hollowing-out than in previous periods of massive technological change. This may be because machines are not just replacing manual human tasks, but cognitive ones too. A recent IMF report made the case that technology and global integration played a dominant role in labor's declining fortunes. Technology alone explains about half of the drop in the labor share of income in the developed countries since 1980.4 Falling prices for capital goods, information and communications technology in particular, have facilitated the expansion of global value chains as firms unbundled production into many tasks that were distributed around the world in a way that minimized production costs. Chart 7 highlights that the falling price of capital goods in the advanced economies went hand-in-hand with rising participation in global supply chains since 1990. Falling capital goods prices also accelerated the automation of routine tasks, contributing especially to job destruction in the developed (high-wage) economies. In other words, firms in the developed world either replaced workers with machinery in areas where technology permitted, or outsourced jobs to lower-wage countries in areas that remained labor-intensive. Both trends undermined labor's bargaining power, depressed labor's share of income, and contributed to inequality. The effects of technology, global integration, population aging and China's economic integration are demonstrated in Chart 8. The world working-age-to-total population ratio rose sharply beginning in the late 1990s. This resulted in an upward trend in China's investment/GDP ratio, and a downward trend in the G7. The upward trend in the G7 capital stock-per-capita ratio began to slow as a result, before experiencing an unprecedented contraction after the Great Recession and Financial Crisis. Chart 7Economic Integration And Falling Capital Goods Prices Economic Integration And Falling Capital Goods Prices Economic Integration And Falling Capital Goods Prices Chart 8Macro Impact Of Labor Supply Shock Macro Impact Of Labor Supply Shock Macro Impact Of Labor Supply Shock The result has been a deflationary global backdrop characterized by demand deficiency and poor potential real GDP growth, both of which have depressed equilibrium global interest rates over the past 20 to 25 years. Transition Phase Chart 9Working-Age Population To Shrink in G7 and China Working-Age Population To Shrink in G7 and China Working-Age Population To Shrink in G7 and China It would appear easy to conclude that these trends will be with us for another few decades because the demographic trends will not change anytime soon. Nonetheless, on closer inspection the global economy is transitioning from a period when cyclical economic pressures and all of the structural trends were pushing equilibrium interest rates in the same direction, to a period in which the economic cycle is becoming less bond-friendly and some of the secular drivers of low interest rates are gradually changing direction. First, the massive labor supply shock of the past few decades is over. The world working-age population ratio has peaked according to United Nations estimates. This ratio is already declining in the major advanced economies and is in the process of topping out in China. The absolute number of working-age people will shrink in China and the G7 countries over the next five years, although it will continue to grow at a low rate for the world as a whole (Chart 9). Unions are unlikely to make a major comeback, but a backdrop that is less labor-abundant should gradually restore some worker bargaining power, especially as economies regain full employment. The resulting upward pressure on real wages will support capital spending as firms substitute toward capital and away from (increasingly expensive) labor. Consumer demand will also receive a boost if inequality moderates and the labor share of income begins to rise. Globalization On The Back Foot Chart 10Globalization Peaking? Globalization Peaking? Globalization Peaking? Second, it is too early to declare globalization dead, but the neo-liberal trading world order that has been in place for decades is under attack. Global exports appear to have peaked relative to GDP and average tariffs have ticked higher (Chart 10). The World Trade Organization has announced that the number of new trade restrictions or impediments outweighed the number of trade liberalizing initiatives in 2016. The U.K. appears willing to sacrifice trade for limits to the free movement of people. The new U.S. Administration has ditched the Trans-Pacific Partnership (TPP) and is threatening to impose punitive tariffs on some trading partners. Anti-globalization policies could paradoxically be positive for capital spending, at least for a few years. If the U.S. were to impose high tariffs on China, for example, it would make a part of the Chinese capital stock redundant overnight. In order for the global economy to produce the same amount of goods and services as before, the U.S. and other countries would need to invest more. Any unwinding of globalization would also be inflationary as it would disrupt international supply chains. Demographics And Saving: From Tailwind To Headwind... Third, the impact of savings in the major advanced economies and China on global interest rates will change direction as well. In the developed world, aggregate household savings will come under downward pressure as boomers increasingly shift into retirement. Economists are fond of employing the so-called life-cycle theory of consumer spending. According to this theory, consumers tend to smooth out lifetime spending by accumulating assets during the working years in order to maintain a certain living standard after retirement. The U.N. National Transfer Accounts Project has gathered data on spending and labor income by age cohort at a point in time. Chart 11 presents the data for China and three of the major advanced economies. The data for the advanced economies suggest that spending tends to rise sharply from a low level between birth and about 15 years of age. It continues to rise, albeit at a more modest pace, through the working years. Other studies have found that consumer spending falls during retirement. Nonetheless, these studies generally include only private spending and therefore do not include health care that is provided by the government. The data presented in Chart 11 show that, if government-provided health care is included, personal spending rises sharply toward the end of life. The profile is somewhat different in China. Spending rises quickly from birth to about 20 years of age, and is roughly flat thereafter. Indeed, consumption edges lower after 75-80 years of age. These data allow us to project the impact of changing demographics on the average household saving rate in the coming years, assuming that the income and spending profiles shown in Chart 11 are unchanged. We start by calculating the average saving rate across age cohorts given today's age structure. We then recalculate the average saving rate each year moving forward in time. The resulting saving rate changes along with the age structure of the population. Chart 11Income And Consumption By Age Cohort Beware Inflection Points In The Secular Drivers Of Global Bonds Beware Inflection Points In The Secular Drivers Of Global Bonds The results are shown in Chart 12. The saving rates for all four economies have been indexed at zero in 2016 for comparison purposes. The aggregate saving rate declines in all cases, falling between 4 and 8 percentage points between 2016 and 2030. Germany sees the largest drop of the four countries. Chart 12Aging Will Undermine Aggregate Saving Aging Will Undermine Aggregate Saving Aging Will Undermine Aggregate Saving The simulations are meant to be suggestive, rather than a precise forecast, because the savings profile across age cohorts will adjust over time. Moreover, governments will no doubt raise taxes to cover the rising cost of health care, providing a partial offset in terms of the national saving rate.5 Nonetheless, the simulations highlight that the major economies are past the point where the baby boom generation is adding to the global savings pool at a faster pace than retirees are drawing from it. The age structure in the major advanced economies is far enough advanced that the rapid increase in the retirement rate will place substantial downward pressure on aggregate household savings in the coming years. It is well known that population aging will also undermine government budgets. Rising health care costs are already captured in our household saving rate projection because the data for household spending includes health care even if it is provided by the public sector. However, public pension schemes will also be a problem. To the extent that politicians are slow to trim pension benefits and/or raise taxes, public pension plans will be a growing drain on national savings. Could younger, less developed economies offset some of the demographic trends in China and the Advanced Economies? Numerically speaking, a more effective use of underutilized populations in Africa and India could go a long way. Nevertheless, deep-seated structural problems would have to be addressed and, even then, it is difficult to see either of these regions turning into the next "China story" given the current backlash against globalization and immigration. ...And The Capex Story Is Largely Behind Us Demographic trends also imply less capital spending relative to GDP, as discussed above. In terms of the impact on global equilibrium interest rates, it then becomes a race between falling saving and investment rates. Some analysts point to the Japanese experience because it is the leading edge in terms of global aging. Bond yields have been extremely low for many years even as the household saving rate collapsed, suggesting that ex-ante investment spending shifted by more than ex-ante savings. Nonetheless, Japan may not be a good example because the deterioration in the country's demographics coincided with burst bubbles in both real estate and stocks that hamstrung Japanese banks for decades. A series of policy mistakes made things worse. Economic theory is not clear on the net effect of demographics on savings and investment. The academic empirical evidence is inconclusive as well. However, a detailed IMF study of 30 OECD countries analyzed the demographic impact on a number of macroeconomic variables, including savings and investment.6 They estimated separate demographic effects for the old-age dependency ratio and the working-age population ratio. Applying the IMF's estimated model coefficients to projected changes in both of these ratios over the next decade suggests that the decline in ex-ante savings will exceed the ex-ante drop in capex requirements by about 1 percentage point of GDP. This is a non-trivial shift. Chart 13Demographics And Capex Requirements Beware Inflection Points In The Secular Drivers Of Global Bonds Beware Inflection Points In The Secular Drivers Of Global Bonds Moreover, our simulations highlight that timing is important. The outlook for the household saving rate depends on the changing age structure of the population and the distribution of saving rates across age cohorts. Thus, the average saving rate will trend down as populations continue to age over the coming decades. In contrast, the impact of demographics on capital spending requirements is related to the change in the growth rate of the working-age population. Chart 13 once again presents our estimates for the demographic component of capital spending. The top panel presents the world capex/GDP ratio that is necessary to maintain a constant capital/output ratio, and the bottom panel shows the change in that ratio. The important point is that the downward adjustment in world capex/GDP related to aging is now largely behind us because most of the deceleration in the growth rate of the working-age population is done. This is in contrast to the household saving rate adjustment where all of the adjustment is still to come. China Is Transitioning Too China must be treated separately from the developed countries because of its unique structural issues. As discussed above, household savings increased dramatically beginning in the mid-1990s (Chart 14). This trend reflected a number of factors, including: the rising share of the working-age population; a drop in the fertility rate, following the introduction of the one-child policy in the late 1970s that allowed households to spend less on raising children and save more for retirement; health care reform in the early 1990s required households to bear a larger share of health care spending; and job security was also undermined by reform of the state-owned enterprises (SOE) in the late 1990s, leading to increased precautionary savings to cover possible bouts of unemployment. These savings tailwinds have turned around in recent years and the household saving rate appears to have peaked. China's contribution to the global pool of savings has already moderated significantly, as measured by the current account surplus. The surplus has withered from about 9% in 2008 to 2½% in 2016. A recent IMF study makes the case that China's national saving rate will continue to decline. The IMF estimates that for every one percentage-point rise in the old-age dependency ratio, the aggregate household saving rate will fall by 0.4-1 percentage points. In addition, the need for precautionary savings is expected to ease along with improvements in the social safety net, achieved through higher government spending on health care. The household saving rate will fall by three percentage points by 2021 according to the IMF (Chart 15). Competitive pressure and an aging population will also reduce the saving rates of the corporate and government sectors. Chart 14China's Savings Rates Have Peaked... China's Savings Rates Have Peaked... China's Savings Rates Have Peaked... Chart 15...Suggesting That External Surplus Will Shrink ...Suggesting That External Surplus Will Shrink ...Suggesting That External Surplus Will Shrink Of course, investment as a share of GDP is projected to moderate too, reflecting a rebalancing of the economy away from exports and capital spending toward household consumption. The IMF expects that savings will moderate slightly faster than investment, leading to a narrowing in the current account surplus to almost zero by 2021. A lot of assumptions go into this type of forecast such that we must take it with a large grain of salt. Nonetheless, the fundamental drivers of China's savings capacity appear to rule out a return to the days when the country was generating a substantial amount of excess savings. Moreover, a return to large current account surpluses would likely require significant currency depreciation, which is a political non-starter given U.S. angst over trade. The risk is that China's excess savings will be less, not more, in five year's time. Tech Is A Wildcard It is extremely difficult to forecast the impact of technological advancement on the global economy. We cannot say with any conviction that the tech-related effects of "hollowing out", "winner-take-all" and the "skills premium" will moderate in the coming years. Nonetheless, these effects have occurred alongside a surge in the world's labor force and rapid globalization of supply chains, both of which reinforced the erosion of employee bargaining power. Looking ahead, technology will still be a headwind for some employees, but at least the transition from a world of excess labor to one that is more labor-scarce will boost workers' ability to negotiate a larger share of the income pie. We will explore the impact of technology on productivity, inflation, growth, and bond yields in a companion report to be published in the next issue. Conclusion: Table 1Key Secular Drivers Beware Inflection Points In The Secular Drivers Of Global Bonds Beware Inflection Points In The Secular Drivers Of Global Bonds The main points we made in this report are summarized in Table 1. All of the structural factors driving real bond yields were working in the same (bullish) direction over the past 30-40 years. Looking ahead, it is uncertain how technological improvement will affect bond prices, but we expect that the others will shift (or have already shifted) to either neutral or outright bond-bearish. No doubt, our views that globalization and inequality have peaked, and that the labor share of income has bottomed, are speculative. These factors may not place much upward pressure on equilibrium yields. Nonetheless, it seems likely that the demographic effect that has depressed capital spending demand is well advanced. We see it shifting from a positive factor for bond prices to a neutral factor in the coming years. It is also clear that the massive positive labor supply shock is over, and is heading into reverse as the global working-age population ratio falls. This may improve labor's bargaining power and the resulting boost consumer spending will be negative for bonds. This may also spark a shift toward using more capital in the production process and thereby place additional upward pressure on global real bond yields. Admittedly, however, this last point requires more research because theory and empirical evidence on it are not clear. Perhaps most importantly, the aging of the population in the advanced economies has reached a tipping point; retirees will drain more from the pool of savings than the working-age population will add to it in the coming years. We have concentrated on real equilibrium bond yields in this report because it is the part of nominal yields that is the most depressed relative to historical norms. The inflation component is only a little below a level that is consistent with central banks meeting their 2% inflation targets in the medium term. There is a risk that inflation will overshoot these targets, leading to a possible surge in long-term inflation expectations that turbocharges the bond bear market. This is certainly possible, as highlighted by a recent Global Investment Strategy Quarterly Strategy Outlook. Pain in bond markets would be magnified in this case, especially if central banks are forced to aggressively defend their targets. Please note that we are not making the case that real global bond yields will quickly revert to pre-Lehman averages. It will take time for the bond-bullish structural factors to unwind. It will also take time for inflation to gain any momentum, even in the United States. Global yields could even drop back to previous lows in the event of another recession. Nonetheless, from a long-term perspective, current market expectations suggest that investors have adopted an overly benign view on the outlook for yields. For example, implied real short-term rates remain negative until 2021 in the U.S. and 2026 in the Eurozone, while they stay negative out to 2030 in the U.K. (Chart 16). We doubt that short-term rates will be negative for that long, given the structural factors discussed above. Another way of looking at this is presented in Chart 17. The market expects the 10-year Treasury yield in ten years to be only slightly above today's spot yield, which itself is not far above the lowest levels ever recorded. Market expectations are equally depressed for the 5-year forward rate for the U.S. and the other major economies. Chart 16Market Expects Negative Short-Term Rates For A Long Time Market Expects Negative Short-Term Rates For A Long Time Market Expects Negative Short-Term Rates For A Long Time Chart 17Forward Rates Very Low Vs. History Forward Rates Very Low Vs. History Forward Rates Very Low Vs. History The implication is that investors should have a bond-bearish bias on a medium- and long-term horizon. Mark McClellan, Senior Vice President The Bank Credit Analyst markm@bcaresearch.com 1 It is true that observed household saving rates fell in some of the advanced economies, such as the United States, at a time when aging should have boosted savings from the mid-1990s to the mid-2000s. This argues against a strong demographic effect on savings. However, keep in mind that we are discussing desired (or ex-ante) savings. Ex-post, savings can go in the opposite direction because of other influencing factors. As discussed below, global savings must equal investment, which means that shifts in desired capital spending demand matter for the ex-post level of savings. 2 Arithmetically, if world trend GDP growth slows by one percentage point, then investment spending would need to drop by about 3½ percentage points of GDP to keep the capital/output ratio stable. 3 David H. Autor, David Dorn, and Gordon H. Hanson, "The China Shock: Learning from Labor-Market Adjustment to Large Changes in Trade," Annual Review of Economics, Vol 8, pp. 205-240 (October 2016). 4 Please see "Understanding The Downward Trend In Labor Income Shares," Chapter 3, IMF World Economic Outlook (April 2017). 5 In other words, while the household saving rate, as defined here to include health care spending by governments on behalf of households, will decline, any associated tax increases will blunt the impact on national savings (i.e., savings across the household, government, and business sectors). 6 Jong-Won Yoon, Jinill Kim, and Jungjin Lee, "Impact Of Demographic Changes On Inflation And The Macroeconomy," IMF Working Paper no. 14/210 (November 2014).

The fundamental drivers of the low rate world are considered by many to be structural, and thus likely to keep global equilibrium bond yields quite depressed by historical standards for years to come. However, some of the factors behind ultra-low interest rates have waned, while others have reached an inflection point. The age structure of world population is transitioning from a period in which aging added to the global pool of savings to one in which aging will begin to drain that pool. Global investment needs will wane along with population aging, but the majority of the effect on equilibrium interest rates is in the past. In contrast, the demographic effects that will depress desired savings are still to come. The net impact will be bond-bearish. Moreover, the massive positive labor supply shock, following the integration of China and Eastern Europe into the world's effective labor force, is over. Indeed, this shock is heading into reverse as the global working-age population ratio falls. This may improve labor's bargaining power, sparking a shift toward using more capital in the production process and thereby placing upward pressure on global real bond yields. It is too early to declare globalization dead, but the neo-liberal trading world order that has been in place for decades is under attack. This could be inflationary if it disrupts global supply chains. Anti-globalization policies could paradoxically be positive for capital spending, at least for a few years. As for China, the fundamental drivers of its savings capacity appear to rule out a return to the days when the country was generating a substantial amount of excess savings. Technological advance will remain a headwind for real wage gains, but at least the transition to a world that is less labor-abundant will boost workers' ability to negotiate a larger share of the income pie. We are not making the case that real global bond yields are going to quickly revert to pre-Lehman averages. Global yields could even drop back to previous lows in the event of another recession. Nonetheless, from a long-term perspective, current market expectations for bond yields are too low. Investors should have a bond-bearish bias on a medium- and long-term horizon. In the September 2016 The Bank Credit Analyst, we summarized the key drivers behind the major global macroeconomic disequilibria that have resulted in deflationary pressure, policy extremism, dismal productivity, and the lowest bond yields in recorded history (Chart II-1). The disequilibria include income inequality, the depressed wage share of GDP, lackluster capital spending, and excessive savings. Chart II-1Global Disequilibria May 2017 May 2017 The fundamental drivers of the low bond yield world are now well documented and understood by investors. These drivers generally are considered to be structural, and thus likely to keep global equilibrium bond yields and interest rates at historically low levels for years to come according to the consensus. Based on discussions with BCA clients, it appears that many have either "bought into" the secular stagnation thesis or, at a minimum, have adopted the view that growth headwinds preclude any meaningful rise in bond yields. However, bond investors might have been lulled into a false sense of security. Yields will not return to pre-Lehman norms anytime soon, but some of the factors behind the low-yield world have waned, while others have reached an inflection point. Most importantly, the age structure of world population is transitioning from a period in which aging added to the global pool of savings to one in which aging will begin to drain that pool. We have reached the tipping point. Equilibrium real bond yields will gradually move higher as a result. But before we discuss what is changing, it is important to review the drivers of today's macro disequilibria. Several of them predate the Great Financial Crisis, including demographic trends, technological advances, and the integration of China's massive workforce and excess savings into the global economy. Ultra-Low Rates: How Did We Get Here? (A) Demographics And Global Savings Chart II-2Global Shifts In The ##br##Saving And Investment Curves May 2017 May 2017 The so-called Global Savings Glut has been a bullish structural force for bonds for the past couple of decades. We won't go through all of the forces behind the glut, but a key factor is population aging in the advanced economies. Ex-ante desired savings rose as baby boomers entered their high-income years. The Great Financial Crisis only served to reinforce the desire to save, given the setback in the value of boomers' retirement nest eggs.1 The corporate sector also began to save more following the crisis. Even more importantly, the surge in China's trade surplus since the 1990s had to be recycled into the global pool of savings. While China's rate of investment was very high, its propensity to save increased even faster, resulting in a swollen external surplus and a massive net outflow of capital. Other emerging economies also made the adjustment from net importers of capital to net exporters following the Asian crisis in the late 1990s. By leaning into currency appreciation, these countries built up huge foreign exchange reserves that had to be recycled abroad. In theory, savings must equal investment at the global level and real interest rates shift to ensure this equilibrium (Chart II-2). China's excess savings, together with a greater desire to save in the developed countries, represented a shift in the saving schedule to the right. The result was downward pressure on global interest rates. (B) Demographics And Global Capital Spending Demographics and China's integration also affected the investment side of the equation. A slower pace of labor force growth in the developed countries resulted in a permanently lower level of capital spending relative to GDP. Slower consumer spending growth, as a result of a more moderate expansion in the working-age population, meant a reduced appetite for new factories, malls, and apartment buildings. Chart II-3 shows that the growth rate of global capital spending that is required to maintain a given capital-to-output ratio has dropped substantially, due to the dramatic slowdown in the growth of the world's working-age population.2 Keep in mind that this estimate refers only to the demographic component of investment spending. Actual capital expenditure growth will not be as weak as Chart II-3 suggests because firms will want to adopt new technologies for competitive or environmental reasons. Nonetheless, the point is that the structural tailwind for global capex from the post-war baby boom has disappeared. Chart II-3Demographics Are A Structural Headwind For Global Capex May 2017 May 2017 (C) Labor Supply Shock And Global Capital Spending While the working-age population ratio peaked in the developed countries years ago, it is a different story at the global level (Chart II-4). The integration of the Chinese and Eastern European workforces into the global labor pool during the 1990s and 2000s resulted in an effective doubling of global labor supply in a short period of time. Relative prices must adjust in the face of such a large boost in the supply of labor relative to capital. The sudden abundance of cheap labor depressed real wages from what they otherwise would have been, thus incentivizing firms to use more labor and less capital at the margin. The combination of slower working-age population growth in the advanced economies and a surge in the global labor force resulted in a decline in desired global capital spending. In terms of Chart II-2, the leftward shift of the investment schedule reinforced the impact of the savings impulse in placing downward pressure on global interest rates. (D) Labor Supply Shock And Income Inequality The wave of cheap labor also aggravated the trend toward greater inequality in the advanced economies and the downward trend in labor's share of the income pie (Chart II-5). In theory, a surge in the supply of labor is a positive "supply shock" that benefits both developed and developing countries. However, a recent report by David Autor and Gordon Hanson3 highlighted that trade agreements in the past were incremental and largely involved countries with similar income levels. The sudden entry of China to the global trade arena, involving a massive addition to the effective global stock of labor, was altogether different. The report does not argue that trade has become a "bad" thing. Rather, it points out that the adjustment costs imposed on the advanced economies were huge and long-lasting, as Chinese firms destroyed entire industries in developed countries. The lingering adjustment phase contributed to greater inequality in the major countries. Management was able to use the threat of outsourcing to gain the upper hand in wage negotiations. The result has been a rise in the share of income going to high-income earners in the Advanced Economies, at the expense of low- and middle-income earners (Chart II-6). The same is true, although to a lesser extent, in the emerging world. Chart II-4Working-Age Population Ratios Have Peaked Working-Age Population Ratios Have Peaked Working-Age Population Ratios Have Peaked Chart II-5Labor Share Of Income Has Dropped Labor Share Of Income Has Dropped Labor Share Of Income Has Dropped Chart II-6Hollowing Out Hollowing Out Hollowing Out Greater inequality, in turn, has weighed on aggregate demand and equilibrium interest rates because a larger share of total income flowed to the "rich" who tend to save more than the low- and middle-income classes. (E) The Dark Side Of Technology Advances in technology also contributed to rising inequality. In theory, new technologies hurt some workers in the short term, but benefit most workers in the long run because they raise national income. However, there is evidence that past major technological shocks were associated with a "hollowing out" or U-shaped pattern of employment. Low- and high-skilled employment increased, but the proportion of mid-skilled workers tended to shrink. Wages for both low- and mid-skilled labor did not keep up with those that were highly-skilled, leading to wider income disparity. Today, technology appears to be resulting in faster, wider and deeper degrees of hollowing-out than in previous periods of massive technological change. This may be because machines are not just replacing manual human tasks, but cognitive ones too. A recent IMF report made the case that technology and global integration played a dominant role in labor's declining fortunes. Technology alone explains about half of the drop in the labor share of income in the developed countries since 1980.4 Falling prices for capital goods, information and communications technology in particular, have facilitated the expansion of global value chains as firms unbundled production into many tasks that were distributed around the world in a way that minimized production costs. Chart II-7 highlights that the falling price of capital goods in the advanced economies went hand-in-hand with rising participation in global supply chains since 1990. Falling capital goods prices also accelerated the automation of routine tasks, contributing especially to job destruction in the developed (high-wage) economies. In other words, firms in the developed world either replaced workers with machinery in areas where technology permitted, or outsourced jobs to lower-wage countries in areas that remained labor-intensive. Both trends undermined labor's bargaining power, depressed labor's share of income, and contributed to inequality. The effects of technology, global integration, population aging and China's economic integration are demonstrated in Chart II-8. The world working-age-to-total population ratio rose sharply beginning in the late 1990s. This resulted in an upward trend in China's investment/GDP ratio, and a downward trend in the G7. The upward trend in the G7 capital stock-per-capita ratio began to slow as a result, before experiencing an unprecedented contraction after the Great Recession and Financial Crisis. Chart II-7Economic Integration And ##br##Falling Capital Goods Prices Economic Integration And Falling Capital Goods Prices Economic Integration And Falling Capital Goods Prices Chart II-8Macro Impact Of ##br##Labor Supply Shock Macro Impact Of Labor Supply Shock Macro Impact Of Labor Supply Shock The result has been a deflationary global backdrop characterized by demand deficiency and poor potential real GDP growth, both of which have depressed equilibrium global interest rates over the past 20 to 25 years. Transition Phase Chart II-9Working-Age Population ##br##To Shrink In G7 And China Working-Age Population To Shrink in G7 and China Working-Age Population To Shrink in G7 and China It would appear easy to conclude that these trends will be with us for another few decades because the demographic trends will not change anytime soon. Nonetheless, on closer inspection the global economy is transitioning from a period when cyclical economic pressures and all of the structural trends were pushing equilibrium interest rates in the same direction, to a period in which the economic cycle is becoming less bond-friendly and some of the secular drivers of low interest rates are gradually changing direction. First, the massive labor supply shock of the past few decades is over. The world working-age population ratio has peaked according to United Nations estimates. This ratio is already declining in the major advanced economies and is in the process of topping out in China. The absolute number of working-age people will shrink in China and the G7 countries over the next five years, although it will continue to grow at a low rate for the world as a whole (Chart II-9). Unions are unlikely to make a major comeback, but a backdrop that is less labor-abundant should gradually restore some worker bargaining power, especially as economies regain full employment. The resulting upward pressure on real wages will support capital spending as firms substitute toward capital and away from (increasingly expensive) labor. Consumer demand will also receive a boost if inequality moderates and the labor share of income begins to rise. Globalization On The Back Foot Chart II-10Globalization Peaking? Globalization Peaking? Globalization Peaking? Second, it is too early to declare globalization dead, but the neo-liberal trading world order that has been in place for decades is under attack. Global exports appear to have peaked relative to GDP and average tariffs have ticked higher (Chart II-10). The World Trade Organization has announced that the number of new trade restrictions or impediments outweighed the number of trade liberalizing initiatives in 2016. The U.K. appears willing to sacrifice trade for limits to the free movement of people. The new U.S. Administration has ditched the Trans-Pacific Partnership (TPP) and is threatening to impose punitive tariffs on some trading partners. Anti-globalization policies could paradoxically be positive for capital spending, at least for a few years. If the U.S. were to impose high tariffs on China, for example, it would make a part of the Chinese capital stock redundant overnight. In order for the global economy to produce the same amount of goods and services as before, the U.S. and other countries would need to invest more. Any unwinding of globalization would also be inflationary as it would disrupt international supply chains. Demographics And Saving: From Tailwind To Headwind... Third, the impact of savings in the major advanced economies and China on global interest rates will change direction as well. In the developed world, aggregate household savings will come under downward pressure as boomers increasingly shift into retirement. Economists are fond of employing the so-called life-cycle theory of consumer spending. According to this theory, consumers tend to smooth out lifetime spending by accumulating assets during the working years in order to maintain a certain living standard after retirement. The U.N. National Transfer Accounts Project has gathered data on spending and labor income by age cohort at a point in time. Chart II-11 presents the data for China and three of the major advanced economies. Chart II-11Income And Consumption By Age Cohort Income And Consumption By Age Cohort Income And Consumption By Age Cohort The data for the advanced economies suggest that spending tends to rise sharply from a low level between birth and about 15 years of age. It continues to rise, albeit at a more modest pace, through the working years. Other studies have found that consumer spending falls during retirement. Nonetheless, these studies generally include only private spending and therefore do not include health care that is provided by the government. The data presented in Chart II-11 show that, if government-provided health care is included, personal spending rises sharply toward the end of life. The profile is somewhat different in China. Spending rises quickly from birth to about 20 years of age, and is roughly flat thereafter. Indeed, consumption edges lower after 75-80 years of age. These data allow us to project the impact of changing demographics on the average household saving rate in the coming years, assuming that the income and spending profiles shown in Chart II-11 are unchanged. We start by calculating the average saving rate across age cohorts given today's age structure. We then recalculate the average saving rate each year moving forward in time. The resulting saving rate changes along with the age structure of the population. The results are shown in Chart II-12. The saving rates for all four economies have been indexed at zero in 2016 for comparison purposes. The aggregate saving rate declines in all cases, falling between 4 and 8 percentage points between 2016 and 2030. Germany sees the largest drop of the four countries. Chart II-12Aging Will Undermine Aggregate Saving Aging Will Undermine Aggregate Saving Aging Will Undermine Aggregate Saving The simulations are meant to be suggestive, rather than a precise forecast, because the savings profile across age cohorts will adjust over time. Moreover, governments will no doubt raise taxes to cover the rising cost of health care, providing a partial offset in terms of the national saving rate.5 Nonetheless, the simulations highlight that the major economies are past the point where the baby boom generation is adding to the global savings pool at a faster pace than retirees are drawing from it. The age structure in the major advanced economies is far enough advanced that the rapid increase in the retirement rate will place substantial downward pressure on aggregate household savings in the coming years. It is well known that population aging will also undermine government budgets. Rising health care costs are already captured in our household saving rate projection because the data for household spending includes health care even if it is provided by the public sector. However, public pension schemes will also be a problem. To the extent that politicians are slow to trim pension benefits and/or raise taxes, public pension plans will be a growing drain on national savings. Could younger, less developed economies offset some of the demographic trends in China and the Advanced Economies? Numerically speaking, a more effective use of underutilized populations in Africa and India could go a long way. Nevertheless, deep-seated structural problems would have to be addressed and, even then, it is difficult to see either of these regions turning into the next "China story" given the current backlash against globalization and immigration. ...And The Capex Story Is Largely Behind Us Demographic trends also imply less capital spending relative to GDP, as discussed above. In terms of the impact on global equilibrium interest rates, it then becomes a race between falling saving and investment rates. Chart II-13Demographics And Capex Requirements May 2017 May 2017 Some analysts point to the Japanese experience because it is the leading edge in terms of global aging. Bond yields have been extremely low for many years even as the household saving rate collapsed, suggesting that ex-ante investment spending shifted by more than ex-ante savings. Nonetheless, Japan may not be a good example because the deterioration in the country's demographics coincided with burst bubbles in both real estate and stocks that hamstrung Japanese banks for decades. A series of policy mistakes made things worse. Economic theory is not clear on the net effect of demographics on savings and investment. The academic empirical evidence is inconclusive as well. However, a detailed IMF study of 30 OECD countries analyzed the demographic impact on a number of macroeconomic variables, including savings and investment.6 They estimated separate demographic effects for the old-age dependency ratio and the working-age population ratio. Applying the IMF's estimated model coefficients to projected changes in both of these ratios over the next decade suggests that the decline in ex-ante savings will exceed the ex-ante drop in capex requirements by about 1 percentage point of GDP. This is a non-trivial shift. Moreover, our simulations highlight that timing is important. The outlook for the household saving rate depends on the changing age structure of the population and the distribution of saving rates across age cohorts. Thus, the average saving rate will trend down as populations continue to age over the coming decades. In contrast, the impact of demographics on capital spending requirements is related to the change in the growth rate of the working-age population. Chart II-13 once again presents our estimates for the demographic component of capital spending. The top panel presents the world capex/GDP ratio that is necessary to maintain a constant capital/output ratio, and the bottom panel shows the change in that ratio. The important point is that the downward adjustment in world capex/GDP related to aging is now largely behind us because most of the deceleration in the growth rate of the working-age population is done. This is in contrast to the household saving rate adjustment where all of the adjustment is still to come. China Is Transitioning Too Chart II-14China's Savings Rates Have Peaked... China's Savings Rates Have Peaked... China's Savings Rates Have Peaked... China must be treated separately from the developed countries because of its unique structural issues. As discussed above, household savings increased dramatically beginning in the mid-1990s (Chart II-14). This trend reflected a number of factors, including: the rising share of the working-age population; a drop in the fertility rate, following the introduction of the one-child policy in the late 1970s that allowed households to spend less on raising children and save more for retirement; health care reform in the early 1990s required households to bear a larger share of health care spending; and job security was also undermined by reform of the state-owned enterprises (SOE) in the late 1990s, leading to increased precautionary savings to cover possible bouts of unemployment. These savings tailwinds have turned around in recent years and the household saving rate appears to have peaked. China's contribution to the global pool of savings has already moderated significantly, as measured by the current account surplus. The surplus has withered from about 9% in 2008 to 2½% in 2016. A recent IMF study makes the case that China's national saving rate will continue to decline. The IMF estimates that for every one percentage-point rise in the old-age dependency ratio, the aggregate household saving rate will fall by 0.4-1 percentage points. In addition, the need for precautionary savings is expected to ease along with improvements in the social safety net, achieved through higher government spending on health care. The household saving rate will fall by three percentage points by 2021 according to the IMF (Chart II-15). Competitive pressure and an aging population will also reduce the saving rates of the corporate and government sectors. Chart II-15...Suggesting That External Surplus Will Shrink ...Suggesting That External Surplus Will Shrink ...Suggesting That External Surplus Will Shrink Of course, investment as a share of GDP is projected to moderate too, reflecting a rebalancing of the economy away from exports and capital spending toward household consumption. The IMF expects that savings will moderate slightly faster than investment, leading to a narrowing in the current account surplus to almost zero by 2021. A lot of assumptions go into this type of forecast such that we must take it with a large grain of salt. Nonetheless, the fundamental drivers of China's savings capacity appear to rule out a return to the days when the country was generating a substantial amount of excess savings. Moreover, a return to large current account surpluses would likely require significant currency depreciation, which is a political non-starter given U.S. angst over trade. The risk is that China's excess savings will be less, not more, in five year's time. Tech Is A Wildcard It is extremely difficult to forecast the impact of technological advancement on the global economy. We cannot say with any conviction that the tech-related effects of "hollowing out", "winner-take-all" and the "skills premium" will moderate in the coming years. Nonetheless, these effects have occurred alongside a surge in the world's labor force and rapid globalization of supply chains, both of which reinforced the erosion of employee bargaining power. Looking ahead, technology will still be a headwind for some employees, but at least the transition from a world of excess labor to one that is more labor-scarce will boost workers' ability to negotiate a larger share of the income pie. We will explore the impact of technology on productivity, inflation, growth, and bond yields in a companion report to be published in the next issue. Conclusion: The main points we made in this report are summarized in Table II-1. All of the structural factors driving real bond yields were working in the same (bullish) direction over the past 30-40 years. Looking ahead, it is uncertain how technological improvement will affect bond prices, but we expect that the others will shift (or have already shifted) to either neutral or outright bond-bearish. Table II-1Key Secular Drivers May 2017 May 2017 No doubt, our views that globalization and inequality have peaked, and that the labor share of income has bottomed, are speculative. These factors may not place much upward pressure on equilibrium yields. Nonetheless, it seems likely that the demographic effect that has depressed capital spending demand is well advanced. We see it shifting from a positive factor for bond prices to a neutral factor in the coming years. It is also clear that the massive positive labor supply shock is over, and is heading into reverse as the global working-age population ratio falls. This may improve labor's bargaining power and the resulting boost consumer spending will be negative for bonds. This may also spark a shift toward using more capital in the production process and thereby place additional upward pressure on global real bond yields. Admittedly, however, this last point requires more research because theory and empirical evidence on it are not clear. Perhaps most importantly, the aging of the population in the advanced economies has reached a tipping point; retirees will drain more from the pool of savings than the working-age population will add to it in the coming years. We have concentrated on real equilibrium bond yields in this report because it is the part of nominal yields that is the most depressed relative to historical norms. The inflation component is only a little below a level that is consistent with central banks meeting their 2% inflation targets in the medium term. There is a risk that inflation will overshoot these targets, leading to a possible surge in long-term inflation expectations that turbocharges the bond bear market. This is certainly possible, as highlighted by a recent Global Investment Strategy Quarterly Strategy Outlook.7 Pain in bond markets would be magnified in this case, especially if central banks are forced to aggressively defend their targets. Please note that we are not making the case that real global bond yields will quickly revert to pre-Lehman averages. It will take time for the bond-bullish structural factors to unwind. It will also take time for inflation to gain any momentum, even in the United States. Global yields could even drop back to previous lows in the event of another recession. Nonetheless, from a long-term perspective, current market expectations suggest that investors have adopted an overly benign view on the outlook for yields. For example, implied real short-term rates remain negative until 2021 in the U.S. and 2026 in the Eurozone, while they stay negative out to 2030 in the U.K. (Chart II-16). We doubt that short-term rates will be negative for that long, given the structural factors discussed above. Chart II-16Market Expects Negative Short-Term Rates For A Long Time Market Expects Negative Short-Term Rates For A Long Time Market Expects Negative Short-Term Rates For A Long Time Another way of looking at this is presented in Chart II-17. The market expects the 10-year Treasury yield in ten years to be only slightly above today's spot yield, which itself is not far above the lowest levels ever recorded. Market expectations are equally depressed for the 5-year forward rate for the U.S. and the other major economies. Chart II-17Forward Rates Very Low Vs. History Forward Rates Very Low Vs. History Forward Rates Very Low Vs. History The implication is that investors should have a bond-bearish bias on a medium- and long-term horizon. Mark McClellan Senior Vice President The Bank Credit Analyst 1 It is true that observed household savings rates fell in some of the advanced economies, such as the United States, at a time when aging should have boosted savings from the mid-1990s to the mid-2000s. This argues against a strong demographic effect on savings. However, keep in mind that we are discussing desired (or ex-ante) savings. Ex-post, savings can go in the opposite direction because of other influencing factors. As discussed below, global savings must equal investment, which means that shifts in desired capital spending demand matter for the ex-post level of savings. 2 Arithmetically, if world trend GDP growth slows by one percentage point, then investment spending would need to drop by about 3½ percentage points of GDP to keep the capital/output ratio stable. 3 David H. Autor, David Dorn, and Gordon H. Hanson, "The China Shock: Learning from Labor-Market Adjustment to Large Changes in Trade," Annual Review of Economics, Vol. 8, pp. 205-240 (October 2016). 4 Please see "Understanding The Downward Trend In Labor Income Shares," Chapter 3 in the IMF World Economic Outlook (April 2017). 5 In other words, while the household savings rate, as defined here to include health care spending by governments on behalf of households, will decline, any associated tax increases will blunt the impact on national savings (i.e. savings across the household, government and business sectors). 6 Jong-Won Yoon, Jinill Kim, and Jungjin Lee, "Impact Of Demographic Changes On Inflation And The Macroeconomy," IMF Working Paper no. 14/210 (November 2014). 7 Please see Global Investment Strategy, "Strategy Outlook: Second Quarter 2017: A Three-Act Play," dated March 31, 2017, available at gis.bcaresearch.com.

Highlights The earnings rebound underway in Corporate America is being driven by more than just higher oil prices. S&P 500 profit margins have stabilized recently, but remain in secular decline. We remain bullish on the dollar and the other "Trump Trades" have legs as well. Uncertainty around tax policy may be restraining business capital spending and C&I loan growth. Feature Chart 1Excluding Energy Earnings Rebounding Excluding Energy Earnings Rebounding Excluding Energy Earnings Rebounding The so-called "Trump trades" have either stalled or partially reversed. The failure to reform Obamacare has dented hopes that the Administration and GOP will get a tax reform package done this year. The S&P 500 is not far off its all time high, but Treasury yields have returned to the bottom of the trading range and the dollar has weakened (although it has risen over the past 3 weeks). We still believe that the Republicans will at least push through tax cuts and some infrastructure spending this year, which will be stimulative for the economy. However, the 12-month outlook for the stock-to-bond ratio does not hinge solely on U.S. fiscal policy. As we have highlighted in the past, the underlying fundamentals for equities are positive, despite the fact that we see more dollar upside (see below). First quarter earnings season is about to kick off, and it should be another good one. Before we discuss the outlook for profits, let's review the fourth quarter of 2016. S&P 500 firms posted profit growth of 6% on a 4-quarter moving total year-over-year basis. The Q4 reading beat consensus bottom-up expectations at the start of earning season but were roughly in line with expectations at the start of Q4 2016 itself. The fourth quarter increase was the best year-over-year EPS gain since Q3 2014 - just after the oil price peak- and the first year-over-year increase in the 4-quarter sum since Q3 2015. Energy sector earnings posted a 6% advance in Q4, as oil prices averaged close to $49 per barrel in Q4 2016, up 17% from Q4 2015. It was the first time that oil prices posted a year-over-year increase in a quarter since Q2 2014. Part of the acceleration in earnings reflects the rise in oil prices from the Q1 2016 bottom, but higher energy prices are not the only factor driving the turnaround (Chart 1). Overall, 9 of the 11 S&P 500 sectors saw positive year-over-year profit gains in Q4 2016, led by technology (13%), financials (12%) and utilities (10%). In addition, Consumer Discretionary, Financials and Health Care all posted solid earnings figures in the last year. Earnings momentum has also picked up in Materials, Real Estate and Utilities, although profit growth in these sectors is also benefiting from favorable comparisons. Eighty-eight percent of technology firms posted Q4 results that beat expectations, as did 80% of health care companies and 75% of financials, so the market was caught somewhat off guard by the pace of the upturn in earnings outside of energy. While earnings grew at 6% year-over-year in Q4 2016, revenues grew just 4% due to low nominal GDP growth last year (although the latter rebounded late in the year). Ten of 11 sectors posted year-over-year revenue increases in Q4, but the revenue gain just matched consensus estimates with only half of firms posting revenues that exceeded already low expectations. In short, the market didn't expect much and didn't get much from revenues in Q4. The Marginal Way: A Top Down View Looking ahead, a secular downtrend in margins will be a headwind for earnings growth in the coming years, as we highlighted in the February 27, 2017 Weekly Report. A "mean reversion" process for margins is underway, as a tight labor market pushes up wages but firms have difficulty passing along the cost pressure in a poor environment for pricing power. For large cap U.S. companies, global GDP is a better proxy for revenue than U.S. GDP. Nominal global GDP growth fell 6% year-over-year in 2015, but rebounded to a 2%+ increase in 2016 and the World Bank expects global GDP to accelerate rapidly to a 6% increase here in 2017. Thus, there is scope for U.S. corporate revenue growth to pick up after a long period of deceleration. Indeed, the risks for global growth are to the upside of consensus estimates in our view (Chart 2). For those industries and sectors with mainly domestic sales (utilities, telecom), U.S. GDP is a better proxy for top line sales. At just 3.0%, U.S. nominal GDP growth was disappointing in 2016, running 340 basis points below its long-term average (6.4%) and nearly a full percentage point shy of the 2010-2014 (post Great Recession but pre-oil price decline) average of 3.8%. We expect nominal GDP growth to accelerate this year, even absent potentially growth-enhancing legislation from Congress on tax cuts, tax reform and infrastructure. Compensation costs represent two thirds of business costs, and various measures of wage gains are slowly climbing as the U.S. economy approaches full employment. Average hourly earnings rose 2.7% in March 2017 versus a year ago, up from a low of 1.5% hit in 2012. The Employment Cost Index is accelerating as well. The Atlanta Fed's Wage tracker has been trending higher for 7 years, not coincidentally, along with service sector inflation. The Atlanta Fed wage tracker shows the same pattern for both job stayers and job seekers (Chart 3). Chart 2Global Growth Accelerating Global Growth Accelerating Global Growth Accelerating Chart 3Wage Pressures Building Wage Pressures Building Wage Pressures Building The quit rate from the BLS's JOLTs data has hit a new cycle high and is within striking distance of an all-time high. This is significant because a high quit rate means that job prospects are favorable and that employees are jumping to new jobs in search of higher wages. In addition, mentions of wages, skilled labor, and shortages in the Fed's Beige Book have been on the upswing for four years (Chart 4). Labor costs are rising faster than selling prices in the non-financial corporate sector, as highlighted by the downtrend in BCA's Profit Margin Proxy (Chart 5, Panel 1). The mean reversion process will continue, but that does not preclude periods of margin expansion. Indeed, margins rose in the third and fourth quarters on a four quarter moving total basis according to S&P data and we would not be surprised to see this continue early in 2017 as nominal GDP growth recovers from last year's depressed pace (Chart 5, Panel 2). Chart 4"Inflation Words" On The Rise "Inflation Words" On The Rise "Inflation Words" On The Rise Chart 5Bullish Profit Model Bullish Profit Model Bullish Profit Model What about the dollar? As we discuss below, BCA believes that the dollar bull market still has legs. A stronger dollar is both a blessing and a curse for margins. All else equal, a stronger dollar lowers the cost of imported goods and thereby boosts margins for import-intensive firms. On the other hand, a strong dollar undermines profits earned overseas. The net impact of dollar strength is negative for overall corporate profits. However, our quantitative work highlights that it does not take much in the way of stronger growth to offset the negative impact on profits from a rise in the dollar. Investors are also concerned about the impact of higher interest rates on corporate income statements, especially given all the corporate debt that has been accumulated. While we agree with the conventional wisdom that interest costs as a percent of sales have likely bottomed for the cycle, and will undermine margins if yields rise, research by the monthly Bank Credit Analyst revealed that it will require a large increase in interest rates to 'move the dial' on interest payments.1 This is because of a long maturity distribution and the fact that the average yield-to-maturity is still so far below the average coupon in the corporate debt indices that average coupons will continue to erode as debt rolls over in the coming years. Chart 6 shows that interest payments as a fraction of GDP will be roughly flat even if the yield curve shifts up by another 100 basis points in the near term. It would require a 200-300 basis point rise in yields to see a meaningful impact on interest payments over the next 1-2 years. The implication is that rising interest costs won't be a key driver of profit margins in our investment horizon. Chart 6U.S. Corporate Sector Interest Payment Projection U.S. Corporate Sector Interest Payment Projection U.S. Corporate Sector Interest Payment Projection Despite our secular view on profit margins, we remain upbeat for EPS growth this year. Our profit model remains constructive. Indeed, EPS growth for the year may not trail (perennially overly optimistic) bottom-up estimates for the year, currently at 10%. In short, we see a potential for upside surprise on earnings this year, although growth will not be as high as our short-term profit model suggests (Chart 5, Panel 3). Bottom Line: We certainly would not rule out a pullback in the S&P 500 on disappointment surrounding a lack of follow-through by Congress and the Trump Administration on a tax cut, tax reform and an infrastructure package. However, fears around margin contraction, the sustainability of the earnings rebound and valuations are overdone. Earnings estimates almost always come down over the course of the year. Moreover, while above-average valuations suggest below average-returns over the next decade, valuation tells us little about returns over the next 12 months. We continue to favor stocks over bonds in 2017. Is The Dollar Bull Over? The dollar has firmed over the past couple of weeks but it remains below the December high in trade-weighted terms. Is this just a consolidation phase? Or has the dollar peaked for this cycle because the maximum policy divergence between the Fed and the other major central banks is now in the price? Indeed, the global growth outlook outside of the U.S. has brightened at a time when some of the so-called "hard" U.S. economic data have disappointed and the promised Trump fiscal stimulus appears to be on the ropes. The European Central Bank (ECB) has already tapered its asset purchase program once and is expected to do so again early in 2018. Some are even speculating that the ECB will lift rates in the not-to-distant future. This raises the possibility that the bund yield curve begins to converge with the Treasury curve, placing upward pressure on the euro versus the dollar. The Eurozone economic data have certainly been stellar so far this year. The PMIs for manufacturing and services both pulled back a bit in March, but remain at levels consistent with continued above-trend growth. The uptrend in capital goods orders bodes well for investment spending over the coming months (Chart 7). In addition, private-sector credit growth has accelerated to the fastest pace since the 2008-09 financial crisis. Our real GDP model for the Eurozone, based on our consumer and business spending indicators, remains quite upbeat for the first half of the year. The Eurozone unemployment rate is falling fast and there is less spare capacity in European labor markets today than was the case in the U.S. when the Fed first hinted at tapering its asset purchases in 2013 (Chart 8). Chart 7Solid Eurozone##br## Economic Data Solid Eurozone Economic Data Solid Eurozone Economic Data Chart 8Less Spare Capacity In Europe Now ##br##Vs. Pre-Taper Tantrum U.S. Less Spare Capacity In Europe Now Vs. Pre-Taper Tantrum U.S. Less Spare Capacity In Europe Now Vs. Pre-Taper Tantrum U.S. Nonetheless, the calm readings on Euro Area core inflation suggest that the ECB does not have to rush to judgment on asset purchases, especially given upcoming elections. Our diffusion index for the components of the CPI points to some upside for core inflation in the coming months, but it fell back to 0.7% in March according to the flash estimate. The ECB will probably not feel comfortable announcing the next tapering until September of this year. But even then, policymakers will apply a heavy dose of "forward guidance" on the outlook for short-term rates in order to avoid an outsized impact on Eurozone bond yields. Some tapering is presumably already discounted in rates and the euro. Chart 9Market Is Reassessing The FOMC Market Is Reassessing The FOMC Market Is Reassessing The FOMC It will be much longer before the Bank of Japan is in any position to begin removing monetary accommodation. We expect that the 0% yield cap on the 10-year JGB to remain in place at least for the remainder of this year, and probably much longer. True, deflationary forces appear to have eased somewhat. Japan is also benefiting from the faster global growth on the industrial side. Nonetheless, the domestic demand story is less positive, with consumer confidence and real retail sales growth languishing. Wages continue to struggle as well. This year's round of Japanese wage negotiations was particularly disappointing, with many manufacturing companies offering pay raises only half as large as those of last year. We continue to see this as the only way out of the low-inflation trap for Japan - keeping Japanese nominal interest rates depressed versus the rest of the world, thus making the yen weaken alongside increasingly unattractive interest rate differentials. On the U.S. side, we believe that the market has over-reacted when the FOMC signaled last month that it was not yet prepared to adjust the 'dot plot.' The market is discounting only two rate hikes over the next 12 months, down by about 10 basis points since the FOMC meeting (Chart 9). The market view is too complacent for three reasons. First, we expect the U.S. "hard" to catch up with the more robust "soft" data readings in the coming months. Second, the FOMC did not signal a more dovish mindset last month. The key message from the March meeting was that the Fed now sees inflation as having finally reached its 2% target, as highlighted by the decision to strip the reference to the "current shortfall of inflation" from the statement. If the U.S. economy performs as we expect, the Fed will have to take a more hawkish tone later this year. The poor (weather-related) March payroll report does not change the Fed outlook. The important point is that the market appears to be at full employment based on FOMC committee projections. In fact at 4.5% in March (the lowest since May 2007) the rate is below the median and midpoint of the FOMC's long-run forecast, of respectively 4.7% and 4.85%. Finally, the market is underestimating the prospects for stimulative tax cuts and infrastructure spending. The Republican's desire to cut taxes will dominate fears of blowing out the budget deficit. The resulting stimulus will add pressure on the FOMC to tighten monetary conditions. Bottom Line: Our views on U.S. fiscal policy and the outlook for the major central banks paint a bullish picture for the dollar and suggest that the other 'Trump trades' still have legs. The dollar has another 10% upside in trade-weighted terms as yield spreads move further in favor of the greenback, but a move of that magnitude wouldn't be a major headwind for U.S. corporate earnings growth and would pale in comparison to the hit earnings took from the 20-25% gain in the dollar in late 2014 through early 2016. Our view remains that the U.S. bond bear phase is not yet over. Revisiting "Weak" U.S. CAPEX The BCA Model for business investment tracks broad capex swings and has been trending down for several months now. Our past research shows that sustainable capital spending cycles only get underway once businesses see clear evidence that consumer final demand is on the upswing. Comments from management during the recent Q4 2016 earnings reporting season were upbeat, but cautious, and there is some evidence (the recent rollover in C&I) loans that businesses may be delaying some portion of capital spending until after tax cuts and or tax reform is enacted by Congress. Part of the macroeconomic narrative for many investors over the past several years is that U.S. growth has been slow this cycle because private investment has been weak. The prolonged nature of "weak" U.S. investment during this economic recovery has been offered as evidence of deep-seated structural problems by many market participants, and arguably remains a factor driving the continued prevalence of the secular stagnation narrative. Two elements of the "weak investment" narrative are undeniably true. First, overall investment has indeed grown at a sluggish pace over the past eight years relative to previous economic expansions. Second, residential investment has certainly been weak by any measure, which is to be expected given that housing was at the epicenter of the subprime financial crisis. However, Chart 10 presents a different perspective about the "weakness" of investment by examining the trend in non-residential fixed asset investment (i.e., capex). The chart shows that, relative to GDP, capex has not been weak at all this cycle: it experienced a V-shaped recovery over the past several years, and has risen either back to its post-1980 average (in nominal terms) or to a new high (in real terms). This highlights that growth in investment, abstracting from the housing effect, has been weak in absolute terms because consumption has been weak, rather than because of some other unexplained structural force. Chart 10Investment Has Not Been Weak Relative To GDP Investment Has Not Been Weak Relative To GDP Investment Has Not Been Weak Relative To GDP More recently, Chart 10 shows that there has been a decline in the capex-to-GDP ratio, which has been a concerning sign for some investors that U.S. growth may be faltering. Until the beginning of last year, this deceleration could have been simply blamed on a collapse in resource investment following the sharp decline in the price of oil that began in mid-2014. But Chart 11 shows that this ceased to be the case through to the fourth quarter, as real capex excluding mining structures has also decelerated sharply. The slowdown in capex last year is echoed by a sharp recent slowdown in U.S. bank lending, and a detailed analysis suggests they may both be (at least somewhat) related to the same cause. Chart 12 presents the 3-month annualized rate of change in commercial & industrial (C&I) loans, along with the U.S. Economic Policy Uncertainty Index. The recent spikes in the latter correspond with the U.K.'s vote to leave the European Union as well as the U.S. election in November, and the chart clearly shows a close correlation between these spikes and the deceleration in C&I loan growth. Indeed, C&I lending had begun to pick up again following the Brexit vote, only to decelerate again after November. Chart 11Oil Accounts For Some, But Not All, ##br##Of Recently Weak CAPEX Oil Accounts For Some, But Not All, Of Recently Weak CAPEX Oil Accounts For Some, But Not All, Of Recently Weak CAPEX Chart 12Tax Rule Certainty May Spur Bank##br## Lending And Investment Tax Rule Certainty May Spur Bank Lending And Investment Tax Rule Certainty May Spur Bank Lending And Investment Uncertainty over Brexit represented legitimate CEO concern about a potential global macro shock, but our view is that the recent uncertainty following the U.S. election has not been driven by fear. This is a crucial distinction with implications for the economic outlook: if the recent uptick has been driven by a dearth of information about how business-friendly fiscal policy will become as a result of the election, then investors are more likely observing uncertainty over how much and when to invest rather than whether to invest. If true, this suggests that weak bank lending and growth in non-resource capex in Q4 has merely been deferred until rule clarity emerges and firms are confident that they will benefit from any investment-related changes to the tax code. In short, far from being a bearish signal about economic activity, recent trends in C&I lending and non-resource capex may actually indicate that firms plan on responding positively to corporate tax relief, suggesting that overall economic growth may improve once the details of the plan are known. Bottom Line: A detailed analysis of recent weakness in C&I lending and non-resource capex points to policy-related uncertainty as the culprit, rather than impending economic weakness or a broad-based contraction in activity. This argues that some capex spending is pent up, and that economic growth will improve following the establishment of tax rule certainty by the Trump administration and/or congressional leadership. John Canally, CFA, Senior Vice President U.S. Investment Strategy johnc@bcaresearch.com Mark McClellan, Senior Vice President The Bank Credit Analyst markm@bcaresearch.com Jonathan LaBerge Vice President, Special Reports Jonathanl@bcaresearch.com 1 Please see The Bank Credit Analyst Monthly Report "Global Debt Titanic Collides With Fed Iceberg?", dated February, 2017, available at bca.bcaresearch.com

Feature The on-going Second Machine Age - the ushering in of Artificial Intelligence (AI) - is obsoleting human jobs at an alarming pace. But an analysis of the data reveals an interesting pattern. Jobs typically done by men have suffered disproportionately, and continue to be in grave danger. On the other hand, jobs that are typically performed by women are thriving. Chart I-1AFemale Labour Participation ##br##Can Rise A Lot Further Female Labour Participation Can Rise A Lot Further Female Labour Participation Can Rise A Lot Further Chart I-1BFemale Labour Participation ##br##Can Rise A Lot Further Female Labour Participation Can Rise A Lot Further Female Labour Participation Can Rise A Lot Further AI Is A Greater Threat To Men Despite decades of progress in workplace equality, most jobs and professions still have a very strong gender bias. The big problem for men is that automation is ideally suited to replace jobs in male-dominated middle-income jobs in manufacturing, construction and transport (Table I-1). For example, the advent of autonomous or semi-autonomous vehicles will destroy livelihoods that involve driving. And 95% of truck drivers are men. Table I-1AI Is A Greater Threat To Men Female Participation: Another Mega-Trend Female Participation: Another Mega-Trend Even among more-skilled jobs, male-dominated professions - such as finance - are under threat. As we explained in The Superstar Economy: Part 2,1 seemingly difficult human tasks that AI finds very easy are those that require the application of complex algorithms and pattern recognition to large quantities of data. This includes tasks such as calculating a credit score or insurance premium, or managing a stock portfolio. Conversely, seemingly easy human tasks that AI finds very difficult are those that require everyday human sensorimotor skills. Tasks such as walking up a flight of stairs or picking up random objects from random places. AI is also incapable (thus far) of reading and responding to peoples' emotions and intentions. A job that requires a range of these human skills is especially secure. This is because AI excels at replicating a narrowly defined task rather than a job which needs a breadth of talents. If your job involves controlling and teaching a disruptive class of seven year olds, or calming a nervous patient before giving an injection, AI will not threaten your livelihood for the foreseeable future. Hence, Education, Human Health and Social Work - the employment sectors that most require a combination of sensorimotor, emotional and communication skills - have seen the strongest job gains over the past two decades. And almost 80% of workers in Education, Human Health and Social Work are women. With AI still in its infancy, the established pattern of job destruction and creation will continue to favour women over men. Improved Parental Leave Helps Women Chart I-2Japan: Labour Force Participation Rate Japan: Labour Force Participation Rate Japan: Labour Force Participation Rate A second structural driver that is boosting female employment is improved parental leave policies. Japan provides an excellent example of what is possible. Starting in 1995, the Japanese government introduced a raft of policies to encourage women to join the labour force: paid maternity leave, subsidised childcare, and paid parental leave for both parents. Today in Japan, both mothers and fathers can take more than a year of paid parental leave at an average rate of 60% of earnings. The policies have had their desired effect. The proportion of Japanese women in the labour force has surged from 57% to 67%, while the male labour participation rate has stayed unchanged at 85%. Therefore, all of the growth in the Japanese labour force through the past 20 years has come from women (Chart I-2). Europe tells a similar tale. Through the past couple of decades, parental leave policies have become steadily more generous. Today, new mothers are guaranteed 58 weeks of paid leave in Germany, 48 in Italy, and 42 in France (Table I-2). New fathers are guaranteed 28 weeks in France and 9 weeks in Germany (Table I-3). Moreover, even after the paid parental leave ends, heavily subsidised childcare costs in the major euro area countries are affordable, averaging around a tenth of the average wage (Table I-4). Table I-2Generous Maternity Pay ##br##In Europe And Japan Female Participation: Another Mega-Trend Female Participation: Another Mega-Trend Table I-3Improving Paternity Pay##br## In Europe And Japan Female Participation: Another Mega-Trend Female Participation: Another Mega-Trend Table I-4Affordable Childcare##br## In Europe And Japan Female Participation: Another Mega-Trend Female Participation: Another Mega-Trend Unsurprisingly, the proportion of European women in the labour force has also surged from 57% to 67%, while the male labour participation rate has stayed flat at 78%. As in Japan, all of the growth in European labour force participation through the past 20 years has come from women (Chart I-3). The surge in female participation also explains why the percentage of the euro area working-age population in employment now stands close to at an all-time high (Chart I-4) - a fact which stuns many people. Chart I-3EU28: Labour Force##br## Participation Rate EU28: Labour Force Participation Rate EU28: Labour Force Participation Rate Chart I-4Increased Female Participation Lifts Euro Area ##br##Employment To Population Near An All Time High Increased Female Participation Lifts Euro Area Employment To Population Near An All Time High Increased Female Participation Lifts Euro Area Employment To Population Near An All Time High The trend is for further improvements in parental leave, with the focus now on improving paternity leave. The important point is that the sharing of parental responsibilities between mothers and fathers allows more women to enter and stay in the labour force. Therefore, we expect the structural rise in European female labour force participation to continue. France at 67% and especially Italy at 55% are still some way behind Germany at 73% and Spain at 70% (Chart I-5, Chart I-6, Chart I-7, Chart I-8 and Chart I-9). For the ultimate end-point in the trend, look to the Scandinavian countries which have had generous parental leave policies for decades. In Sweden, labour force participation for women is almost identical to that for men: 80% versus 83%. If the EU eventually reaches the Scandinavian end-point, it would mean another 20 million women in the EU labour force (Feature Chart). The one major world economy that goes against the trend is the United States. Over the past 20 years, the proportion of American women who are in the labour force has actually declined from 70% to 67%. Why? Possibly because in the U.S. new parents have no guarantee of paid leave. And the cost of childcare has been increasing much faster than overall inflation. Childcare now costs close to half the average wage, making it unaffordable for many low-income workers. Nevertheless, even in the U.S., labour force participation for women is outperforming that for men. Because the decline in participation for women is mild, whereas the decline for men is more severe (Chart I-10). Chart I-5Germany:##br## Labour Force Participation Rate Germany: Labour Force Participation Rate Germany: Labour Force Participation Rate Chart I-6France: ##br##Labour Force Participation Rate France: Labour Force Participation Rate France: Labour Force Participation Rate Chart I-7Italy:##br## Labour Force Participation Rate Italy: Labour Force Participation Rate Italy: Labour Force Participation Rate Chart I-8Spain:##br## Labour Force Participation Rate Spain: Labour Force Participation Rate Spain: Labour Force Participation Rate Chart I-9U.K.:##br## Labour Force Participation Rate U.K.: Labour Force Participation Rate U.K.: Labour Force Participation Rate Chart I-10U.S.: ##br##Labour Force Participation Rate U.S.: Labour Force Participation Rate U.S.: Labour Force Participation Rate The Structural Investment Theme: Personal Products The two factors driving increased female participation in the total labour force have much further to run. First, AI is a much greater threat to jobs and professions that are male-dominated than to those that are female-dominated. Second, further improvements in parental leave policies will allow labour force participation for women to gradually converge with that for men. All mega-trends have an associated structural investment theme. And in this case the theme is the Personal Products sector. According to Euromonitor, 90% of personal product sales are to women. For cosmetics, the proportion is close to 100%. Therefore, as the percentage of women in the labour force continues to rise, the sales and profits of Personal Products will continue to outperform those from other sectors. The highly defensive nature of personal product demand is also a big advantage in a lower-growth world. Once again, Japan provides an excellent example of what is possible. Since its crisis in 1990, overall equity market profits and prices have gone nowhere, but the profits of the Personal Products sector - led by Shiseido and Kao - have increased fivefold. The European Personal Products sector is now following the Japanese template with a reassuring similarity (Chart I-11). Chart I-11Japan Is The Template Japan Is The Template Japan Is The Template Buy and hold L'Oréal, Beiersdorf and Unilever as long-term investments. Dhaval Joshi, Senior Vice President European Investment Strategy dhaval@bcaresearch.com 1 Published on January 19, 2017 and available at eis.bcaresearch.com

Highlights Duration & Fed Policy: The longer risk assets can withstand rising rates, the higher will be the ultimate resting place for Treasury yields. Maintain below-benchmark duration on a 6-12 month horizon and add a short fed funds futures trade to profit from increased Fed hawkishness in the near-term. Yield Curve: While the long-run trend will be for the yield curve to flatten as the Fed hiking cycle progresses, rising inflation expectations will cause the curve to steepen between now and the end of the year. Maintain a position long the 5-year bullet, short a duration-matched 2/10 barbell to profit from a steeper curve on a 6-9 month horizon. Feature Say Uncle Chart 1More Tightening To Come More Tightening To Come More Tightening To Come The Fed lifted rates last week but kept its median projected path for future rate hikes unchanged. Judging from the market's reaction, this was a more dovish outcome than was anticipated. Since last Wednesday's meeting the dollar is down 0.5%, junk spreads have tightened 10 basis points and the 2/10 yield curve has steepened 1 bp. In other words, financial conditions have continued to ease even as the Fed took another step toward more restrictive policy. All in all, money markets are now discounting only a slightly slower pace of rate hikes than the Fed's median forecast (Chart 1) and financial conditions suggest that further incremental tightening is in store. The financial conditions component of our Fed Monitor1 is above zero, meaning that financial conditions are more accommodative than the long-run average, and the Chicago Fed's Adjusted Financial Conditions Index also shows that conditions are easy relative to the strength of the economy (Chart 1, bottom panel). New York Fed President William Dudley has previously described how the Fed incorporates financial conditions into its decision making:2 Chart 2The Fed Policy Loop The Yield Curve On A Cyclical Horizon The Yield Curve On A Cyclical Horizon All else equal, if financial conditions tighten sharply, then we are likely to proceed more slowly. In contrast, if financial conditions were not to tighten at all or only very little, then - assuming the economic outlook hadn't changed significantly - we would likely have to move more quickly. In the end, we will adjust the policy stance to support financial market conditions that we deem are most consistent with our employment and inflation objectives. We have also described this process in the context of our Fed Policy Loop3 (Chart 2). In essence, the Fed will continue to nudge rate hike expectations higher until financial conditions tighten excessively. At that point - because with inflation below target the Fed still has an interest in supporting the recovery - it will quickly shift to a more dovish stance. Chart 3Short Jan 2018 Fed Funds Futures Short Jan 2018 Fed Funds Futures Short Jan 2018 Fed Funds Futures One implication of the Fed Policy Loop is that the longer risk assets can withstand rising rates, the higher will be the ultimate resting place for the fed funds rate and Treasury yields. As such, we continue to recommend a below-benchmark duration allocation on a 6-12 month horizon. Another implication is that because markets shrugged off the latest rate increase, Fed policy is likely to turn more hawkish in the very near term. We therefore recommend investors add a tactical trade: short the January 2018 fed funds futures contract (Chart 3). We calculate that this trade will return 11 bps in a scenario where the Fed lifts rates twice more before the end of the year and 37 bps in a scenario where the funds rate is raised three times. However, we do not expect to hold this trade until the end of the year. Rather, we expect the Fed will nudge rate expectations higher in the next month or two and that these gains will be realized over a much shorter horizon. We also add a caveat that, in the present environment, it is safer to implement any "hawkish Fed trades" in either fed funds futures or the overnight index swap market. The Eurodollar market does not provide the same potential for gains because the LIBOR / OIS spread is currently elevated and could tighten to offset the profits from rising fed funds rate expectations (Chart 3, bottom panel). Fed hawkishness also argues for a flatter yield curve in the very near term. While this could materialize, we continue to hold our position in the 5-year bullet over a duration-matched 2/10 barbell - a trade designed to profit from a steeper 2/10 slope. For reasons described in the next section we believe the yield curve will steepen between now and the end of the year, although the risks are tilted toward flattening in the very near term and in 2018 and beyond. What Drives The Yield Curve? In this week's report we present an overview of the main drivers of the slope of the Treasury yield curve. Specifically, we identify (i) the fed funds rate, (ii) inflation expectations, (iii) implied volatility and (iv) unit labor costs as factors that correlate strongly with the slope of the yield curve on a cyclical horizon. We review the outlook for each of these factors and conclude that the Treasury yield curve has room to steepen between now and the end of the year. Beyond that, the curve will likely resume flattening as inflationary pressures start to bite and the Fed's rate hike cycle picks up steam. Chart 4Fed Rate Hikes Flatten The Curve Fed Rate Hikes Flatten The Curve Fed Rate Hikes Flatten The Curve 1. The Fed Funds Rate Not surprisingly, the slope of the Treasury curve correlates very strongly with the level of short rates (Chart 4). Typically, short-maturity yields are much more influenced by the expected path of Fed rate hikes than long-maturity yields. As such, when the Fed is lifting rates the yield curve tends to bear-flatten - both the 2-year and 10-year Treasury yields rise, but the 2-year rises more quickly. In contrast, when the Fed is cutting rates the yield curve tends to bull-steepen - both the 2-year and 10-year Treasury yields fall, but the 2-year falls more quickly. In a typical cycle the yield curve will start to flatten as the Fed lifts rates and will eventually become completely flat when the end of the rate hike cycle is reached and the fed funds rate is at its "equilibrium" or "terminal" level. Usually, at that point in the cycle, the Fed will keep policy too tight in an effort to rein in inflation. This causes the economy to slow and the yield curve to invert, signaling the start of the next recession. A recent BCA Special Report4 speculates that if the federal government succeeds in delivering sizeable fiscal stimulus, inflationary pressures could start to build next year, leading to a more rapid pace of Fed rate hikes and a flat or inverted yield curve by the end of 2018. This would be consistent with a recession in 2019. In terms of the behavior of the yield curve, this is not far off from the Fed's own projections. At present, the median FOMC projection calls for the fed funds rate to reach its equilibrium level of 3% by the end of 2019. If this forecast plays out, it means that the 2/10 Treasury slope must flatten by roughly 117 bps between now and then. Turning back to Chart 4, we see that the Treasury curve has already flattened considerably even though the Fed has only raised rates three times. This means that either the equilibrium fed funds rate is much lower than the Fed's 3% projection and the 2/10 slope will reach zero with a much lower fed funds rate, or that the curve flattening is overdone and the curve has room to steepen before it resumes its cyclical flattening trend. As is explained below, we favor the latter interpretation. 2. Inflation Expectations The 5-year/5-year forward TIPS breakeven inflation rate is also highly correlated with the slope of the yield curve (Chart 5). As long-dated inflation expectations increase the yield curve tends to steepen, and vice-versa. Interestingly, the positive correlation between long-dated inflation expectations and the slope of the Treasury curve persists even when the Fed is hiking rates. Notice that in the 1999 rate hike cycle, the yield curve did not start to flatten until the 5-year/5-year breakeven fell. Also, in the 2004-06 hike cycle, curve flattening ebbed just as the breakeven started to widen. Chart 5Rising TIPS Breakevens Steepen The Curve Rising TIPS Breakevens Steepen The Curve Rising TIPS Breakevens Steepen The Curve Charts 6 and 7 show the relationship between the 2/10 Treasury slope and the 5-year/5-year breakeven in more detail. Chart 6 shows the correlation between monthly changes in the 2/10 Treasury slope and the 5-year/5-year breakeven using all available data back to January 1999. We see that a positive correlation between the slope and the breakeven prevailed in 64% of monthly observations, while only 36% of months displayed a negative correlation. Chart 62/10 Nominal Treasury Slope Vs. TIPS Breakeven ##br##Inflation Rate 5-Year/5-Year Forward (February 1999 - Present) The Yield Curve On A Cyclical Horizon The Yield Curve On A Cyclical Horizon Chart 72/10 Nominal Treasury Slope Vs. TIPS Breakeven Inflation Rate 5-Year/5-Year ##br##Forward During Fed Tightening Cycles (June 1999 To May 2000 & June 2004 To June 2006) The Yield Curve On A Cyclical Horizon The Yield Curve On A Cyclical Horizon In Chart 7, we focus exclusively on the past two Fed tightening cycles (1999-2000 & 2004-2006). Not only does a linear regression show an even stronger correlation than was achieved with the full sample, but we also see that a positive correlation between the slope and the breakeven existed in 73% of monthly observations, while only 27% of months displayed a negative correlation. At present, core PCE inflation is still below the Fed's 2% target and different measures of inflation expectations are all well below levels that prevailed during prior rate hike cycles (Chart 8). In other words, the Fed must proceed slowly enough with rate hikes to ensure that long-dated inflation expectations continue to trend higher, which argues for a steeper yield curve until inflation and inflation expectations are more firmly anchored around the Fed's target. For the 5-year/5-year forward TIPS breakeven inflation rate we think a range of 2.4% to 2.5% would signal that inflation expectations are well anchored around the Fed's target. 3. Volatility Implied interest rate volatility - as measured by the MOVE volatility index - is another factor that correlates with the yield curve on a cyclical horizon (Chart 9). In theory, higher rate volatility should coincide with a steeper yield curve, all else equal, and this is exactly the correlation we observe. Chart 8Fed Wants Inflation Expectations To Rise Fed Wants Inflation Expectations To Rise Fed Wants Inflation Expectations To Rise Chart 9Higher Vol Steepens The Curve Higher Vol Steepens The Curve Higher Vol Steepens The Curve Let's consider that there is a risk premium applied to taking a unit of duration risk (usually called the term premium) and that said risk premium is larger for longer-maturity bonds that carry more duration risk. All else equal, the risk premium applied to one unit of duration risk should be larger when rate volatility is higher. This should also coincide with a steeper yield curve, since there is more duration risk at the long-end of the curve. In a recent report,5 we concluded that the level of disagreement among forecasters about future GDP growth and T-bill rates were the two most important drivers of cyclical swings in implied rate volatility, the Global Economic Policy Uncertainty Index has at times also played a role (Chart 9, bottom 3 panels). Chart 10Higher Unit Labor Costs Flatten The Curve Higher Unit Labor Costs Flatten The Curve Higher Unit Labor Costs Flatten The Curve At the moment, the amount of forecaster disagreement about future GDP growth is near its lows since 1990 and T-bill forecast disagreement has, until recently, been suppressed by the zero lower bound on interest rates. All this implies that the balance of risks favors higher implied interest rate volatility in the months ahead, which will apply steepening pressure to the yield curve. 4. Unit Labor Costs Unit labor costs are the final yield curve indicator we discuss in this report. Since faster wage growth tends to coincide with Fed tightening and slowing wage growth tends to correlate with Fed easing, it makes sense for wage indicators to be inversely correlated with the slope of the yield curve. While it is broadly true that all wage indicators show a reasonable inverse correlation with the slope of the curve, unit labor costs are the best. The reason is that unit labor costs (compensation per unit produced) actually measure both wage growth (compensation per hour) and labor productivity (output per hour) (Chart 10). It turns out that the yield curve can flatten in the traditional way - a bear-flattening driven by rising wages and Fed tightening - but occasionally it can also bull-flatten if the market starts to discount a lower equilibrium (or terminal) fed funds rate. We might expect this sort of curve behavior in an environment of extremely low productivity growth, and this is exactly what has occurred during the past few years. Notice in Chart 10 that compensation per hour does not explain the curve flattening that started in 2014, but unit labor costs do because they also factor in incredibly low productivity growth. In the longer-run, we would strongly expect unit labor costs to remain in an uptrend. Wage growth is accelerating and there are structural headwinds that will prevent productivity growth from returning to the levels seen at the height of the IT revolution in the late 1990s and early 2000s. As was discussed last year in a Special Report from our Global Investment Strategy service,6 the rate of human capital accumulation is in a secular downtrend as is the share of workers in their 40s - the age cohort when people are most productive. However, there has also been a cyclical component to the productivity slowdown and it is possible that productivity growth could accelerate somewhat in the near-term as the cycle matures. The capital stock per worker correlates strongly with productivity growth (Chart 11), and while capital investment has been depressed for most of the recovery there are finally some signs that it may return (Chart 12). Chart 11Productivity Held Back By Lack Of Investment Productivity Held Back By Lack Of Investment Productivity Held Back By Lack Of Investment Chart 12Getting Optimistic About Capex Getting Optimistic About Capex Getting Optimistic About Capex In fact, it is even conceivable that more rapid wage growth itself might encourage firms to replace labor with capital, causing traditional measures of wage growth to accelerate relative to unit labor costs. Also, the prospect of tax reform and regulatory relief could give capital spending a boost - it has already led to a jump higher in small business optimism (Chart 12, bottom panel). Unit labor costs will likely continue to accelerate on a cyclical investment horizon, applying flattening pressure to the yield curve. But this flattening pressure would be mitigated to the extent that there is any cyclical rebound in productivity growth. Yield Curve Strategy Upon consideration of the four factors described above, we conclude that while the slope of the yield curve will likely be close to zero sometime in late 2018, curve flattening won't start in earnest until late this year or early next year when inflation expectations are higher (2.4% to 2.5% on long-dated TIPS breakevens) and core PCE inflation is firmly anchored around the Fed's 2% target. This conclusion is based on our observations that: TIPS breakevens and the slope of the curve tend to be positively correlated, even during rate hike cycles. Interest rate volatility is more likely to rise than fall. Unit labor costs are likely to remain in an uptrend on a cyclical horizon, but there is scope for them to level-off if we see a modest late-cycle rebound in productivity growth. To position for a steeper yield curve between now and the end of this year we continue to recommend that investors favor the 5-year Treasury note relative to a duration-matched position in a 2-year/10-year barbell. Long bullet/short barbell trades tend to outperform when the yield curve steepens, and our model suggests that the 5-year yield is currently very cheap relative to the 2/10 slope (Chart 13). We have been recommending this trade since December 20, 2016 and it has so far returned +2 bps even though the 2/10 slope has flattened 13 bps during that time. The strong positive carry means that not much curve steepening is required for the trade to realize strong positive gains. Chart 13The 5-Year Bullet Is Cheap On The Curve The 5-Year Bullet Is Cheap On The Curve The 5-Year Bullet Is Cheap On The Curve Ryan Swift, Vice President U.S. Bond Strategy rswift@bcaresearch.com 1 For further details on our Fed Monitor please see U.S. Bond Strategy Weekly Report, "Buy The Back-Up In Junk Spreads", dated March 14, 2017, available at usbs.bcaresearch.com 2 https://www.newyorkfed.org/newsevents/speeches/2015/dud150605 3 Please see U.S. Bond Strategy Weekly Report, "Caught In A Loop", dated September 29, 2015, available at usbs.bcaresearch.com 4 Please see BCA Special Report, "Beware The 2019 Trump Recession", dated March 7, 2017, available at usbs.bcaresearch.com 5 Please see U.S. Bond Strategy Weekly Report, "The Road To Higher Vol Is Paved With Uncertainty", dated February 14, 2017, available at usbs.bcaresearch.com 6 Please see Global Investment Strategy Special Report, "Weak Productivity Growth: Don't Blame The Statisticians", dated March 25, 2016, available at gis.bcaresearch.com Fixed Income Sector Performance Recommended Portfolio Specification

Highlights Despite our tactical bullish stance, the cyclical outlook remains firmly negative for the yen, with a 12-month target for USD/JPY above 120. The BoJ is currently committed to an inflation overshoot, with this solid commitment, a strong economy will be able to lift inflation expectations, depress real interest rates, and hurt the yen. The key improvements pointing to higher inflation expectations are: Already positive inflation expectation dynamics, the closing of the output gap, the removal of the fiscal drag, the tightness in the labor market, and the end of the private-sector deleveraging. The tactical environment suggests that nimble traders with short investment horizons should stay short USD/JPY for now. Longer-term investors may want to add to short bets on the yen on further weaknesses. Feature We have espoused a cyclically bearish stance on the yen since September when the BoJ began targeting the price of money instead of the quantity of money, aiming for stable JGB yields around 0%.1 More recently, we have been buyers of the yen on a tactical basis. Here, we are reviewing whether this tactical call should morph into a cyclical bullish stance on the yen or whether the primary trend for the yen still points lower. Ultimately, we expect USD/JPY to punch through 120 on a 12 month basis. The Liquidity Trap Our framework to analyze the yen rests on one key assumption: Japan remains mired in liquidity trap dynamics. As we have pointed out before, the key symptom of this disease is evident in the Land of the Rising Sun: Loan demand has become irresponsive to changes in private sector borrowing costs (Chart I-1). In this environment, we can experience strange dynamics. As we argued in details a few months ago, when both in a liquidity trap and at the lower bound of interest rates, the demand for money is infinite, and interest rates are independent of the level of output in the economy.2 In other words, a decrease in exports, government spending, or investment, hurts demand without affecting nominal interest rates (Chart I-2, middle panel). However, in the long run, decreases in aggregate demand exert downward pressure on prices, and thus, lower inflation expectations today (Chart I-2, bottom panel). The opposite is true for a positive demand shock. Chart I-1The Symptom Of Disease The Symptom Of The Disease The Symptom Of The Disease Chart I-2The Thing That Should Not Be JPY: Climbing To The Springboard Before The Dive JPY: Climbing To The Springboard Before The Dive In this topsy-turvy world, a negative shock to growth, by decreasing inflation expectations, pushes up real interest rates, and thus the exchange rate. Meanwhile, a positive shock increases inflation expectations, pulling down real rates and the exchange rate as well. This is fundamental as USD/JPY continues to trade closely in line with real rate differentials between the U.S. and Japan (Chart I-3). Chart I-3USD/JPY: No Money Illusion Here USD/JPY: No Money Illusion Here USD/JPY: No Money Illusion Here This is even truer now that the Bank of Japan is both trying to keep 10-year JGB yields near 0%, and has promised to keep a very accommodative monetary policy in place until inflation has overshoot the price stability target of an average inflation rate of 2% over the whole business cycle. In other words, the BoJ's inflation target is near symmetrical and monetary policy will only harden once previous inflation undershoots below 2% have been compensated by an extended period of inflation overshoot. Also, we expect the BoJ to stay committed to this policy. Not only does Abenomics remain popular in Japan, but we expect Kuroda to be re-appointed to lead the BoJ. Moreover, the last two members of the policy committee not appointed by Abe will see their terms end in 2017. After this year, the BoJ committee will fully represents Abe's wishes. Under this framework, the key to expect the yen to fall is therefore not valuation, nor the current account outlook - two factors pointing to a higher yen - but whether or not the economy and inflation expectations can improve durably on a cyclical basis. In the next section, we explore the key positive economic developments underpinning our negative JPY stance. Bottom Line: As the BoJ is strongly committed to maintaining an extremely dovish stance until inflation overshoots by a wide-enough margin to compensate for previous undershoots, key economic improvements in Japan should lead to higher inflation expectations, falling Japanese real interest rates, and a much weaker yen. The Five Samurais We see five reasons to remain bearish the JPY: Inflation expectation dynamics, the closing output gap, the disappearance of the fiscal drag, the labor market tightness, and the end of the Japanese private sector's deleveraging. Factor 1: Inflation Expectations Are Already Unhinged Even before the BoJ aggressively targeted 0% JGB yields, Japanese inflation expectations were on an improving path. During the 2012 summer, markets began correctly anticipating the December electoral victory of Shinzo Abe, apprehending that his BoJ was about to massively ramp up quantitative easing. Japanese 5-year/5-year forward CPI swaps soon decoupled from the rest of the world and the U.S. (Chart I-4). Chart I-4The BoJ Policy Has Already Borne Fruit The BoJ Policy Has Already Borne Fruits The BoJ Policy Has Already Borne Fruits Chart I-5The Mechanics Of Price-Level Targeting JPY: Climbing To The Springboard Before The Dive JPY: Climbing To The Springboard Before The Dive So strong has the perceived commitment of the BoJ to higher inflation been that Japanese inflation expectations never tanked the way U.S. ones did after 2014. These dynamics contributed to keep Japanese real rates depressed relative to U.S. ones. Moreover a virtuous circle was created where lower real rates supercharged the USD/JPY's rally, lifting it by more than 60% from 77 in September 2012 to 125 in June 2015, and this further supported Japanese inflation expectations. In the summer of 2015, as EM and commodity prices began imploding on the growing expectation of a Chinese economic hard landing, Japanese inflation expectations did relapse, strengthening the yen rally. But again, unlike in the U.S., Japanese CPI swaps never fell to new lows, pointing to some improving dynamics for the domestic component of Japanese inflation expectation formations. Going forward, we expect Japanese inflation expectations to move further up. The price level targeting mechanism put in place by the BoJ last fall reinforces inflationary dynamics (Chart I-5). Any anticipated tightening in monetary policy in response to economic improvements has been pushed further away in the future, in a world where inflation may be higher locally and globally. Additionally, if global and local inflation rises, because nominal interest rates are pegged at low levels, the increase in inflation expectations puts additional downward pressure on real rates, further stimulating the domestic economy, further weakening the yen, and further boosting inflation expectations. The circuits for positive feedback loops are being laid in place. Factor 2: The Output Gap Based on the OECD's estimates, the Japanese output gap has now moved into positive territory for the first time since 2007-2008, the last episode where Japan experienced anything close to inflation (Chart I-6). Prior to then, the last time the Japanese output gap was as positive as it will be in 2017 was in 1993, among the last years when Japanese core inflation was still above 1%. While this reflects the global phenomenon of low productivity growth, the low level of supply expansion in Japan has been augmented by the 2% decline in the labor force since 1998. This means that the capacity constraints in the Japanese economy are easy to reach even if average real GDP growth has only been 0.8% since 2010. The cyclical improvements in the business cycle only point toward an increasingly positive output gap and rising inflationary pressures. To begin with, business confidence and PMIs are all very robust (Chart I-7). Chart I-6No More Slack In Japan No More Slack In Japan No More Slack In Japan Chart I-7Japanese Businessmen Feel Good Japanese Businessmen Feel Good Japanese Businessmen Feel Good The strength of the U.S. ISM index suggests that Japanese exports have more upside (Chart I-8) as well. Not only does a stronger Japanese trade balance contributes to a larger positive output gap, but also, strong export growth has often been the key precursor to higher capex in Japan (Chart I-8, bottom panel). Finally, the credit dynamics remain supportive. Bank loan growth has not slowed much, despite the large tightening in Japanese monetary conditions in 2016. With conditions now easing in the country, we expect the credit impulse, which has bottomed around the zero line, to re-accelerate going forward, supporting excess demand above potential GDP growth (Chart I-9). Together, all these factors suggest that the improvement in the Japanese shipments-to-inventory ratio witnessed since March 2016 will continue to lift Japanese inflation expectations higher (Chart I-10). Chart I-8Strong Japanese Exports ##br##Will Filter To Capex Strong Japanese Exports Will Filter To Capex Strong Japanese Exports Will Filter To Capex Chart I-9The Japanese Credit ##br##Impulse Will Rebound The Japanese Credit Impulse Will Rebound The Japanese Credit Impulse Will Rebound Chart I-10Upward Momentum In ##br##Japanese Inflation Expectations Upward Momentum In Japanese Inflation Expectations Upward Momentum In Japanese Inflation Expectations Factor 3: Fiscal Policy Another key factor that has hampered the Japanese economy since 2013 has been the large fiscal belt-tightening experience by the country. In the wake of the 2011 Tohoku earthquake, the government primary deficit blew up to 7.7% of potential GDP in 2011. It will hit 3.5% for 2017, but the IMF does not forecast much more narrowing of the government budget gap (Chart I-11). This signifies that the great brake that slowed the Japanese economy and prevented a rise in inflation is being lifted. In fact, we expect the Japanese government deficit to increase again. First, Abe's upper house electoral victory last summer was built on a campaign of larger government spending. Second, with an approval rating of 56% four years into his premiership, Abe remains a highly popular prime minister for a country plagued by 15 changes of government since 1990. This is a vote of confidence by the Japanese public toward his "Abenomics" program. Finally, military spending is likely to increase. As recently as 2005, Japan's and China's defense budgets were the same; today, China outspends Japan by four times (Chart I-12). In an increasingly unstable Asia-Pacific region, where China, Russia, and North Korea are all conducting more independent foreign policy agendas, Japan will be forced to fend for itself with more military spending, underscoring the relatively hawkish agenda of the Abe administration on this front. This will require more spending by Tokyo in this arena. Chart I-11Vanishing Japanese##br## Fiscal Drag Vanishing Japanese Fiscal Drag Vanishing Japanese Fiscal Drag Chart I-12The Geopolitical Imperative To Increase ##br##Japanese Government Spending The Geopolitical Imperative To Increase Japanese Government Spending The Geopolitical Imperative To Increase Japanese Government Spending Factor 4: The tightening Labor Market The Japanese labor market has now become very tight and key supply-side adjustments are behind us. The job-openings-to-applicants ratio stands at July 1991 levels, the last time when Japan was able to generate any durable wage growth. Additionally, the level of participation of women in the labor force is very elevated. The employment-to-population ratio for prime-age females stands at 74%, well above the 71.4% level of the U.S. today, and just as high as the U.S. in 2000, when that ratio was at its highest (Chart I-13). Additionally, despite a shrinking labor force and population, the total number of employed individuals stands at 65 million, the highest level since 1999 (Chart I-14). Hiring growth is also experiencing its most vigorous upswing in 20 years. Unsurprisingly, nominal wages have been growing since 2013, the longest upswing since 2004 to 2006, and wages are now at their highest level since 2009 (Chart I-14, middle panel). Chart I-13The Japanese Labor Market Is Very Tight (I) The Japanese Labor Market Is Very Tight (I) The Japanese Labor Market Is Very Tight (I) Chart I-14The Japanese Labor Market Is Very Tight (II) The Japanese Labor Market Is Very Tight (II) The Japanese Labor Market Is Very Tight (II) With the economy remaining robust, the output gap being closed, and the fiscal drag disappearing, this tightening in the labor-market should lead to additional wage gains in Japan. As the labor market slack dissipates further, we expect Japanese employment growth to slow and wages to accelerate their upward path. It is true that the Japanese labor market duality still constitutes a structural damper on Japanese wages, but for now, the very important positive cyclical factors noted above should overpower this long-term negative. Only with additional reform of the labor market will this duality dissipate structurally. Factor 5: End Of The Private Sector Deleveraging The last factor that has turned the corner in Japan is the evolution of the private sector's deleveraging. Non-financial private debt fell from 220% of GDP in 1994 to 160% of GDP today, after having stabilized since 2009 (Chart I-15). At these levels, the Japanese non-financial private debt to GDP is in line with the worldwide average of 157%, much below China's 210%, as well as below the levels recorded in Canada, Australia, New Zealand or Sweden. This development is key for many reasons. First, since 2011, Japanese households have in fact re-levered, with their debt load rising by 6.5% since their trough. This means that Japanese households are generating demand in excess of their earnings, and are therefore a source of inflation in the country. Second, the end of deleveraging has coincided with an end to the decline in Japanese land prices that has put downward pressure on all prices since 1991 (Chart I-16). Finally, the rising debt load of the Japanese government is no longer just a compensating mechanism for the deficiency in demand created by the private sector's sector deleveraging. In fact, like for households, government dissaving is now purely adding to the aggregate demand of Japan, and at the margin, is inflationary. Unsurprisingly, since 2012, periods of accelerating growth in the Japanese broad money supply have now been associated with periods of weakness in the yen (Chart I-17). This highlights the fact that money creation is now generating some increase in inflation expectations as the private sector is not furiously building its savings anymore and as the Kuroda BoJ is not leaning against inflationary developments. Chart I-15Private Sector Deleveraging Is Over Private Sector Deleveraging Is Over Private Sector Deleveraging Is Over Chart I-16Land Prices Are Not A Source Of Deflation Anymore Land Prices Are Not A Source Of Deflation Anymore Land Prices Are Not A Source Of Deflation Anymore Chart I-17Money Matters Money Matters Money Matters Putting It All Together In our view, in an environment where Japan is beginning to generate domestic inflationary pressures of its own, where the output gap is now positive, where the government is not putting a brake on growth anymore, where the labor market is at its tightest in decades, and where private sector deleveraging is not an handicap anymore, any improvement in global growth is likely to result in further increases in Japanese inflation expectations. Our sister service, Global Investment Strategy is long Japanese CPI swaps, a trade we agree with. In the context of FX, with the BoJ firmly on an easing path, rising Japanese inflation expectations will only depress Japanese real rates, exactly as the Fed becomes more aggressive. As a result, on a 12-18 months basis, the downside for the yen is very large. What About Trump? Chart I-8Japan FDI Profile JPY: Climbing To The Springboard Before The Dive JPY: Climbing To The Springboard Before The Dive President Trump wants to see a lower dollar to achieve his goal of creating manufacturing jobs in the U.S. Much ink has been spilled on the potential emergence of a Plaza 2.0 accord. We disagree. The U.S. has very little leverage to boost the value of the yen. The Bank of Japan's policy is designed to generate domestic inflationary pressures, the yen is only a casualty of this policy. In fact, with inflation expectations having been so low for so long, no country in the world can better justify having a very loose monetary policy setting than Japan. Also, the 97% surge in the yen that followed the Plaza accord of 1985 caused Japanese interest rates to stay too low relative to the state of the economy. As a result, a massive debt bubble ensued that lifted the economy further, but then prompted the bust which Japan still pays for. Today, the Japanese are unlikely to want to repeat the same mistake. While we do think that deleveraging has ended in Japan, a country with a falling population is unlikely to begin a new private-sector debt supercycle either. Finally, China continues to be an economy that saves too much. This means that China can either allocate these savings domestically through the debt market or export them internationally through its current account surplus. We expect Chinese authorities, who are already very worried by the high debt load in China to choose the second option for the next two years. As a result, BCA foresees further declines in the RMB over the next 12 to 18 months. In this environment, the Japanese would find it very difficult to remain competitive in the Chinese market if their currency rises as the RMB weakens.3 That being said, Trump will want some concessions out of the Japanese. Already, the February 10 meeting between the U.S. president and PM Abe is giving us a glimpse of things to come. Japanese non-tariff barriers on U.S. products are likely to decrease, potentially in the agricultural and automotive field especially. Additionally, Japan still runs a large current account surplus and therefore, a large capital account deficit. We expect Japanese FDIs in the U.S. to only grow going forward. The main beneficiary is likely to be the automotive sector as it would be the key mechanism for Japanese firms to avoid paying large tariffs / punitive taxes and still access the vital U.S. market (Chart I-18). Moreover, this fits well within Trump's agenda as it creates manufacturing jobs in the U.S. Call it a win-win situation if you will. Not Time To Close Short USD/JPY Yet Despite this very negative cyclical view on the yen, we remain committed to our tactical short USD/JPY position: For one, positioning on the yen remains too extreme (Chart I-19). Second, as argued by our European Investment Strategy service, we may be on the cusp of a mini down cycle in the credit impulse, suggesting a temporary deceleration in the G10.4 The recent collapse in quarterly credit growth in the U.S. points exactly in this direction (Chart I-20). Because U.S. 10-year bond yields are so tightly linked to global economic surprises, negative surprises could put temporary downward pressure on Treasury yields (Chart I-21). A move lower in yields would be very supportive of the yen, even if only for a few months. Chart I-19Speculators Are Still Too ##br##Short JPY Tactically Speculators Are Still Too Short JPY Tactically Speculators Are Still Too Short JPY Tactically Chart I-20Falling Short-Term Credit##br## Impulse In The U.S. Falling Short-Term Credit Impulse In The U.S. Falling Short-Term Credit Impulse In The U.S. Chart I-21Falling Surprises Can##br## Temporarily Help Bond Prices Falling Surprises Can Temporarily Help Bond Prices Falling Surprises Can Temporarily Help Bond Prices Third, the dollar correction is not over. Sentiment and positioning on the dollar represent tactical hurdles that need to be overcome before the greenback can resume its ascent. Also French OAT / German bunds spreads are at distressed levels, having only been higher at the height of the euro crisis in 2012, and not far off the levels experienced during the ERM crisis of the early 1990s (Chart I-22). This suggests that the risk of a Le Pen presidency is now well known. We agree that the impact of such an event would be enormous, but the 34.5% odds currently assigned to it on Oddschecker are too great, especially now that Bayrou - a centrist politician - is not entering the race and putting his support behind Macron. Finally, the dollar has followed a textbook wave pattern since October. A continuation of this pattern suggests that the DXY has downside toward 97-98 (Chart I-23). Chart I-22OAT / Bund Spreads Price In A Lot Of Negatives OAT / Bund Spreads Price In A Lot Of Negatives OAT / Bund Spreads Price In A Lot Of Negatives Chart I-23A Textbook Wave Pattern In The Dollar A Textbook Wave Pattern In The Dollar A Textbook Wave Pattern In The Dollar The ultimate factor in favor of the continuation of the yen correction is the higher degree of complacency that has settled globally. Our Global Complacency indicator, based on the G10 stock-to-bond ratio, commodity prices, and the VIX is at an extremely elevated level warning of a potential risk-off event globally. Any rollover in this very mean-reverting indicator would prompt a further weakness in USD/JPY as well as AUD/JPY, especially if the BoJ doesn't increase stimulus in the meantime (Chart I-24). Chart I-24AUnless The BoJ Eases Further, Too Much ##br##Complacency Equals Tactically Long JPY Unless The BoJ Eases Further, Too Much Complacency Equals Tactically Long JPY Unless The BoJ Eases Further, Too Much Complacency Equals Tactically Long JPY Chart I-24BUnless The BoJ Eases Further, Too Much ##br##Complacency Equals Tactically Long JPY Unless The BoJ Eases Further, Too Much Complacency Equals Tactically Long JPY Unless The BoJ Eases Further, Too Much Complacency Equals Tactically Long JPY Bottom Line: Tactical investors should continue shorting USD/JPY for the moment. More cyclical players can begin deploying capital to short the yen as the cyclical outlook for this currency remains dire, but better opportunity to sell this currency are likely to emerge over the coming months. A dollar-cost averaging strategy seems wise at this point. Mathieu Savary, Vice President Foreign Exchange Strategy mathieu@bcaresearch.com 1 Please see the Foreign Exchange Strategy Weekly Report, "How do You Say "Whatever It Takes" In Japanese?", dated September 23, 2016, available at fes.bcaresearch.com 2 Please see the Foreign Exchange Strategy Weekly Report, "Down The Rabbit Hole", dated April 15, 2016, available at fes.bcaresearch.com 3 For a more detailed discussion on the RMB, please see the Global Investment Strategy Weekly Report, "Does China Have A Debt Problem Or A Savings Problem?", dated February 24, 2017, available at gis.bcaresearch.com 4 For a more detailed discussion of the mini-cycle, please see the European Investment Strategy Weekly Report, "Slowdown: How And When?", dated February 2, 2017, available at eis.bcaresearch.com Currencies U.S. Dollar Chart II-1USD Technicals 1 USD Technicals 1 USD Technicals 1 Chart II-2USD Technicals 2 USD Technicals 2 USD Technicals 2 The U.S. economy is giving a green light to the Fed to hike. Headline CPI is at 2.5% annually, and core CPI is at 2.3%; Retail sales beat expectations at 0.4% MoM; The core CPI measure is evidence that the U.S. economy is fundamentally strong and dynamic. Real GDP now stands 11% above its pre-recession peak, and it is approaching the Congressional Budget Office's estimate of potential output. The unemployment and output gap are also close to their long-term levels. With the economy closing in on its potential, it is only natural that FOMC participants "expressed the view that it might be appropriate to raise the federal funds rate again fairly soon" in the Minutes. Although a risk of disappointment from Trump's fiscal proposal is possible, the economy's momentum will continue. Report Links: Updating Our Long-Term FX Value Models - February 17, 2017 Risks To The Cyclical Dollar View - February 3, 2017 Dollar Corrections, EM Outlook, Global Liquidity, And Protectionism - January 27, 2017 The Euro Chart II-3EUR Technicals 1 EUR Technicals 1 EUR Technicals 1 Chart II-4EUR Technicals 2 EUR Technicals 2 EUR Technicals 2 The euro area remains robust, with this week's data showing a strong outperformance: German, French and overall euro area PMI increased and beat expectations across all measures, with the exception of France which only outperformed on the Composite measure; Euro area producer prices strengthened to a 2.4% annual pace; After seeing some downside from worries about a Le Pen victory, markets have calmed François Bayrou, a centrist, announced an alliance with presidential candidate Emmanual Macron, adding a resistance to the euro's downside. Substantial volatility can still be expected, however, as a Le Pen victory is not completely out of the realm of possibility, which means that the euro can see some weakness in the near term. Report Links: Updating Our Long-Term FX Value Models - February 17, 2017 The French Revolution - February 3, 2017 GBP: Dismal Expectations - January 13, 2017 The Yen Chart II-5JPY Technicals 1 JPY Technicals 1 JPY Technicals 1 Chart II-6JPY Technicals 2 JPY Technicals 2 JPY Technicals 2 Positive signs continue to emerge in Japanese data: Industrial production yearly growth came in at 3.2% Nikkei Manufacturing PMI came in at 53.5, outperforming expectations Japan's Leading Economic Index came at 104.8, the highest level since 2015 These economic developments are good news for the BoJ, as it shows them that their price level targeting and yield curve control measures seem to be working. However the objective of these measures is not to achieve these marginal improvements in the economy. The objective is to catapult Japan out of the liquidity trap it is in, which means that these measures will likely stay in place for a while. Therefore, on a cyclical basis we remain short the yen, as we expect USD/JPY to reach 120 on a 12 to 18 month horizon. Report Links: Updating Our Long-Term FX Value Models - February 17, 2017 Dollar Corrections, EM Outlook, Global Liquidity, And Protectionism - January 27, 2017 Update On A Tumultuous Year - January 6, 2017 British Pound Chart II-7GBP Technicals 1 GBP Technicals 1 GBP Technicals 1 Chart II-8GBP Technicals 2 GBP Technicals 2 GBP Technicals 2 Recent data has painted a mixed picture for the U.K. Industrial and manufacturing production yearly growth came in at 4.3% and 4% respectively. Both measures blew past expectations. Also, in spite of the dramatic fall in the pound, Inflation seems to be relatively contained, as both core and headline numbers came in below expectation at 1.8% and 1.6% respectively. However not everything is good news. Yearly growth for retail sales and retail sales ex fuel underperformed expectations coming at 1.5% and 2.6%, respectively. Additionally, wage growth has been limited, as average weekly earnings yearly growth came below expectations at 2.6%. We continue to be bullish on the pound, particularly against the euro as any additional political risks caused by Brexit are now well known by participants, making the pound very cheap, especially if one takes into account real rate differentials. Report Links: Updating Our Long-Term FX Value Models - February 17, 2017 Outlook: 2017's Greatest Hits - December 16, 2016 The Pound Falls To The Conquering Dollar - October 14, 2016 Australian Dollar Chart II-9AUD Technicals 1 AUD Technicals 1 AUD Technicals 1 Chart II-10AUD Technicals 2 AUD Technicals 2 AUD Technicals 2 The AUD has been the top performing currency against the USD out of the G10, having appreciated 7.11% since the beginning of the year. This rally is increasingly tenuous. Full-time employment has struggled to pick up, while part-time employment increased by 4%. This will hamper wage growth and consumption going forward. This is important as consumption is already 58% of the economy. Meanwhile, net exports have made a negative contribution to GDP growth for almost two years. In fact, Australian exports to China subtracted 1% of GDP growth last year, due to a decline in commodity prices. Going forward, a limited upside in commodity prices and an end to the Chinese easing cycle can exacerbate this decline. On a technical basis, AUD/USD has sustained momentum since the beginning of the year, with the RSI displaying overbought levels since mid-January. The cross is also approaching a key resistance level, pointing to growing risks ahead. Report Links: Updating Our Long-Term FX Value Models - February 17, 2017 Risks To The Cyclical Dollar View - February 3, 2017 Outlook: 2017's Greatest Hits - December 16, 2016 New Zealand Dollar Chart II-11NZD Technicals 1 NZD Technicals 1 NZD Technicals 1 Chart II-12NZD Technicals 2 NZD Technicals 2 NZD Technicals 2 Recent data for New Zealand has not been particularly positive and have weighed on the kiwi: Retail sales underperformed, growing by 0.8% QoQ against expectations of 1.1%. Business NZ PMI fell to 51.6 from last month's 54.5. Nevertheless, a closer look at the data paints a much brighter picture: the decline in NZ PMI seems to have been primarily due to bad weather conditions, which means that the strong fundamentals of the kiwi economy should show up in the data once seasonal factors start to dissipate. Therefore, we are bullish on the NZD versus the AUD, as the structural backdrop for these countries could not be further apart, yet the market is now pricing less than a 10 basis points difference from here until the end of the year. Report Links: Updating Our Long-Term FX Value Models - February 17, 2017 Risks To The Cyclical Dollar View - February 3, 2017 Outlook: 2017's Greatest Hits -December 16, 2016 Canadian Dollar Chart II-13CAD Technicals 1 CAD Technicals 1 CAD Technicals 1 Chart II-14CAD Technicals 2 CAD Technicals 2 CAD Technicals 2 Canadian employment numbers came out seemingly strong, with a net change in employment of 48,300 and a decrease in the unemployment rate to 6.8%. However, these numbers mask numerous underlying inconsistencies. The decrease in unemployment was the result of a robust part-time employment growth of 5.6%, not the 0.3% growth in full-time employment. Wage growth remains subdued, with average hourly earnings of permanent workers currently increasing at a 1% annual pace, compared to 3.3% a year ago. Furthermore, hours worked have declined by 0.8%, exacerbating the weakness of full-time employment's contribution to activity. Retail sales underperformed expectations, contracting at a 0.5% monthly pace; the measure excluding Autos also contracted at a 0.3% pace. Increasing household debt and festering labor market complications are likely to weigh on consumer confidence. An uncertain outlook on trade developments is an additional handicap to future CAD strength. Report Links: Updating Our Long-Term FX Value Models - February 17, 2017 Outlook: 2017's Greatest Hits - December 16, 2016 When You Come To A Fork In The Road, Take It - November 4, 2016 Swiss Franc Chart II-15CHF Technicals 1 CHF Technicals 1 CHF Technicals 1 Chart II-16CHF Technicals 2 CHF Technicals 2 CHF Technicals 2 During the last couple of weeks, fear of a Eurosceptick government in Europe's second biggest economy, has lowered EUR/CHF below the implied floor that the SNB has had for the last couple of years. Indeed, last week, as La Pen surged on French presidential polls, this crossed reached 1.063, its lowest level since August 2015. This is bad news for Switzerland, as economic data continues to indicate that the country has not been able to shake off the shackles of deflation: Headline inflation outperformed expectations as it finally exited deflationary territory, coming in at 0%. Industrial production contracted by 3.3% on a year on year basis Given this deflationary backdrop, the SNB will continue to try to limit the downside for this cross. However, on the months leading to the French elections, the floor will continue to get tested. Report Links: Updating Our Long-Term FX Value Models - February 17, 2017 Outlook: 2017's Greatest Hits - December 16, 2016 Long-Term FX Valuation Models: Updates And New Coverages - September 30, 2016 Norwegian Krone Chart II-17NOK Technicals 1 NOK Technicals 1 NOK Technicals 1 Chart II-18NOK Technicals 2 NOK Technicals 2 NOK Technicals 2 Inflation seems to be abating in Norway as core and headline inflation numbers fell sharply from last month reading, coming in at 2.1% and 2.8% respectively. This is the result of various factors: First, the inflation caused by the collapse of the krone is starting to fade away. From 2014 to 2016, the krone collapsed along with oil prices. This selloff in the krone passed through inflation to the Norwegian economy via rising imported goods, with a lag. Today, roughly one year after the NOK bottomed, the effects of the currency on inflation is starting to dissipate. Furthermore, labor market dynamics in Norway are anything but inflationary as wage growth is contracting by 4% and although unemployment is low, the Norges Bank has pointed out that is in largely caused by a fall in the participation rate. Thus, given that high inflation is receding, the Norges Bank will keep its easing bias for the time being. Report Links: Updating Our Long-Term FX Value Models - February 17, 2017 Outlook: 2017's Greatest Hits -December 16, 2016 The Pound Falls To The Conquering Dollar - October 14, 2016 Swedish Krona Chart II-19SEK Technicals 1 SEK Technicals 1 SEK Technicals 1 Chart II-20SEK Technicals 2 SEK Technicals 2 SEK Technicals 2 The February 2017 Monetary Policy Statement illustrated a clear dovish stance. Governors and economists at the Riksbank are paranoid about risks emanating from a strong currency and political developments. Tensions from a recently strong SEK have created worries about a potential slowdown in inflation. The Bank has therefore reiterated the possibility of an intervention if the Krona's appreciation is too rapid, making it a very real possibility. A questionable political outlook from the U.S. and the euro area has further hampered the Riksbank's optimism. The euro area is a particular risk since it represents a large source of Sweden's growth, and any damage to the monetary union will have a catastrophic effect on Sweden. Because of these reasons, the Riksbank explicitly stated that it is "still prepared to make monetary policy more expansionary if the upward trend in inflation were to be threatened and confidence in the inflation target weakened." Report Links: Updating Our Long-Term FX Value Models - February 17, 2017 Outlook: 2017's Greatest Hits - December 16, 2016 One Trade To Rule Them All - November 18, 2016 Trades & Forecasts Forecast Summary Core Portfolio Tactical Trades Closed Trades