Asset Allocation

Highlights Portfolio Strategy The easy money has already been made in the liquidity-to-growth theme-levered long S&P energy/short global gold miners pair trade. Lock in profits and move to the sidelines, for now. Similarly, book gains in the long S&P materials/short S&P utilities market-neutral trade. A stealthy macro shift, at the margin, suggests that a more challenging phase lies ahead for this relative share price ratio. Recent Changes Book 18.3% profits in the long S&P energy/short global gold miners pair trade today. Take profits in excess of 8.6% in the long materials/short utilities pair trade today. Table 1 Feature Equities continued to defy gravity last week, vaulting to fresh all-time highs. Seasonality (or the pending Santa rally) appears to have trumped any "buy on rumor sell the tax news" jitters, at a time when macro data continue to surprise to the upside. Heading into 2018, easier fiscal policy will likely offset some of the uneasiness of the Fed's ongoing tightening cycle as we postulated in early October.1 Synchronized global economic and capex growth remain the key macro themes that dominate markets. The latest GDP revisions in the G3 confirm our global capex upcycle bias: U.S., euro area and, especially, Japanese gross fixed capital formation are on fire (Chart 1). Importantly, once the tax bill related dust settles, profits will come back to the forefront as a key stock market driver. In that regard, the news on the EPS front is ebullient and, along with the forward multiple, all that matters. Table 2 shows annual SPX returns going back to 1979, and breaks down the composition of the capital (not total) return into two components: forward earnings growth and the forward P/E multiple (January 1979 is the first IBES data point for forward EPS SPX estimates). Chart 1Synchronized Global Capex Table 2Disentangling SPX Returns Currently, sell side analysts expect 11% EPS growth for 2018, and our sense is that 8-12% EPS growth is achievable next year, a message that our SPX EPS macro model corroborates (Chart 2). Keep in mind that there is no tax cut penciled into our EPS model's numbers. Chart 2SPX EPS Macro Model Flashing Green What is interesting from the multiple/EPS analysis is that over the last four decades when forward profit growth was in this high single-digit / low double-digit range (ten iterations), the multiple expanded modestly (on average, adding 2.6 percentage points to the market's return) and EPS did the heavy lifting (explaining, on average, roughly 80% of the S&P 500's 12.9% average annual return, Table 3). If we consider periods when EPS growth was positive but below 8% (eleven iterations), SPX returns are close to 10%, on average, with EPS and the multiple contributing almost equally to the market's return. One caveat is that two recessionary years and the dot com bust are part of this segment skewing the results to the downside (Table 3).2 Table 3Disentangling SPX Returns Continued Nevertheless, if history at least rhymes, were EPS growth to stay positive next year and hit the 8-12% mark, then a profit driven low double-digit broad equity market return is likely. If profits disappoint and grow between 0-8%, barring recession, empirical evidence suggests that equity returns will still prove healthy. Adding it up, the path of least resistance is higher for equities on a cyclical 9-12 month horizon. Granted, since Brexit the SPX has rallied in a near straight line up and a healthy and temporary pause for breath is likely in Q1/2018. As a result, this week we are booking impressive gains in two tactical market-neutral trades we initiated in late-August and mildly de-risking our portfolio. Lock In Profits In The Long Energy/Short Gold Producers Trade In late-August we initiated a liquidity-to-growth handoff levered market-neutral trade: long S&P energy/short global gold miners. Over the past four months this trade is up 18.3%. It also sports a positive annual dividend carry of 200bps. With the equity market overshoot phase likely going on hiatus sometime in early 2018 is it still prudent to hold this high-octane intra-commodity and market-neutral trade? The short answer is no. Nothing in terms of macro data has changed to trip up this pair trade. If anything, the handoff of global liquidity to economic growth has gained steam in the past few months. Global GDP, IP, manufacturing PMIs, global trade (Chart 3) and gross capital formation are all growing simultaneously across all of the G7 and most of the EMs. Even China's economy seems to have stabilized. The Fed announced its plans to wind down its balance sheet as expected in September and the BoE and BoC have both tightened monetary policy. Even the ECB announced a halving of the size of its monthly purchases in late-October (but extended it for nine months). All these central bank (CB) moves suggest that, at the margin, the global liquidity injection is reversing, with CBs actually mopping up liquidity. This is de facto negative for the shiny metal and gold mining equities as interest rates are headed higher (Chart 4). Chart 3Brisk Global Growth... Chart 4...Higher Rates... Moreover, geopolitical uncertainty is steadily receding, especially now that the Senate also passed a tax bill, and a final bill will likely soon be signed into law.3 Historically in times of duress, safe haven assets are bid up and vice versa, and the current low policy uncertainty backdrop is conducive to additional gains in the relative share price ratio (policy uncertainty shown inverted, Chart 5). Meanwhile, on the relative operating front, energy stocks have the upper hand versus gold miners. The oil and gas rig count has resumed its advance and remains 150% clear of the lows hit during the depths of the global manufacturing recession of late-2015/early-2016. Anecdotes of global oil majors comfortably registering positive EPS, in the new era of $50/bbl oil, and reinstating stock buybacks and eliminating scrip dividends (RDS, BP & ENI) suggest that the worst is behind the industry. In contrast, safe haven asset demand is in retreat and will continue to weigh on global gold ETF flows. Anecdotally, the BITCOIN/ICO/cryptocurrency mania may also steal some of bullion's thunder, as this mania is capturing investor's imagination. Either a flare up in global geopolitical risk or a global growth scare could cause investors to start shifting capital into gold ETFs. Our relative EPS models do an excellent job in capturing this energy positive/gold negative backdrop and continue to suggest that energy profits will outpace gold mining EPS (Chart 6). Chart 5...And Diminishing Uncertainty##br## Still Bode Well For The Trade Chart 6But We Do Not Want To##br## Overstay Our Welcome If these different macro and operational forces all emit an unambiguously bullish signal for S&P energy shares compared with global gold miners, why book profits? Our sense is that there are high odds of a pullback in Q1/2018 and from a portfolio management and risk perspective it is prudent to lock in handsome profits in excess of 18.3% in a four month period. There are high odds that most of these key drivers are reflected in relative share prices versus late-August. Relative valuations are pricier today and technicals are also flashing yellow (bottom panel, Chart 4). We deem that the easy money has already been made in this market-neutral trade, despite the still favorable relative macro backdrop. This was a tactical three-to-six month pair trade that has mostly played out and we would not like to overstay our welcome. Were the broad market pullback to occur in the upcoming quarter, and the ratio to trade significantly lower, we would not hesitate to reinstate this pair trade. Our cyclical strategy is to "buy the broad market dip" and remain opportunistic on a tactical basis. Bottom Line: Lock in 18.3% profits in the long S&P energy/short global gold miners pair trade and move to the sidelines for now. Take Profits In Materials Vs. Utilities Similar to booking gains in the liquidity-to-growth levered market-neutral long S&P energy/short global gold miners pair trade, we also recommend taking profits in the reflation levered long S&P materials/short S&P utilities pair trade. Since its late-August inception this market-neutral trade has generated returns in excess of 8.6% and added alpha to our portfolio. While overall macro conditions continue to underpin the relative share price ratio, some cracks are appearing on the surface. Global reflation has matured and synchronized global growth is as good as it gets. The ISM manufacturing and services surveys have ticked down in sympathy recently, warning that the easy gains are behind this market neutral trade (Chart 7). Worrisomely, our relative sector Cyclical Macro Indicators are sniffing out this marginal shift in the macro backdrop and suggest that a more challenging phase lies ahead for the relative share price ratio (Chart 8). BCA's view remains that a sizable selloff in the bond markets is the most likely scenario in 2018. This is one of our key themes for next year, and given that this trade typically moves in lockstep with interest rates, the path of least resistance is higher. Nevertheless, the fact that this ratio has not kept up with the slingshot recovery in the stock-to-bond (S/B) ratio is slightly disconcerting. The top panel of Chart 9 shows that the gap between the S/B and the materials/utilities ratios has widened further since late-August. Chart 7As Good As It Gets? Chart 8Fatigue Signs Chart 9More Balanced Backdrop=Move To The Sidelines On the operating front, our relative EPS models are also showing signs of fatigue. Materials profits cannot expand indefinitely at the breakneck pace observed since the 2016 trough, at a time when utilities EPS have stabilized. Currently, the relative earnings models suggest that materials are on an even keel with utilities (Chart 9). Tack on rising odds of a healthy broad market pullback in Q1/2018, and from a risk management perspective we would rather de-risk the portfolio a notch by locking in near double-digit gains since inception in this volatile pair trade. Bottom Line: Book gains of 8.6% in the long S&P materials/short S&P utilities pair trade. Anastasios Avgeriou, Vice President U.S. Equity Strategy anastasios@bcaresearch.com 1 Please see BCA U.S. Equity Strategy Weekly Report, "Can Easy Fiscal Offset Tighter Monetary Policy?" dated October 9, 2017, available at uses.bcaresearch.com. 2 For reference and completion purposes Table 3 also tabulates the results during EPS contractions (nine iterations) and in profit boom times, i.e. forward EPS growth north of 12% (nine iterations). 3 Please see BCA U.S. Equity Strategy & Geopolitical Strategy Special Report, "Tax Cuts Are Here - Equity Sector Implications," dated December 11, 2017, available at uses.bcaresearch.com. Current Recommendations Current Trades Size And Style Views Favor small over large caps and stay neutral growth over value.

Recommended Allocation Highlights We are late cycle. Strong growth could turn in 2018 from a positive for risk assets into a negative. More risk-averse investors may thus want to turn cautious. But the last year of a bull run can be profitable, and we don't expect a recession until late 2019. For now, therefore, our recommendations remain pro-risk and pro-cyclical. We may turn more defensive in 2H 2018 if the Fed tightens above equilibrium. We expect inflation to pick up in 2018, which will lead the Fed to hike maybe four times. This will push long rates to 3%, and strengthen the U.S. dollar. Equities should outperform bonds in this environment. We prefer euro zone and Japanese equities over U.S., and remain underweight EM. Late-cycle sectors such as Financials and Industrials, should do well. We also favor corporate bonds and private equity. Feature Overview Fin de cycle Global economic growth in 2017 was robust for the first time since the Global Financial Crisis (Chart 1). Forecasts for 2018 put growth slightly lower, but are likely to be revised up. However, as the year rolls on, the strong economic momentum may turn from being a positive for risk assets into a negative. U.S. output is now above potential, according to IMF estimates. As Chart 2 shows, historically recessions - and consequently equity bear markets - have usually come within a year or two of the output gap turning positive. With the economy operating above capacity, inflation pressures force the Fed to tighten monetary policy, which eventually causes a slowdown. Chart 1Growth Finally On A Firm Footing Global Growth Has Accelerated Chart 2Recessions Follow Output Gap Closing That is exactly how BCA sees the next couple of years panning out, leading to a recession perhaps in the second half of 2019. U.S. inflation was soft in 2017, but underlying inflation pressures are picking up, with core CPI inflation having bottomed, and small companies saying they are raising prices (Chart 3). Add to that wage pressures (with unemployment heading below 4% in 2018), tax cuts (which might boost growth by 0.2-0.3% points in their first year) and a higher oil price (we expect Brent to average $67 a barrel during the year), and core PCE inflation is likely to rise to 2%, in line with the Fed's expectations. This means the market is too sanguine about the risk of monetary tightening in the U.S. It has priced in less than two rates hikes in 2018, compared to the Fed's three dots, and almost nothing after that (Chart 4). If inflation picks up as we expect, four rate hikes in 2018 could be on the cards. Chart 3Inflation Pressures Picking Up Chart 4Market Still Underpricing Fed Hikes The consequences of this are that bond yields are likely to rise. Despite a significant market repricing since September of Fed behavior, long-term rates have not risen much, leading to a flattening yield curve (Chart 5). The market has essentially priced in that inflation will not rebound and that, consequently, the Fed will be making a policy mistake by hiking further. If, therefore, we are correct that inflation does reach 2%, the yield curve would be likely to steepen over the next six months, with the 10-year U.S. Treasury yield reaching 3% by mid-year. Other developed economies, however, have less urgency to tighten monetary policy and we, therefore, see the U.S. dollar appreciating. The only other major economy with a positive output gap currently is Germany (Chart 6). However, the ECB will continue to set policy for the weaker members of the euro area, and output gaps in France (-1.8% of GDP), Italy (-1.6%) and Spain (-0.7%) remain significantly negative. In the absence of inflation pressures, the ECB won't raise rates until late 2019. Japan, too, continues to struggle to bring inflation up the BOJ's 2% target and the Yield Curve Control policy will therefore stay in place, meaning that a rise in global rates will weaken the yen. Chart 5Is Fed Making A Policy Mistake? Chart 6Still A Lot Of Negative Output Gaps This sort of late-cycle environment is a tricky one for investors. The catalysts for strong performance in equities that we foresaw a few months ago - U.S. tax cuts and upside surprises in earnings - have now largely played out. Global earnings will probably rise next year by around 10-12%, in line with analysts' forecasts. With multiples likely to slip a little as the Fed tightens, high single-digit performance is the best that investors should expect from equities. The macro environment which we expect, would be more negative for bonds than positive for equities. That argues for the stock-to-bond ratio to continue to rise until closer to the next recession (Chart 7). And, for now, none of the recession indicators we have been consistently monitoring over the past months is flashing a warning signal (Chart 8). Chart 7Stock-To-Bond Ratio Likely To Rise Further Chart 8Recession Warning Signals Still Not Flashing More risk-averse investors might chose to reduce their exposure to risk assets now, given how close we are to the end of the cycle. But this would be at the risk of leaving some money on the table, since the last year of a bull run can often be the most profitable (remember 1999?). We, therefore, maintain our recommendation for pro-cyclical and pro-risk tilts: overweight equities versus bonds, overweight credit, overweight higher-beta equity markets and sectors, and a preference towards riskier alternative assets. We may move towards a more defensive stance in mid to late 2018, when we see clearer signs that the Fed has tightened above equilibrium or that the risk of recession is rising. Garry Evans, Senior Vice President Global Asset Allocation garry@bcaresearch.com What Our Clients Are Asking What Will Be The Impact Of The U.S. Tax Cuts? It is not a done deal, but it still seems likely (notwithstanding the Democratic victory in Alabama) that the U.S. House and Senate will agree a joint tax bill to pass before the end of the year. Since the two current bills have only minor differences, it is possible to make some estimates of the macro and sector impacts of the tax reform. The Joint Committee on Taxation estimates that the cuts will reduce government revenue by $1.4 trillion over 10 years - or $1 trillion (5% of GDP) once positive effects on growth are accounted for. The Treasury argues that tax reform (plus deregulation and infrastructure development) will push GDP growth to 2.9% and therefore government revenues will increase by $300 billion. BCA's estimate is that GDP growth will be boosted by 0.2-0.3% in 2018 and 2019.1 For businesses, the key tax changes are: 1) a reduction in the headline corporate rate from 35% to 21%; 2) immediate expensing of capital investment; 3) a limit to deduction of interest expenses to 30% of taxable income; 4) a move to a territorial tax system from a worldwide one, with a 10% tax on repatriation of past profits held overseas; 5) curbs for some deductions, such as R&D, domestic production and tax-loss carry-forwards. Corporate tax cuts will give a one-off boost to earnings, since the effective tax rate is currently over 25% (Chart 9, panel 1), with telecoms, utilities and industrials likely to be the biggest beneficiaries. This is not fully priced into stocks, since companies with high tax rates have seen their stock prices rise only moderately (Chart 9, panel 2). BCA's sector strategists expect that capex will especially be boosted: they estimate that the one-year depreciation increases net present value by 14% (Table 1).2 This should be positive for the Industrials sector (supplying the capital goods) and for Financials (which will see increased demand for loans). We are overweight both. Chart 9Tax Cuts Should Boost Earnings Table 1 Is Bitcoin A Bubble, And What Happens When It Bursts? The recent surge in prices (Chart 10) of virtual currencies has pushed Bitcoin and aggregate cryptocurrency market cap to $275 billion and $500 billion respectively. The recent violent run-up certainly bears a close resemblance to classic bubbles, but the impact of a sharp correction should be minimal on the real economy and traditional capital markets. As mentioned above, the market cap of cryptocurrencies has reached $500 billion. Globally, there is about $6 trillion in currency3 outstanding, so the value of virtual currencies is now 8% that of traditional fiat currency. Additionally, an estimated 1000 people own about 40% of the world's total bitcoin, for an average of about $105 million per person. At the moment, the macro impact has been constrained by the fact that most people are buying bitcoins as a store of value (Chart 11) or vehicle for speculation, rather than as a medium of exchange. However, when the public begins to regard them as legitimate substitutes for traditional fiat currencies, their impact will be felt on the real economy. Chart 10A Classic Bubble Chart 11Bitcoin Trading Volume By Top Three Currencies That would raise the issue of regulation. The U.S. government generates close to $70 billion per year as "seigniorage revenue." Governments across the world have no intention of losing this revenue, and would most likely introduce their own competitors to bitcoin. Until then, the biggest potential impact of these private currencies might be to spur inflation in the fiat currencies in which their prices are measured. That would be bad for government bonds, but potentially good for stocks. A further risk - and a similarity with the real estate bubble of 2007 - is the use of leverage. The news of a Tokyo-based exchange (BitFyler) offering up to 15x leverage for the purchase of bitcoins has spooked investors. However, the U.S. housing market is valued at $29.6 trillion, almost 60 times that of cryptocurrencies. Finally, the 19th century free banking era in the U.S., which at one point saw 8000 different currencies in circulation, experienced multiple banking crises. A world with myriad private currencies all competing with one another would be similarly unstable. Why Did The U.S. Dollar Weaken In 2017, And Where Will It Go In 2018? Chart 12Positioning And Relative Rates Supportive For USD We were wrong to be bullish on U.S. dollar at the start of 2017. We think the dollar weakness during most of the year can be attributed to the fact that investors were massively long the dollar at the end of 2016 (Chart 12, panel 2), which made the market particularly vulnerable to surprises. Several surprises did come: inflation softened in the U.S. but strengthened in the euro area. There were also positive geopolitical surprises in Europe - for example the victory of Emmanuel Macron in the French presidential election - while the failure to repeal Obamacare in the U.S. raised investors' concerns on the administration's ability to undertake fiscal stimulus. As a result, the U.S. dollar depreciated against euro despite widening interest rate differentials (Chart 12 panel 4) in 2017. Chart 13late Cycle Outperformance Since investors are now aggressively short the dollar, the hurdle for the greenback to deliver positive surprises is much lower than a year ago. Since the Senate passed the Republican tax bill in early December, we have already seen some recovery in the dollar (Chart 12, panel 1). As the labor market continues to firm, with GDP running above potential, U.S. inflation should finally start to pick up in 2018, which will allow the Fed to hike rates, possibly as many as four times during the year. This will contrast with the macro situation overseas: Japan and Europe are likely to continue loose monetary policy to maintain the momentum in their economies. All this should be supportive of the dollar. Are Convertible Bonds Attractive Over The Next 12 Months? With valuations for traditional assets expensive and investors' thirst for yield continuing, the market is in need of alternative sources of return. Convertible bonds offer a hybrid credit/equity exposure, giving investors the option to participate in rising equity markets but with less risk. An allocation to convertibles could prove attractive for the following reasons: Convertible bonds typically outperform high-yield debt in the late stages of bull markets, because of their relatively lower exposure to credit spreads. Junk spreads have a history of starting to widen before equity bear markets begin. Fifty percent of the convertibles index comprises issuance from small-cap and mid-cap firms. Although equity valuations are expensive, prices should continue to rise as long as inflation stays low. Additionally, our U.S. Investment Strategy service thinks that small-cap equities will outperform large caps in the coming months, partly because the likely cuts in U.S. corporate taxes will disproportionately benefit smaller companies. Convertible bonds do appear somewhat cheap relative to equities (Chart 13, panel 3) but, on balance, there is not a strong valuation case for the asset class. Equities appear fairly valued relative to junk bonds, and convertibles are trading at an elevated investment premium. However, valuation is not likely to be a significant headwind to the typical late-cycle outperformance of convertibles versus high yield. biggest near-term risk for convertibles relative to high yield stems from the technology sector, which makes up 35% of the convertibles index. Technology convertible bonds have strongly outperformed their high-yield counterparts in recent months (Chart 13, panel 4), and are possibly due for a period of underperformance. We recommend investors stay cautious on technology convertibles. Other Than U.S. Tips, What Other Inflation-Linked Bonds Do You Like? Our research shows that inflation-linked bonds (ILBs) are a good inflation hedge in a rising inflationary environment.4 With our house view of rising inflation in 2018, we have been overweight U.S. Tips over nominal Treasury bonds as the U.S. is the most liquid market for inflation-linked bonds, with a market cap of over US$ 1.2 trillion. Outside the U.S., we favor ILBs in Japan and Australia, while we suggest investors to avoid ILBs in the U.K. and Germany (even though the U.K. linkers' market is the second largest after the U.S.), for the following two key reasons: First, even though inflation is below target in Japan, Australia and the euro area, while above target in the U.K., in all of these markets, inflation has bottomed, as shown in Chart 14. Second, our breakeven fair-value models, which are based on trade-weighted currencies, the Brent oil price in local currencies, and stock-to-bond total-return ratios, indicate that ILBs are undervalued in Japan and Australia, while overvalued in the U.K. and Germany, as shown in Chart 15. Chart 14Inflation Dynamics Chart 15Where to Buy Inflation? The shorter duration (in real terms) of ILBs are an added bonus which fits well with our overall underweight duration positioning in the government bond universe. Global Economy Overview: Growth in developed economies remains strong and there is little in the data to suggest it will slow. This is likely to push up inflation and interest rates, especially in the U.S., over the next six to 12 months. Prospects for emerging markets, however, are less encouraging given that China is likely to slow moderately as it pushes ahead with reforms. U.S.: U.S. growth momentum remains very strong. GDP growth in the past two quarters has come in over 3%, and NowCasts for Q4 point to 2.9-3.9%. The Citigroup Economic Surprise Index (Chart 16, panel 1) has surged since June, and the Manufacturing ISM is at 53.9 and the Non-Manufacturing at 57.4 (panel 2). The worst that can be said is that momentum will be unable to continue at this rate but, with business confidence high, wage growth likely to pick up in 2018, and some positive impacts from tax cuts, no significant slowdown is in sight. Euro Area: Given its stronger cyclicality and ties to the global trade cycle, euro zone growth has surprised on the upside even more strongly than in the U.S. The Manufacturing PMI reached 60.6 in December (its highest level since 2000), and GDP growth in Q3 accelerated to 2.6% QoQ annualized. The euro's strength in 2017 seems to have done little to dent growth, and even weaker members of the euro zone such as Italy have seen improving GDP growth (1.7% in Q3). With the ECB reining back monetary easing only slightly, and banking problems shelved for now, growth should remain resilient in early 2018. Japan: Retail sales saw some weakness in October (-0.2% YoY), probably because of bad weather, but elsewhere data looks robust. Q3 GDP came in at 1.3% QoQ annualized and export growth remains strong at 14% YoY. There are even some signs of life in the domestic economy, with wages finally picking up a little (+0.9% YoY), driven by labor shortages among part-time workers, and consumer confidence at a four-year high. Inflation has been slow to rise, but at least core core inflation (the Bank of Japan's favorite measure) is now in positive territory at +0.2%. Emerging Markets: Chinese credit and monetary series, historically good lead indicators for the real economy, continue to decline (M2 growth in October of 8.8% was the lowest since data started in 1996). But, for now, economic growth has held up, with the Manufacturing and Non-Manufacturing PMIs both stably above 50 (Chart 17, panel 3). Key will be how much the government's moves to deleverage the financial system and implement structural reform in 2018 will slow growth. Elsewhere in emerging markets, economic growth remains sluggish, with GDP growth in Brazil barely rebounding to 1.4% YoY, Russia to 1.8%, and India slowing to 6.3% (down from over 9% in early 2016). Chart 16Growth Momentum Very Strong Chart 17Will China And EM Slow in 2018? Interest rates: We expect U.S. inflation to pick up in 2018, as the lagged effects of 2017's stronger growth and the weak dollar start to come through, amid higher oil prices and rising wages. We, along with the Fed, expect core PCE inflation to rise to 2% during the year. This means the Fed is likely to raise rates four times, compared to market expectations of twice. Consequently, we see the 10-year Treasury yield over 3% by mid-year. In the euro zone, the still-large output gap means inflation is less likely to surprise on the upside, allowing the ECB to keep negative rates until well into 2019. The Bank of Japan is unlikely to alter its Yield Curve Control, given the signal this would send to the market when inflation expectations are still well below its 2% target (Chart 17, panel 4). Chart 18Equities: Priced for Perfection Global Equities Still Cautiously Optimistic: Our pro-cyclical equity positioning in 2017 worked very well in terms of country allocation (overweight euro zone and Japan in the DM universe) and global sector allocation (favoring cyclicals vs defensives). The two calls that did not pan out were underweight EM equities vs. DM equities, which was partially offset by our positive stance on China within the EM universe, and the overweight of Energy, which was the worst performing sector of the year. The stellar equity performance in 2017 was largely driven by strong earnings growth. Margins improved in both DM and EM; earnings grew in all sectors, and analysts remained upbeat (Chart 18). Another important contributor to 2017 performance was the extraordinary performance of the Tech sector, especially in China: globally, tech returned 41.9%, outperforming the MSCI all country index by 18.9%. GAA's philosophy is to take risk where it is mostly likely be rewarded. In July, we took profits in our Tech overweight and used the funds to upgrade Financials to overweight from neutral. Then in October we started to reduce tracking risk by scaling down our active country bets, closing our overweight in the U.S. to reduce the underweight in EM. BCA's house view is for synchronized global growth to continue in 2018, but a possible recession in late 2019. We are a little concerned that equity markets are priced for perfection, given that our earnings model indicates a deceleration in the coming months mostly due to a base effect. As such, our combination of "close to shore" country allocation and "pro-cyclical" sector allocation is appropriate for the next 9-12 months. Country Allocation: Still Favor DM Over EM Chart 19China: From Tailwind to Headwind for EM ? Our longstanding call of underweight EM vs. DM since December 2013 was gradually reduced in scale, first in March 2016 (to -5 percentage points from -9) and then in October 2017 (further to -2 points). Going forward, investors should continue to maintain this slight underweight position in EM vs. DM. First, our positive stance on China proved to be timely as shown in Chart 19, panel 4, with China outperforming EM by 54.1% since March 2016, and by 18.8% in 2017. Back then our positive stance on China was supported by attractive valuations (bottom panel) and our view that Chinese politics would be supportive for global growth in the run up to the 19th Party Congress. Now BCA's Geopolitical Strategists think that "China politics are shifting from a tailwind to a headwind for global growth and EM assets".5 In addition, Chinese equities are no longer valued at a discount to the EM average (bottom panel). Second, BCA's currency view is for continued strength in the USD, especially against emerging market currencies. This does not bode well for EM/DM performance in US dollar terms (Chart 19, panel 1). Third, EM money growth leads profit growth by about three months (Chart 19, panel 2). The rolling over in money growth indicates that the currently strong earnings growth may lose steam going forward, while relative valuation is in the fair-value zone (Chart 19, panel 3). Sector Allocation: Stay Overweight Energy Our pro-cyclical sector positioning has worked well in aggregate as the market-cap-weighted cyclical index significantly outperformed the defensive index in 2017. This positioning is also in line with BCA's house view of synchronized global growth and higher inflation expectations, which translates into two major sector themes: capex recovery and rising interest rates. (Please see detailed sector positioning on page 24.) Within the cyclical space, however, the Energy sector did not perform as expected in 2017 (Chart 20). It returned only 3.4%, underperforming the global aggregate by 19.6%. For the next 9-12 months, we recommend investors to stay overweight this underdog of 2017. Chart 20Energy Stocks Lagging Oil Price First, the energy sector is a major beneficiary from a capex recovery. There are already signs of a recovery in basic resources investment in the U.S.6 Second, the energy sector's relative return lagged oil price performance in 2017. Given the generally close correlation between earnings and the oil price, and between analyst earnings revisions and OECD oil inventory growth, earnings in the sector should outpace the broad market. Third, based on price-to-cash earnings, the energy sector is still trading at about a 30% discount to the broad market, and offers a much higher dividend yield (about 1.2 points higher) than the broad market. Even though these discounts are in line with historical averages, they are still supportive of an overweight. Government Bonds Maintain Slight Underweight Duration. One important theme for 2018 will be a resumption of the cyclical uptrend in inflation.7 The implications are that both nominal bond yields and break-even inflation rates will be higher in 2018. We have been underweight duration in government bonds since July 2016. Now with the U.S. 10-year Treasury yield at 2.35%, much lower than its fair value of 2.81%, there is considerable upside risk for global bond yields from current low levels. Investors should continue to underweight duration in global government bonds Maintain Overweight Tips Vs. Treasuries. The base-case forecast from our U.S. bond strategists is that the Tips breakeven rate will rise to 2.4-2.5% as U.S. core PCE reaches the Fed's 2% target, probably sometime in the middle of 2018. Compared to the current level of 1.87%, 10-yr Tips would have upside of 33-38 bps, an important source of return in the low-return fixed-income space (Chart 21, bottom panel). In terms of relative value, Tips are now slightly cheaper than nominal bonds, also supportive of the overweight stance. Underweight Canadian Government Bonds. BCA's Global Fixed Income Strategy has taken profits in their short Canada vs. U.S. and U.K. tactical position, as the market has become too aggressive in pricing in more rate hikes in Canada. Strategically, however, the underweight of Canada (Chart 22) in a hedged global portfolio is still appropriate because: 1) the output gap has closed in Canada, according to Bank of Canada estimates, and so any additional growth will translate into higher inflation; and 2) the rising CAD will not deter the BoC from more rate hikes if the oil prices remain strong. Chart 21U.S. Bond Yields Have Further To Rise Chart 22Strategic Underweight Canadian Bonds Corporate Bonds Our overweights through most of 2017 on spread product worked well: U.S. investment grade (IG) bonds returned around 290 bps over Treasuries in the year to end-November, and high-yield bonds almost 600 bps. Returns over the next 12 months are unlikely to be as attractive. Spreads (Chart 24) are now close to historic lows: the U.S. IG bond spread, at 90 bps, is only about 30 bps above its all-time record. High-yield valuations look a little more attractive: based on our model of probable defaults over the next 12 months, the default-adjusted spread over U.S. Treasuries is likely to be around 240 bps (Chart 25). In both cases, however, investors should expect little further spread contraction, meaning that credit is now no more than a carry trade. However, in an environment where rates remain fairly low and investors continue to stretch for yield, that pick-up will remain attractive in the absence of a significant turn-down in the economic cycle. The key to watch is the shape of the yield curve. An inverted yield curve in history has been an excellent indictor of the end of the credit cycle. We expect the yield curve to steepen somewhat in H1 2018, before flattening again and then inverting late in the year. Spread product is likely, therefore, to produce decent returns until that point. Thereafter, however, the deterioration of U.S. corporate health over the past three years (Chart 23) could mean a sharp sell-off in corporate bonds. This might be exacerbated by the recent popularity of open-ended mutual funds and ETFs: a small widening of spreads could be magnified by a panicked sell-off in such funds. Chart 23Rising Leverage May Worsen Sell-Off Chart 24Credit Spreads Close To Record Lows Chart 25But Default - Adjusted, Junk Still Looks Attractive Commodities Energy: Bullish Energy prices performed strongly in H2 2017, and we expect bullish sentiment to continue. OPEC 2.0 is likely to maintain production discipline, and will maintain its promised 1.8mm b/d production cuts through the end of 2018. Our estimates for global demand growth are higher than those of other forecasters. This, along with potential unplanned production outages in Iraq, Libya and Venezuela (together accounting for 7.4mm b/d of production at present), drives our above-consensus price forecast of $67 a barrel for Brent crude during 2018. Industrial Metals: Neutral Since China accounts for more than 50% of world base-metal consumption, prices will continue to be highly dependent on developments there. (Chart 26, panel 4). Since the government is trying to accelerate environmental and supply-side reforms, domestic production capacity for base metals will shrink, which will be a positive for global metals prices. However, a focus on deleveraging in the financial sector and restructuring certain industries could slow Chinese GDP growth, reducing base-metal demand. Precious Metals: Neutral Gold has risen by 12% in 2017, supported by an uncertain geopolitical environment coupled with low interest rates. We believe that geopolitical uncertainties will persist and may even intensify, and that inflation may rise in the U.S., which would be positives for gold (Chart 26, panel 3). Based on BCA's view that stock market could be at risk from the middle of 2018,8 a moderate gold holding is warranted as a safe-haven asset. However, rising interest rate and a potentially stronger U.S. dollar are likely to limit the upside for gold. Currencies USD: The currency is down over 6% on a trade-weighted basis over the past 12 months (Chart 27). Looking into 2018, the USD is likely to perform well in the first half. U.S. inflation should gather steam in the first two to three quarters, and the Fed will be able at least to follow its dot plot - something interest rate markets are not ready for. As investors remain short the USD, upside risk to U.S. interest rates should result in a higher dollar. Chart 26Bullish Oil, Neutral Metals Chart 27Dollar Likely To Appreciate EM/JPY: Carry trades are a key mechanism for redistributing global liquidity, and they have recently begun to lose steam. A crucial reason for this has been the policy tightening in China which has been the key driver of growth in EM economies. Additionally, Japanese flows have been chasing momentum into EM assets. Further tightening in EM could reverse the flows and initiate a flight to safety, favoring the yen relative to EM currencies. CHF: The currency continues to trade at a 5% premium to its PPP fair value against the euro. However, after considering Switzerland's net international investment position at 130% of GDP, the trade-weighted CHF trades in line with fair value. The CHF will continue to behave as a risk-off currency, and so long as global volatility remains well contained, EUR/CHF will experience appreciating pressure. GBP: Sterling continues to look cheap, trading at an 18% discount to PPP against the USD. However, Brexit remains a key problem. If future immigration is limited, the U.K. will see lower trend growth relative to its neighbors, forcing its equilibrium real neutral rate downward. Consequently, it will be more difficult to finance the current account deficit of 5% of GDP. Until negotiations with the EU come closer to completion, the pound will continue to offer limited reward and plenty of volatility. Alternatives Chart 28Favor Private Equity and Farmland Alternative assets under management (AUM) have reached a record $7.7 trillion in 2017. Lower fees and a broader range of investment types have helped attract more capital. Private equity remains the most popular choice,9 driven by its strong performance and transparency. Many investors have also shifted part of their allocations toward potentially higher-return private debt programs. Return Enhancers: Favor Private Equity Vs. Hedge Funds In 2017 so far, private equity has returned 12.1%, whereas hedge funds have managed only a 5.9% return (Chart 28). We expect private-equity fund-raising to continue into 2018, but with a larger focus on niche strategies with more favorable valuations. Additionally, deploying capital gradually not only provides for vintage-year diversification, but also creates opportunities for investors to benefit from potential market corrections. We continue to favor private equity over hedge funds outside of recessions. During a recession, we recommend investors take shelter in hedge funds with a macro mandate. Inflation Hedges: Favor Direct Real Estate Vs. Commodity Futures In 2017 to date, direct real estate has returned 5.1%, whereas commodity futures are down over 3.7%. Direct real estate as an asset class continues to provide valuable diversification, lower volatility, steady yields and an illiquidity premium. However, a slowdown in U.S. commercial real estate (CRE) has made us more cautious on the overall asset class. With regards to the commodity complex, the long-term transition of the global economy to a more renewables-focused energy base will continue the structural decline in commodity demand. We continue to stress the structural and long-term nature of our negative recommendation on commodities. Volatility Dampeners: Favor Farmland & Timberland Vs. Structured Products In 2017 to date, farmland and timberland have returned 3.2% and 2.1% respectively, whereas structured products are up 3.7%. Farmland continues to outperform timberland. The slow U.S. housing recovery has added downward pressure to timberland returns. Investors can reduce the volatility of a traditional multi-asset portfolio with inclusion of farm and timber assets. For structured products, low spreads in an environment of tightening commercial real estate lending standards and falling CRE loan demand, warrant an underweight. Risks To Our View We think upside and downside risks to our central scenario for 2018 - slowing but robust economic growth, and continuing moderate outperformance of risk assets - are roughly evenly balanced. On the negative side, perhaps the biggest risk is China, where the slowdown already suggested in the monetary data (Chart 29) could be exacerbated if the government pushes ahead aggressively with structural reforms. Geopolitical risks, which the market over-emphasized in 2017, seem under-estimated now.10 U.S. trade policy, Italian elections, and North Korea all have potential to derail markets. Also, when the U.S. yield curve is as flat as it is currently, small risks can be blown up into big sell-offs. This is particularly so given over-stretched valuations for almost all asset classes. Chart 29China Monetary Conditions Suggest A Slowdown Table 2How Will Trump Try To Influence The Fed? The most likely positive surprise could come from a dovish Fed. New Fed chair Jay Powell is something of an unknown quantity, and the White House could use the three remaining Fed vacancies to push the Fed to keep rates low, so as not to offset the positive effect of the tax cuts. Without these new appointees, the Fed would have a slightly more hawkish bias in 2018 (Table 2). The intellectual argument for hiking only slowly would be, as Janet Yellen said last month: "It can be quite dangerous to allow inflation to drift down and not to achieve over time a central bank's inflation target." The Fed has missed its 2% target for five years. It is possible to imagine a situation where the Fed increasingly makes excuses to keep monetary policy easy (encouraged, for example, by a short-lived sell-off in markets or a slowdown in China) and this causes a late-cycle blow-out, similar to 1999. 1 Please see Global Investment Strategy Weekly Report, "When To Get Out," dated December 8, 2017 available at gis.bcaresearch.com. 2 Please see U.S. Equity Strategy Insight Report, "Tax Cuts Are Here - Sector Implications," dated December 12, 2017, available at uses.bcaresearch.com. 3 CBNK Survey: Monetary Base, Currency in Circulation. Source: IMF - International Financial Statistics. 4 Please see Global Investment Strategy Special Report, "Two Virtuous Dollar Circles," dated October 28, 2016, available at gis.bcaresearch.com. 5 Please see Geopolitical Strategy Special Report, "China: Party Congress Ends ... So What?" dated November 1, 2017, available at gps.bcaresearch.com. 6 Please see U.S. Equity Strategy Weekly Report, "High-Conviction Calls," dated November 27, 2017, available at uses.bcaresearch.com. 7 Please see The Bank Credit Analyst, "Outlook 2018 - Policy And The Markets: On A Collision Course," dated 20 November 2017, available at bca.bcaresearch.com. 8 Please see The Bank Credit Analyst, "Outlook 2018 - Policy And The Markets: On A Collision Course," dated November 20, 2017, available at bca.bcaresearch.com. 9 Source: BNY Mellon - The Race For Assets; Alternative Investments Surge Ahead. 10 Please see Geopolitical Strategy Weekly Report, "From Overstated To Understated Risks," dated November 22, 2017, available at gps.bcaresearch.com. GAA Asset Allocation

Dear Client, I am currently traveling in Europe visiting clients. This week, in lieu of a regular report, I am sending along a research report written by my colleague at BCA Global Asset Allocation. The topic covers one of the more fascinating "alternative" parts of the fixed income universe - catastrophe bonds. I trust that you will find this report insightful and useful. Best regards, Robert Robis, Senior Vice President Global Fixed Income Strategy Highlights Catastrophe bonds ("cat bonds") have recently been receiving a lot of investor attention because, after this summer's large hurricanes, they are now attractively priced. We explain the mechanics of this market, and analyze cat bonds' historic risk-return characteristics. Cat bonds have historical annualized returns of 7.4%, with volatility of only 3.0%, making them an attractive risk-adjusted investment. However, they are exposed to "cliff risk", creating a return distribution with negative skew and large excess kurtosis. But cat bonds offer interesting portfolio diversification benefits, since financial and economic shocks have minimal impact on cat bond returns. The reinsurance market tends to be cyclical, with premiums rising following a catastrophe and decreasing during a period of calm. Feature Introduction In 1992, Hurricane Andrew caused $17 billion in losses, more than twice the value of the insured property, and forced many insurers into bankruptcy. As the global economy has grown in size since then, the monetary value of insured events has risen steadily. However, increasing regulatory hurdles in the form of higher reserve requirements have led to capacity constraints (Chart 1) in the traditional insurance industry. In 2005, Hurricane Katrina, which caused $108 billion in losses, strengthened the case for the introduction of catastrophe bonds and other insurance-linked securities that helped ease financial burdens in the insurance industry, for several reasons. First, catastrophe bonds give access to the deepest, most liquid, and efficient sources of capital. Second, the securitization of reinsurance capital has created a secondary market where risk exposures can be transferred within the investor community. Third, insurance firms have the ability to move some exposures off their books, thereby allowing them to underwrite larger risks that they would otherwise lack the capacity to cover. According to S&P Global Ratings, the market for cat bonds and other insurance-linked securities is estimated to be about $86 billion. Other insurance-linked securities include industry loss warranties (ILW), collateralized reinsurance contracts, and reinsurance sidecars. Cat bonds are the only insurance-linked securities that publicly trade on a secondary market. The recent increase in natural catastrophes has led to surging supply in the cat bond market. Record issuance in the first and second quarters of 2017 has pushed the size of the outstanding cat bond market to over $30 billion (Chart 2) for the first time. This comes after a period prior to this year with fewer catastrophes and where bond pricing has been stable, which led to increased deal sizes. In this Special Report, we run through the mechanics of the cat bond structure and market. We analyze historical risk-return characteristics (Chart 3) and compare them to other major asset classes. Since insurance-linked securities are known to have very low correlation with other assets, we test their potential diversification benefits within a traditional portfolio. Finally, we analyze their historical performance in periods of financial market stress and rising interest rate environments, which are two of the biggest worries for investors. Our conclusions are that: Chart 1Capacity Constraints Chart 2Record Issuance In 2017 Chart 3Risk-Return Profile The reinsurance market is cyclical, with premiums increasing following a catastrophe and decreasing following a period of calm. Realized volatility in the cat bond market is low. However, returns have a negative skew with an extremely fat-tailed distribution relative to other traditional assets. The addition of cat bonds to a traditional multi-asset portfolio has tremendous diversification benefits. The largest improvement to risk-adjusted returns comes from substituting equities with cat bonds. Financial crises have minimal impact on cat bond returns. However, depending on the magnitude of catastrophe losses, there could be varied regional impacts. Investors can customize the risk-return profile by altering the attachment and exhaustion points, and also by diversifying across trigger types. Mechanics Of Cat Bonds Despite the increasing popularity of cat bonds, their non-conventional structure is understood by only a limited number of investors. A better understanding of the characteristics of this financial instrument makes analyzing risk and return more straightforward. The key features (Chart 4) of a catastrophe bond are as follows. An insurer looking to reduce certain exposures will create a special purpose vehicle (SPV), also known as the issuer, to assist with the transaction. The issuer/SPV sells reinsurance protection to the sponsoring firms and simultaneously issues a cat bond to the investor. The proceeds from the bond sale are managed in a segregated collateral account to generate the floating-rate component of the coupon payable to investors. The fixed component of the coupon is financed through reinsurance premiums paid by the sponsoring firm to the issuer or SPV. Traditionally, cat bonds used a total return swap where a counterparty guaranteed the liquidity and performance of a collateral account. This forced investors and sponsors to rely on the creditworthiness of the swap provider. In 2007, two cat bonds that used Lehman Brothers as a swap counterparty were forced into default because of illiquid collateral assets and mismatched maturities. Nowadays, the assets managed in the collateral account are invested only in U.S. Treasury money market funds or structured notes from the International Bank for Reconstruction and Development (IBRD). The final settlement of the bond is binary: 1) if no trigger event occurs before the bond maturity, the SPV returns the principal to investors along with the final coupon; 2) if a catastrophe hits and the bond is triggered, the principal in the collateral account is used to settle the claims of the sponsoring firms. Cat bonds are typically used to cover a piece of risk exposure in the sponsor's book. For example, a cat bond could cover indemnities exceeding $1 billion up to $1.2 billion, making the bond issue size equal to $200 million. The $1 billion is called the attachment point, and the $1.2 billion is called the exhaustion point, at which point the principal is exhausted and investors are not liable for any further claims. The tranche with the higher attachment point will be of higher quality, but with a lower rate of return. The reinsurance industry is cyclical, which makes contract pricing more volatile than investors might expect. The Rate on Line (Chart 5) index can be seen as a yield on the insurance contracts underwritten in the industry. Market conditions can be split into two phases: Chart 4Mechanics Of Cat Bonds Chart 5Cyclical Reinsurance Premiums Soft Market: Following many years of limited or minor catastrophes, reinsurance premiums are pressured downward and bond prices rise. In these circumstances, demand for cat bonds will be limited as coupon income will be less attractive. Hard Market: A major catastrophe will significantly erode the capital available in the insurance industry, thereby creating a supply shortage that pushes up reinsurance premiums. In these conditions, cat bond issuance will rise, driven by attractive coupon income. Investors can manage the premium cycle by slightly increasing risk at the portfolio level in a softening market (falling premiums) and reducing risk in hardening market (rising premiums). The recent catastrophes should drive up reinsurance premiums, but the sheer weight of money searching for yields in the current environment might make the uplift surprisingly modest compared to the past. Given that cat bonds have a binary payout feature, investors need to understand the trigger type (Table 1) used in the contract. In the early days, most bonds were issued with an indemnity trigger, but the type of trigger (Chart 6) has become more varied over time. The type of trigger used in the cat bond has the following impacts: If the trigger used in the bond takes longer to settle, the investor can be involved in a long drawn-out legal battle with the sponsoring firm looking to settle claims. This could in turn force the bond beyond maturity and keep investor funds locked up at significantly lower rates of return. Table 1Understanding Trigger Types Chart 6Choosing The Right Trigger Type Investors also need to understand the level of basis risk sponsoring firms are exposed to with different trigger types. In the context of cat bonds, basis risk is when the settlement payout from cat bonds differs from the actual portfolio losses incurred by sponsoring firms. If they have basis risk, investors will have to deal with moral hazard, where sponsoring firms will have incentive to underwrite excessive risks. Historical Risk & Return Investing in catastrophe bonds is essentially a "short gamma" strategy, where investors are selling insurance and collecting premium with the hope of options not being triggered during the maturity of the bond. Attractive historical returns (Table 2) have been the result of lower-than-expected principal write-downs given limited catastrophes. In the early years, cat bonds as an asset class were not fully understood by the broader market, creating a "novelty premium" up until 2010. Subsequently, low interest rates have had a profound impact on all traditional assets, making cat bond yields relatively attractive. Realized volatility has been extremely low since the investor collects regular coupons in the absence of a catastrophe that triggers a payout. This makes risk-adjusted returns very attractive compared to other major assets. However, because of the extreme tail risk, there exists a big negative skew along with high excess kurtosis. Cat bonds are exposed to "cliff risk" - the likelihood that the tranche's notional value will be exhausted once settlement claims reach the attachment point. The two main sources of risk that investors need to be mainly concerned about, however, are: 1) insurance risk that cat bonds assume, and 2) credit risk associated with the collateral account. An attractive feature of cat bonds is that poor performance tends to be self-correcting, as seen in the reinsurance cycle. Following a particularly destructive natural disaster, a number of factors such as increased insurance demand, the reduced capacity of insurance firms, and upward revisions to probability models serve to increase insurance premiums and potential returns to insurance-linked securities. For example, after the 2011 Japanese Tohoku earthquake and tsunami, insurance premiums were pushed up by around 50% for earthquake risk and 20% for other catastrophe risk. The likelihood of incurring negative returns is far lower than the chance of benefitting from positive returns. Cat bonds have achieved positive monthly returns 92% of the time (Table 3). The recent hurricane season in the U.S. was the first time returns turned negative on a 12-month basis. Table 2Historical Risk-Return Analysis (January 2002 - November 2017) Table 3Only Fifteen Months Of Negative Returns Finally, there have been many comparisons between cat bonds and high-yield credit. While high-yield debt performance is tied to market and economic cycles lasting about 10 years, that of cat bonds is tied to low probability catastrophes. Frequency of loss in junk bonds is greater than it is for cat bonds. However, the potential principal loss is greater for cat bonds, because they have almost zero recovery value. Diversification & Portfolio Impact Cat bonds' performance is linked to factors such as natural disasters, longevity risk, or life insurance mortality, and not to broader financial market risks. However, in periods of economic stress, markets experience a flight to quality and correlations between risk assets increase. Therefore, the benefits of portfolio diversification dissolve when they are needed most. This is not the case with cat bonds, however, as correlations with other assets (Table 4) have remained stable over time. This makes them a potentially useful diversification instrument in multi-asset portfolios. Table 4Cross-Asset Correlation (January 2002 - November 2017) To test this, we perform a typical portfolio analysis whereby we add cat bonds to a conventional portfolio and investigate the impact on the return and risk of the portfolio (Chart 7). Starting with the most traditional allocation of 60% equities and 40% bonds, we augment the portfolio with a 10% allocation to cat bonds and come up with the following results: Replacing equities with cat bonds leads to the largest reduction in portfolio volatility, and a small decrease in annualized returns. This new portfolio generates equity-like returns, but with a smaller correlation with stocks. Replacing traditional fixed income with cat bonds leads to a large increase to annualized returns, while the impact on volatility is virtually non-existent. The largest positive impact on risk-adjusted returns occurs when cat bonds replace equities, because the reduction in volatility is substantially greater than the increase in returns when cat bonds replace traditional bonds. We also ranked the MSCI All-Country World equity and Bloomberg Barclays Global Aggregate Bond indices from worst to best monthly returns and then overlaid the corresponding cat bond returns for each ranked month (Chart 8). This technique removes randomness from the time series in order to view the relative randomness of the other. We have the following findings: Cat bonds have had only three months that delivered a return less than -2%. These were -2.1% in September 2005 during Hurricane Katrina, -3.6% in March 2011 during the Tohoku earthquake and tsunami in Japan, and -5.8% in September 2017 after the severe hurricanes in Texas, Florida and the Caribbean. Other than catastrophe-related events, cat bond returns have been stable. Cat bonds displayed no reaction when equities had their most negative months. But they tend to have relatively stronger returns when equities also have positive months. Cat bonds performed well in both good and bad months for traditional fixed income. This shows that causes of traditional bond market losses and cat bond principal loss have little or no bearing on one another. Since cat bonds have a large negative skew and high excess kurtosis, investors can potentially lose all their capital if the bonds are triggered. When allocating to cat bonds, investors need to maintain a well-diversified position in order to minimize the risk of complete capital wipeout. This can be done by carefully picking bonds covering different perils (i.e. earthquakes, wind, extreme mortality), regions and trigger types (Chart 9). As a broader range of perils come to the market, investors will find increasing avenues for diversification within the asset class. Investors can also benefit from very low correlations within the cat bond universe, where returns from cat bonds covering a specific peril have no bearing on returns from cat bonds covering another peril. Chart 7Portfolio Diversification Chart 8Attractive Monthly Returns Chart 9Diversifying Across Perils, Coupon Rate And Expected Loss Financial Market Stress Having established that underlying market developments have no bearing on cat bond performance, we want to address two further important questions: 1) do financial crises affect cat bond returns? 2) do natural catastrophes trigger financial crises? Looking at previous global market crisis scenarios dating back to 2008 (Chart 10), we see that cat bonds had positive absolute returns during all crisis periods. The only period with negative cat bond returns was during the 2008 Lehman Brothers' collapse, when the bank was the swap counterparty for two bonds that defaulted. Large natural catastrophes do not affect broader capital markets, but do tend to have a large local impact. In August 2005, Hurricane Katrina, with damages totaling $108 billion, became the costliest hurricane to date in the U.S. The hurricane triggered a cat bond, and the index was down 2.1%, but there was no noticeable lingering impact on the U.S. economy. On the other hand, the earthquake and tsunami in Tohoku on March 11, 2011 had devastating effects. With damages exceeding $300 billion (approximately 5% of Japanese GDP), the cat bond index dropped 3.6%, and Japanese equities collapsed 7.3%. Moreover, a big earthquake in a major city or region such as Tokyo or California could have the capacity to trigger a global recession. Finally, looking at past major catastrophes (Chart 11), we see that existing cat bond prices do not fully recover to their pre-catastrophe levels. Accordingly, picking up bonds at a discount may not generate the expected return as price levels struggle to fully recover. Chart 10Outperformance Across The Board Chart 11Not A Full Recovery Interest Rate & Inflation Hedge Traditional bonds with fixed coupon payments underperform in a rising rate environment. Since cat bonds receive a floating-rate coupon along with the fixed premium, they are largely immune to rising rates. When central banks hike rates, the principal of the bonds invested in money market assets will produce a higher return, thereby offering investors a powerful shield against possible inflation, as well. Since the total coupon received by investors includes a fixed and floating component, cat bonds have a lower modified duration relative to similar maturity traditional bonds. Conclusion Despite their abnormal return distributions, we recommend investors allocate capital from their "alternatives" bucket toward cat bonds. Against a backdrop of low yields and investor complacency, cat bonds are highly attractive given their potential for consistently robust returns and, perhaps most importantly, tremendous diversification benefits. Still, allocations should be relatively small given the illiquid nature of the cat bond market, and diversification among bonds and issuers is critical due to the potential for large losses in the event that a cat bond is triggered. Aditya Kurian, Research Analyst Global Asset Allocation adityak@bcaresearch.com

Highlights Catastrophe bonds ("cat bonds") have recently been receiving a lot of investor attention because, after this summer's large hurricanes, they are now attractively priced. We explain the mechanics of this market, and analyze cat bonds' historic risk-return characteristics. Cat bonds have historical annualized returns of 7.4%, with volatility of only 3.0%, making them an attractive risk-adjusted investment. However, they are exposed to "cliff risk", creating a return distribution with negative skew and large excess kurtosis. But cat bonds offer interesting portfolio diversification benefits, since financial and economic shocks have minimal impact on cat bond returns. The reinsurance market tends to be cyclical, with premiums rising following a catastrophe and decreasing during a period of calm. Feature Introduction In 1992, Hurricane Andrew caused $17 billion in losses, more than twice the value of the insured property, and forced many insurers into bankruptcy. As the global economy has grown in size since then, the monetary value of insured events has risen steadily. However, increasing regulatory hurdles in the form of higher reserve requirements have led to capacity constraints (Chart 1) in the traditional insurance industry. In 2005, Hurricane Katrina, which caused $108 billion in losses, strengthened the case for the introduction of catastrophe bonds and other insurance-linked securities that helped ease financial burdens in the insurance industry, for several reasons. First, catastrophe bonds give access to the deepest, most liquid, and efficient sources of capital. Second, the securitization of reinsurance capital has created a secondary market where risk exposures can be transferred within the investor community. Third, insurance firms have the ability to move some exposures off their books, thereby allowing them to underwrite larger risks that they would otherwise lack the capacity to cover. According to S&P Global Ratings, the market for cat bonds and other insurance-linked securities is estimated to be about $86 billion. Other insurance-linked securities include industry loss warranties (ILW), collateralized reinsurance contracts, and reinsurance sidecars. Cat bonds are the only insurance-linked securities that publicly trade on a secondary market. The recent increase in natural catastrophes has led to surging supply in the cat bond market. Record issuance in the first and second quarters of 2017 has pushed the size of the outstanding cat bond market to over $30 billion (Chart 2) for the first time. This comes after a period prior to this year with fewer catastrophes and where bond pricing has been stable, which led to increased deal sizes. In this Special Report, we run through the mechanics of the cat bond structure and market. We analyze historical risk-return characteristics (Chart 3) and compare them to other major asset classes. Since insurance-linked securities are known to have very low correlation with other assets, we test their potential diversification benefits within a traditional portfolio. Finally, we analyze their historical performance in periods of financial market stress and rising interest rate environments, which are two of the biggest worries for investors. Our conclusions are that: Chart 1Capacity Constraints Chart 2Record Issuance In 2017 Chart 3Risk-Return Profile The reinsurance market is cyclical, with premiums increasing following a catastrophe and decreasing following a period of calm. Realized volatility in the cat bond market is low. However, returns have a negative skew with an extremely fat-tailed distribution relative to other traditional assets. The addition of cat bonds to a traditional multi-asset portfolio has tremendous diversification benefits. The largest improvement to risk-adjusted returns comes from substituting equities with cat bonds. Financial crises have minimal impact on cat bond returns. However, depending on the magnitude of catastrophe losses, there could be varied regional impacts. Investors can customize the risk-return profile by altering the attachment and exhaustion points, and also by diversifying across trigger types. Mechanics Of Cat Bonds Despite the increasing popularity of cat bonds, their non-conventional structure is understood by only a limited number of investors. A better understanding of the characteristics of this financial instrument makes analyzing risk and return more straightforward. The key features (Chart 4) of a catastrophe bond are as follows. An insurer looking to reduce certain exposures will create a special purpose vehicle (SPV), also known as the issuer, to assist with the transaction. The issuer/SPV sells reinsurance protection to the sponsoring firms and simultaneously issues a cat bond to the investor. The proceeds from the bond sale are managed in a segregated collateral account to generate the floating-rate component of the coupon payable to investors. The fixed component of the coupon is financed through reinsurance premiums paid by the sponsoring firm to the issuer or SPV. Traditionally, cat bonds used a total return swap where a counterparty guaranteed the liquidity and performance of a collateral account. This forced investors and sponsors to rely on the creditworthiness of the swap provider. In 2007, two cat bonds that used Lehman Brothers as a swap counterparty were forced into default because of illiquid collateral assets and mismatched maturities. Nowadays, the assets managed in the collateral account are invested only in U.S. Treasury money market funds or structured notes from the International Bank for Reconstruction and Development (IBRD). The final settlement of the bond is binary: 1) if no trigger event occurs before the bond maturity, the SPV returns the principal to investors along with the final coupon; 2) if a catastrophe hits and the bond is triggered, the principal in the collateral account is used to settle the claims of the sponsoring firms. Cat bonds are typically used to cover a piece of risk exposure in the sponsor's book. For example, a cat bond could cover indemnities exceeding $1 billion up to $1.2 billion, making the bond issue size equal to $200 million. The $1 billion is called the attachment point, and the $1.2 billion is called the exhaustion point, at which point the principal is exhausted and investors are not liable for any further claims. The tranche with the higher attachment point will be of higher quality, but with a lower rate of return. The reinsurance industry is cyclical, which makes contract pricing more volatile than investors might expect. The Rate on Line (Chart 5) index can be seen as a yield on the insurance contracts underwritten in the industry. Market conditions can be split into two phases: Chart 4Mechanics Of Cat Bonds Chart 5Cyclical Reinsurance Premiums Soft Market: Following many years of limited or minor catastrophes, reinsurance premiums are pressured downward and bond prices rise. In these circumstances, demand for cat bonds will be limited as coupon income will be less attractive. Hard Market: A major catastrophe will significantly erode the capital available in the insurance industry, thereby creating a supply shortage that pushes up reinsurance premiums. In these conditions, cat bond issuance will rise, driven by attractive coupon income. Investors can manage the premium cycle by slightly increasing risk at the portfolio level in a softening market (falling premiums) and reducing risk in hardening market (rising premiums). The recent catastrophes should drive up reinsurance premiums, but the sheer weight of money searching for yields in the current environment might make the uplift surprisingly modest compared to the past. Given that cat bonds have a binary payout feature, investors need to understand the trigger type (Table 1) used in the contract. In the early days, most bonds were issued with an indemnity trigger, but the type of trigger (Chart 6) has become more varied over time. The type of trigger used in the cat bond has the following impacts: If the trigger used in the bond takes longer to settle, the investor can be involved in a long drawn-out legal battle with the sponsoring firm looking to settle claims. This could in turn force the bond beyond maturity and keep investor funds locked up at significantly lower rates of return. Table 1Understanding Trigger Types Chart 6Choosing The Right Trigger Type Investors also need to understand the level of basis risk sponsoring firms are exposed to with different trigger types. In the context of cat bonds, basis risk is when the settlement payout from cat bonds differs from the actual portfolio losses incurred by sponsoring firms. If they have basis risk, investors will have to deal with moral hazard, where sponsoring firms will have incentive to underwrite excessive risks. Historical Risk & Return Investing in catastrophe bonds is essentially a "short gamma" strategy, where investors are selling insurance and collecting premium with the hope of options not being triggered during the maturity of the bond. Attractive historical returns (Table 2) have been the result of lower-than-expected principal write-downs given limited catastrophes. In the early years, cat bonds as an asset class were not fully understood by the broader market, creating a "novelty premium" up until 2010. Subsequently, low interest rates have had a profound impact on all traditional assets, making cat bond yields relatively attractive. Realized volatility has been extremely low since the investor collects regular coupons in the absence of a catastrophe that triggers a payout. This makes risk-adjusted returns very attractive compared to other major assets. However, because of the extreme tail risk, there exists a big negative skew along with high excess kurtosis. Cat bonds are exposed to "cliff risk" - the likelihood that the tranche's notional value will be exhausted once settlement claims reach the attachment point. The two main sources of risk that investors need to be mainly concerned about, however, are: 1) insurance risk that cat bonds assume, and 2) credit risk associated with the collateral account. An attractive feature of cat bonds is that poor performance tends to be self-correcting, as seen in the reinsurance cycle. Following a particularly destructive natural disaster, a number of factors such as increased insurance demand, the reduced capacity of insurance firms, and upward revisions to probability models serve to increase insurance premiums and potential returns to insurance-linked securities. For example, after the 2011 Japanese Tohoku earthquake and tsunami, insurance premiums were pushed up by around 50% for earthquake risk and 20% for other catastrophe risk. The likelihood of incurring negative returns is far lower than the chance of benefitting from positive returns. Cat bonds have achieved positive monthly returns 92% of the time (Table 3). The recent hurricane season in the U.S. was the first time returns turned negative on a 12-month basis. Table 2Historical Risk-Return Analysis (January 2002 - November 2017) Table 3Only Fifteen Months Of Negative Returns Finally, there have been many comparisons between cat bonds and high-yield credit. While high-yield debt performance is tied to market and economic cycles lasting about 10 years, that of cat bonds is tied to low probability catastrophes. Frequency of loss in junk bonds is greater than it is for cat bonds. However, the potential principal loss is greater for cat bonds, because they have almost zero recovery value. Diversification & Portfolio Impact Cat bonds' performance is linked to factors such as natural disasters, longevity risk, or life insurance mortality, and not to broader financial market risks. However, in periods of economic stress, markets experience a flight to quality and correlations between risk assets increase. Therefore, the benefits of portfolio diversification dissolve when they are needed most. This is not the case with cat bonds, however, as correlations with other assets (Table 4) have remained stable over time. This makes them a potentially useful diversification instrument in multi-asset portfolios. Table 4Cross-Asset Correlation (January 2002 - November 2017) To test this, we perform a typical portfolio analysis whereby we add cat bonds to a conventional portfolio and investigate the impact on the return and risk of the portfolio (Chart 7). Starting with the most traditional allocation of 60% equities and 40% bonds, we augment the portfolio with a 10% allocation to cat bonds and come up with the following results: Replacing equities with cat bonds leads to the largest reduction in portfolio volatility, and a small decrease in annualized returns. This new portfolio generates equity-like returns, but with a smaller correlation with stocks. Replacing traditional fixed income with cat bonds leads to a large increase to annualized returns, while the impact on volatility is virtually non-existent. The largest positive impact on risk-adjusted returns occurs when cat bonds replace equities, because the reduction in volatility is substantially greater than the increase in returns when cat bonds replace traditional bonds. We also ranked the MSCI All-Country World equity and Bloomberg Barclays Global Aggregate Bond indices from worst to best monthly returns and then overlaid the corresponding cat bond returns for each ranked month (Chart 8). This technique removes randomness from the time series in order to view the relative randomness of the other. We have the following findings: Cat bonds have had only three months that delivered a return less than -2%. These were -2.1% in September 2005 during Hurricane Katrina, -3.6% in March 2011 during the Tohoku earthquake and tsunami in Japan, and -5.8% in September 2017 after the severe hurricanes in Texas, Florida and the Caribbean. Other than catastrophe-related events, cat bond returns have been stable. Cat bonds displayed no reaction when equities had their most negative months. But they tend to have relatively stronger returns when equities also have positive months. Cat bonds performed well in both good and bad months for traditional fixed income. This shows that causes of traditional bond market losses and cat bond principal loss have little or no bearing on one another. Since cat bonds have a large negative skew and high excess kurtosis, investors can potentially lose all their capital if the bonds are triggered. When allocating to cat bonds, investors need to maintain a well-diversified position in order to minimize the risk of complete capital wipeout. This can be done by carefully picking bonds covering different perils (i.e. earthquakes, wind, extreme mortality), regions and trigger types (Chart 9). As a broader range of perils come to the market, investors will find increasing avenues for diversification within the asset class. Investors can also benefit from very low correlations within the cat bond universe, where returns from cat bonds covering a specific peril have no bearing on returns from cat bonds covering another peril. Chart 7Portfolio Diversification Chart 8Attractive Monthly Returns Chart 9Diversifying Across Perils, Coupon Rate And Expected Loss Financial Market Stress Having established that underlying market developments have no bearing on cat bond performance, we want to address two further important questions: 1) do financial crises affect cat bond returns? 2) do natural catastrophes trigger financial crises? Looking at previous global market crisis scenarios dating back to 2008 (Chart 10), we see that cat bonds had positive absolute returns during all crisis periods. The only period with negative cat bond returns was during the 2008 Lehman Brothers' collapse, when the bank was the swap counterparty for two bonds that defaulted. Large natural catastrophes do not affect broader capital markets, but do tend to have a large local impact. In August 2005, Hurricane Katrina, with damages totaling $108 billion, became the costliest hurricane to date in the U.S. The hurricane triggered a cat bond, and the index was down 2.1%, but there was no noticeable lingering impact on the U.S. economy. On the other hand, the earthquake and tsunami in Tohoku on March 11, 2011 had devastating effects. With damages exceeding $300 billion (approximately 5% of Japanese GDP), the cat bond index dropped 3.6%, and Japanese equities collapsed 7.3%. Moreover, a big earthquake in a major city or region such as Tokyo or California could have the capacity to trigger a global recession. Finally, looking at past major catastrophes (Chart 11), we see that existing cat bond prices do not fully recover to their pre-catastrophe levels. Accordingly, picking up bonds at a discount may not generate the expected return as price levels struggle to fully recover. Chart 10Outperformance Across The Board Chart 11Not A Full Recovery Interest Rate & Inflation Hedge Traditional bonds with fixed coupon payments underperform in a rising rate environment. Since cat bonds receive a floating-rate coupon along with the fixed premium, they are largely immune to rising rates. When central banks hike rates, the principal of the bonds invested in money market assets will produce a higher return, thereby offering investors a powerful shield against possible inflation, as well. Since the total coupon received by investors includes a fixed and floating component, cat bonds have a lower modified duration relative to similar maturity traditional bonds. Conclusion Despite their abnormal return distributions, we recommend investors allocate capital from their "alternatives" bucket toward cat bonds. Against a backdrop of low yields and investor complacency, cat bonds are highly attractive given their potential for consistently robust returns and, perhaps most importantly, tremendous diversification benefits. Still, allocations should be relatively small given the illiquid nature of the cat bond market, and diversification among bonds and issuers is critical due to the potential for large losses in the event that a cat bond is triggered. Aditya Kurian, Research Analyst Global Asset Allocation adityak@bcaresearch.com

Highlights Dear Client, I'm on the road this week teaching the BCA Academy in Chicago. Instead of our regular Weekly Report, we are sending you a Special Report written by my colleague Juan Manuel Correa. His piece, "Riding the Wave: Momentum Strategies in Foreign Exchange Markets," focuses on the application of momentum strategies in the FX space. More specifically, Juan lays out the case that momentum is now pointing to upside in the U.S. dollar. I trust you find his report both informative and enjoyable. Best regards, Mathieu Savary, Vice President, Foreign Exchange Strategy Feature Merchant: In this chaos of opinions, which is the most prudent? Shareholder: To go in the direction of the waves, and not fight against powerful currents - Confusion de Confusiones, Joseph de la Vega, 1688. Since the invention of financial markets, momentum has captivated the minds of investors, economists and general speculators. As early as 1688, the Spanish merchant Jose de la Vega became the first market observer to document the powerful forces of momentum in the primitive financial markets of Amsterdam.1 Since then, a number of academic studies have confirmed that momentum strategies deliver significant excess returns, even when traditional risk factors are taken into account.2 Because the success of momentum flies in the face of the Efficient Market Hypotheses, academia has tried to understand this phenomenon. Transaction costs, short-selling constraints and unsophisticated market participants have been among some of the explanations advanced and more widely accepted. However, there is still no real consensus as to why momentum strategies work. Foreign exchange markets present themselves as a fascinating space to study momentum, given that FX markets are:3 a) Very liquid, and possess very low transaction costs; b) Include no short selling constraints; c) Are populated by very sophisticated investors. So how successful are momentum strategies in foreign exchange markets? More specifically: In what time frame does momentum work best? In which currencies or crosses are momentum strategies more effective? Are there any macroeconomic factors that influence the success of a momentum strategy? Generally, momentum in financial markets is defined as the positive correlation between past and future returns. Momentum can either refer to time series momentum (buy/sell a currency which has had positive/negative returns) or cross-sectional momentum (buy the best-performing currencies and sell the worst-performing currencies). In this report, we will focus on time-series momentum. We use moving average crossovers to generate signals. We chose this technique as it is commonly used by practitioners, and it provides an easy and flexible buy/sell signal. When a short-term moving average crosses a long-term one from below, we buy the cross. Conversely, when it crosses it from above, we short the cross. While it is true that this technique does not follow the strict definition of momentum, it is a close enough proxy, as it takes into account the relative acceleration of the price. Furthermore, we tested 15 different combinations of moving averages on all 45 crosses in the G10, on a sample of nearly 29 years. By doing this we do not bias our analysis to dollar pairs or to any particular strategy. For more details on the methodology, please see Appendix A. Wave Watching: Observations On Historical Returns Our strategies consist of 15 different combinations of 1-month, 2-month, 3-month, 6-month, 12-month and 24-month moving averages. On average, momentum strategies had an annualized spot return of 0.5% and a carry return of 0.9% from when our sample period started in January 1989 to its end in October 2017 (Chart I-1). Furthermore, most strategies provided positive returns on average (see Appendix B) while substantially decreasing drawdowns (see Appendix D, Table 1). Chart I-1Momentum Across History However, some strategies performed better than others. On average, we found that momentum strategies based on the "medium-term" - i.e. when the slower of the two moving averages necessary to generate the crossovers was either 130-days (6-months) or 260-days (12-months) - tended to perform best. In terms of nomenclature in our comparative study, we named each strategy by summing the number of days in the faster moving average and the slower one. The resulting number is the total amount of days considered by the strategy. This way shorter term-focused strategies have lower numbers while longer-term focused strategies have higher numbers (Appendix A, Table 1). We found that risk-adjusted returns for strategies focused on the short term tend to be low: they rise as strategies become more focused on medium-term horizons, and then they drop again when longer term moving-average crossovers are used, following a "hump" pattern (Chart I-2). This pattern holds across the majority of FX crosses (see Appendix C). Our results are consistent with the literature on momentum on other assets classes. Generally, short-term returns tend to be reverting: if an asset's return last month was positive it will likely be negative the following month. The reversal effect tends to also be present in the long-term: if an asset experienced strong positive returns on a multi-year horizon, it is likely to offer negative returns in the subsequent time period. On the other hand, positive return auto correlation, the staple of traditional momentum strategies, tends to be strongest in medium-term time frames.4 Next, we examined the carry component of the strategies. On average, momentum strategies are long carry currencies slightly more often than not, and vice versa with funding currencies. As a result, momentum strategies tend to generate a positive carry (Chart I-3). Chart I-2Medium Term Focused Strategies ##br##Perform Best Chart I-3Momentum Strategies Favor ##br##Carry Currencies... This result is robust across strategies and across currency pairs (see Appendix B & C). Of the 675 different return indexes generated by our various moving average crossover signals, only 108 had a negative carry. So, are momentum strategies and carry strategies one and the same? Not quite. When we tested the correlation between the returns of our G10 carry strategy Index and the returns of all 15 of our momentum indexes, we found it to be nearly zero. Furthermore, we found that the spot returns of momentum strategies tended to increase in periods of increasing G10 implied volatility (Chart I-4). This stands in stark contrast to carry strategies, which are allergic to any increase in volatility.5 Chart I-4...But Momentum Also Likes Volatility We also tested for which crosses momentum strategies worked best. We found that commodity crosses tend to be the worst performers, with the least reliable and least rewarding signals. Meanwhile, pairs involving the yen or the U.S. dollar in one of the legs tended to perform the best by a wide margin, in both spot terms and carry terms (Chart I-5). Chart I-5AMomentum Winners: ##br##USD And JPY Crosses Chart I-5BMomentum Winners: ##br##USD And JPY Crosses Bottom Line: Historically, momentum strategies have provided positive returns. However, medium term-focused strategies tend to perform best. Momentum strategies also tend to produce positive carry, even though their spot return rises along with volatility. Finally, crosses involving a USD or JPY leg tend to provide the best momentum returns. Characteristics Of Momentum: Wave Patterns And Surfing Lessons We opted to take an unconventional approach from the plethora of academic research trying to understand momentum. However, to do so, we needed to momentarily step away from financial markets and instead dive in another field where riding waves is paramount: surfing. Diagram 1Oceanic Wave Patters Oceanic waves are produced by the wind. When wind blows across the surface of the ocean, the force is transferred to the water and generates swell, which is a group of travelling waves.6 However not all swell is created equally. There are two main types of swell: groundswell and windswell. Groundswell is the result of powerful winds or storms thousands of miles away from shore. These strong storm systems far away in the ocean tend to generate smooth and infrequent waves. These are the best waves for surfing, as these waves create enough power for a surfer to gain great balance and thus, ride the wave for a long period of time (Diagram 1 - Top Panel). On the other hand, windswell refers to swell created by local winds. These local winds tend to generate smaller waves and choppy waters, which makes for lower-quality surfing (Diagram 1 - Bottom Panel). This insight from surfing can be translated to financial markets. Much like a surfer at the beach, a momentum player would prefer smooth waves in the currencies he or she trades, as these types of waves can provide consistent signals that he or she can take advantage of. We therefore tested whether currencies that behave like groundswell tend to have higher risk-adjusted momentum returns than currencies that behave like windswell. How can we test this numerically? We found that volatility is not the right measure to capture this particular wave pattern, as it does not account for smoothness (see Appendix D). Instead, we measured smoothness by calculating a cross's average 1-year fractal dimension,7 a modification of an indicator championed by BCA's European Investment Strategy's Dhaval Joshi. A low average fractal dimension over that 1-year window indicates that more often than not a cross has been following a smooth trend, while an elevated fractal dimension indicates a cross that has been range-bound.8 We invert this number, giving higher numbers to smoother, trending crosses and lower numbers to jagged, noisy crosses. We call this the "Wave Smoothness Indicator," and it turns out to be highly correlated to risk-adjusted momentum returns for crosses in the G10, particularly if we take out managed crosses like EUR/CHF, EUR/SEK, and EUR/NOK (Chart I-6). To further illustrate this point, we sorted all crosses by their median risk-adjusted returns across all the moving-average crossover strategies we tested. We then looked at the five crosses where our momentum strategies delivered the higher risk-adjusted returns against the five crosses where the strategies fared the worst (Chart I-7A & Chart I-7B). The best currencies to execute momentum strategies have long and smooth cycles, while the worst ones exhibit much more noise. Chart I-6Wave Dynamics And Momentum Returns Chart I-7AGroundswell: Paradise For Momentum Surfers Chart I-7BWindswell: No Wave Riding In Choppy Waters As a result, it is apparent that smoothness is a crucial factor behind successful momentum trading, at least in the FX space. For example, while AUD/NZD displays long cycles, these gyrations are not smooth. Consequently, moving-average crossover strategies work badly for this cross, as it is too noisy to provide reliable buy/sell signals. Bottom Line: Analogous to the dynamic between surfers and oceanic waves, currencies that have long and smooth cycles (groundswell) tend to provide better returns than currencies which have small and noisy cycles (windswell). Storm Warning: Macro Determinants Of Momentum What factors make a currency behave more like groundswell as opposed to windswell? In order to gain some understanding, let's look at the crosses where momentum strategies worked best in our sample: the USD crosses and the JPY crosses. The yen and the dollar experience such strong and broad-based trends that for any cross, simply being correlated to the trade-weighted dollar and the trade-weighted yen makes for a good predictor of whether this currency pair will experience strong momentum-continuation behavior. Moreover, in line with our results above, crosses with a high correlation to these currencies also tend to exhibit stronger groundswell patterns (Chart I-8). What is so special about the dollar and the yen? The oceanic waves once again offer a clue. Recall that groundswell is generated by powerful oceanic storms. Similarly, the trade-weighted dollar and yen are ultra-sensitive to two of the most powerful forces in the global economy: global trade dynamics and global risk aversion (Chart I-9). Chart I-8JPY And USD Determine Wave ##br##Patterns In Currency Markets Chart I-9The Powerful Winds Of ##br##The Global Economy Global trade and risk aversion generate strong and well-defined waves, which makes any cross that is highly correlated to them fertile ground for implementing momentum strategies. Moreover, due to their sheer strength, these economic forces are subject to extremely strong feedback loops that reinforce the groundswell pattern present in "momentum" currencies. How exactly do these feedback loops work? Let's begin with the USD. The U.S. economy has a low beta to global growth, as it is a relatively closed economy where manufacturing represents a small share of both employment and gross value-added. Thus, when global trade accelerates, the U.S. economy does not benefit as much as other large blocs, and the dollar depreciates (Chart I-10). However, a fall in the dollar also helps global trade, as the world economy, particularly EM economies, carry large liabilities in U.S. dollars. Thus, when the dollar falls, the cost of financing global trade decreases, which in turn generates more trade, more investment, and more growth. This is a very powerful feedback loop. Although related, the yen cycle is slightly different, as it is more related to risk aversion and liquidity, given that the yen is the funding currency of choice for carry traders. When global economic activity is strong, carry trades distribute funds from places where liquidity is plentiful like Japan to places that offer high-return at the cost of higher risk (Chart I-11). So long as returns are elevated in the nations sporting high-carry currencies, more liquidity flows into these economies, supporting additional growth and returns. However, this virtuous cycle can become a vicious one when volatility rises, as liquidity can be quickly drained when Japanese investors repatriate home funds from abroad, and carry traders close their positions, selling the high-carry currency and covering their shorts in the funding ones. This not only appreciates the yen relatively to riskier currencies but also worsens the economic outlook and return profile of the carry currencies.9 Chart I-10The U.S. Economy Is Less ##br##Sensitive To Global Growth Chart I-11Japan Is The World's ##br##Provider Of Liquidity These dynamics also explain why momentum strategies tend to be more frequently long-carry currencies than funding ones. Simply put, risk-on cycles tend to be longer than risk-off ones. Chart I-12 shows how momentum strategies tend to overweight funding currencies on the rare occasions when volatility spikes, which makes their spot returns higher than their carry returns during those instances. On the other hand, when volatility is low, momentum strategies buy carry currencies, adding an additional benefit beyond their spot returns. Chart I-12Momentum Overweighs Carry More Often, ##br##Because Greed Is More Common Than Fear Meanwhile, risk-off cycles may be short-lived but they tend to be very intense. Thus, buying the funding currencies as they start generating higher momentum can deliver very quick, very powerful gains. This also helps elucidate the seeming paradox whereby momentum trades in the FX space see an accelerating pace of gains when volatility rises. This makes momentum strategies more agile than carry strategies. Importantly, understanding the link between momentum and the exposure to global factors like global trade as well as risk aversion explains why pairs where both legs of the cross are commodity currencies perform so badly as momentum plays. Much like windswell is generated by local winds, crosses from commodity producers like AUD/NOK or AUD/NZD have a diminished sensitivity to global factors, and instead are mostly driven by relative commodity dynamics or even relative domestic dynamics - forces akin to a localized wind system. With all of the above considered, we conclude the following: In the G10 currency space, momentum strategies will provide high profits on crosses that are driven by powerful systematic forces, and will provide lower returns from crosses driven by more idiosyncratic forces. It thus seems that an investor profiting from momentum in the FX space is not exploiting a market inefficiency, in the strictest academic terms, but rather a fundamental trait of each currency. Finally, we are not suggesting moving-average crossovers are the only mean to generate momentum-based buy and sell signals for currencies. But MA crossovers are a simple yet powerful indicator that provides timing signals in the foreign exchange market. Bottom Line: Currencies that are driven by powerful systematic forces will provide better momentum returns than currencies driven by weak idiosyncratic forces. Global forces like trade dynamics and risk aversion will generate groundswell-like wave patterns that are optimal for momentum strategies. Investment Implications Based on the observations made in this report, we have created a list of five rules of thumb for investors to consider when using momentum in currency markets: When using moving averages to assess momentum, the slower of the two moving averages should have a rolling window between 6-months and 12-months in order to generate superior signals. This gives credence to the commonly used 200-day moving average. Meanwhile, the faster of the moving averages should not exceed 3-months. Currencies that have long, powerful and smooth cycles (groundswell) will tend to provide better returns that currencies that have short, choppy and weak cycles (windswell). Moreover, currencies with a groundswell pattern will tend to be driven by powerful systematic factors, while currencies with a windswell pattern will be driven by weaker idiosyncratic factors. More specifically, investors should try to capture momentum in global risk aversion and global trade. The currencies that best follow these criteria are the JPY and USD crosses. What is momentum telling us now? The financial world continues to be in a risk-on mood. As glee rather than fear has taken hold of investors, momentum continues to point to further downside in the yen (Chart I-13). Chart I-13Plentiful Liquidity Is Supporting Momentum##br## In This Risk-On Environment... Chart I-14...But Global Growth Is##br## Starting To Peak On the other hand, momentum seems to be favoring the dollar right now. Global trade is very strong, but signs are accumulating that it may begin to slow after a spectacular couple of years. The faster moving 1-month/6-month moving-average crossover signals that the dollar is a buy, while the 1-month/200-day is also relatively close (Chart I-14). This means that at the very least, investors should be reducing their short dollar exposures. Juan Manuel Correa, Research Analyst juanc@bcaresearch.com Mathieu Savary, Vice President Foreign Exchange Strategy mathieu@bcaresearch.com 1 Gray, Wesley R., and Jack R. Vogel. "Quantitative Momentum a Practitioner's Guide to Building a Momentum-Based Stock Selection System." Quantitative Momentum a Practitioner's Guide to Building a Momentum-Based Stock Selection System, Wiley, 2016. 2 Jegadeesh, Narasimhan and Sheridan Titman, "Returns to Buying Winners and Selling Losers: Implications for Stock Market Efficiency" Journal of Finance, 48(1): 65-91 (1993) 3 Lukas Menkhoff, Lucio Sarno, Maik Schmeling and Andreas Schrimpf, "Currency Momentum Strategies" (2011) 4 Gray, Wesley R., and Jack R. Vogel. "Quantitative Momentum a Practitioner's Guide to Building a Momentum-Based Stock Selection System." Quantitative Momentum a Practitioner's Guide to Building a Momentum-Based Stock Selection System, Wiley, 2016. 5 Please see Foreign Exchange Strategy Special Report, titled "Carry Trades: More than Pennies And Steamrollers", dated May 6, 2016, available at fes.bcaresearch.com 6 "Wave Energy, Decay and Direction." Surfline.com, 2017, www.surfline.com/surfology/surfology_forecast_index.cfm. 7 Bruno, R. and Raspa, G. (1989). Geostatistical characterization of fractal models of surfaces. In Geostatistics, Vol. 1 (M. Armstrong, ed.) 77-89. Kluwer, Dordrecht. 8 For more insights into application of fractals in finance please see European Investment Strategy Special Report, titled "Fractal Dimension And Market Turning Points", dated July 24, 2014, available at eis.bcaresearch.com 9 For a more detailed discussion of how carry trades generate virtuous and vicious circles in the economies of high-carry currencies, please see Foreign Exchange Strategy Weekly Report, titled "Canaries In The Coal Mine Alert: EM/JPY Carry Trades", dated December 1, 2017, available at fes.bcaresearch.com Appendix A: Methodology Appendix AFormula 1 Table 1Days Used By Each Strategy Appendix B: Momentum By Strategy Chart II-1A1-Month/2-Month Momentum Strategy Chart II-1B1-Month/2-Month Momentum Strategy Chart II-2A1-Month/3-Month Momentum Strategy Chart II-2B1-Month/3-Month Momentum Strategy Chart II-3A1-Month/6-Month Momentum Strategy Chart II-3B1-Month/6-Month Momentum Strategy Chart II-4A1-Month/12-Month Momentum Strategy Chart II-4B1-Month/12-Month Momentum Strategy Chart II-5A1-Month/24-Month Momentum Strategy Chart 5B1-Month/24-Month Momentum Strategy Chart II-6A2-Month/3-Month Momentum Strategy Chart II-6B2-Month/3-Month Momentum Strategy Chart II-7A2-Month/6-Month Momentum Strategy Chart II-7B2-Month/6-Month Momentum Strategy Chart II-8A2-Month/12-Month Momentum Strategy Chart II-8B2-Month/12-Month Momentum Strategy Chart II-9A2-Month/24-Month Momentum Strategy Chart II-9B2-Month/24-Month Momentum Strategy Chart II-10A3-Month/6-Month Momentum Strategy Chart II-10B3-Month/6-Month Momentum Strategy Chart II-11A3-Month/12-Month Momentum Strategy Chart II-11B3-Month/12-Month Momentum Strategy Chart II-12A3-Month/24-Month Momentum Strategy Chart II-12B3-Month/24-Month Momentum Strategy Chart I-13A6-Month/12-Month Momentum Strategy Chart II-13B6-Month/12-Month Momentum Strategy Chart II-14A6-Month/24-Month Momentum Strategy Chart II-14B6-Month/24-Month Momentum Strategy Chart 15A12-Month/24-Month Momentum Strategy Chart II-15B12-Month/24-Month Momentum Strategy Appendix C: Momentum By Currency Legs Chart III-1 Chart III-2 Chart III-3 Chart III-4 Chart III-5 Chart III-6 Chart III-7 Chart III-8 Chart III-9 Chart III-10 Appendix D: Other Data Chart IV-1Volatility Does Not Fully Explain ##br##Momentum Returns Chart IV-2Volatility Does Not Fully Explain ##br## Momentum Returns Table 1Worst Sample 1-Month Return Trades & Forecasts Forecast Summary Core Portfolio Tactical Trades Closed Trades

Overweight Selected Companies Dear Client, This week I am away visiting clients in Australia, so we are sending you this report written by my colleague Oleg Babanov (Emerging Market Equity Sector Strategy). Oleg identifies select companies in Austria as excellent conduits to emerging market growth whilst maintaining high standards of corporate governance. Oleg also has a list of top stocks in Poland, Russia and Turkey. Please contact us if you would like to see those additional picks. Dhaval Joshi Highlights We are recommending an overweight position in select Austrian companies on a long-term (one year-plus) time horizon. Austrian-listed companies traditionally have high exposure to Central and Eastern Europe (CEE) and other Emerging Markets (EM), while offering superior corporate governance standards, which secures a premium to EM peers. At the same time, geographically diversified revenues stemming from developed and emerging markets support less-volatile earnings growth and outperformance over the long-term. Table 1Single-Stock Statistics On Select Austrian Companies* Austrian Companies - EM Focused... Companies in Austria have traditionally been active in both Western Europe, with a main focus in Austria and Germany, as well as in the CEE region, providing investors with a unique access to both kind of markets. Sectors with high exposure include financials, with around 56% in emerging markets, consumer discretionary, with 46%, and materials with 45%. Furthermore, in terms of company count, pretty much every listed company in the materials as well as the real estate sector has exposure to emerging markets (Chart I-1A, Chart I-1B, Chart I-1C, Chart I-1D, Chart I-1E, Chart I-1F). Chart I-1AGeographical Revenue Breakdown Austria: ##br##Consumer Discretionary Chart I-1BGeographical Revenue Breakdown Austria: ##br##Financials Chart I-1CGeographical Revenue Breakdown Austria:##br## IT Chart I-1DGeographical Revenue Breakdown Austria:##br## Materials Chart I-1EGeographical Revenue Breakdown Austria: ##br##Real Estate Chat I-1FGeographical Revenue Breakdown Austria:##br## Utilities ...And With High Corporate Governance Standards The Austrian ATX equity index has significantly outperformed the MSCI EM index on both a long-term (+21% over five years and +27% over three years) and short-term time horizon (+12%) (Chart I-2A & Chart 1-2B). Chart I-2AFive-Year Performance: ##br##Austrian ATX Index Vs. MXEF Index Chart I-2BYTD Performance:##br## Austrian ATX Index Vs. MXEF Index We believe part of this outperformance is warranted by better corporate governance standards of Austrian companies, which score highly compared to their emerging market peers on all metrics, with the exception of environmental disclosure (Chart I-3A, Chart I-3B, Chart I-3C, Chart I-3D).1 Effectively such companies are offering investors access to emerging markets with less corporate risk, and better management and disclosure standards. Chart I-3AESG Disclosure Comparison Chart I-3BSocial Disclosure Comparison Chart I-3CEnvironment Disclosure Comparison Chart I-3DGovernance Disclosure Comparison Based on the findings above, we have created a portfolio of six companies from the consumer discretionary, financials, real estate and industrials sectors, combining exposure to emerging markets with a high ESG score and sound operational and financial performance (Table I-2). Table I-2Select Overweight Companies And ##br##12-Month Beta Vs. MSCI EM Sector Specifics Price performance over the past five years has been strong, with our overweight basket outperforming the broad MSCI EM index by 53% (Chart I-4). Valuations between Austrian banks and companies from other sectors are diverging. While non-bank companies are trading at a 16% premium to EM peers on a P/E basis, Austrian banks are trading at a 14% discount to the EM Banks Index on a price-to-book comparison (Chart I-5). Chart I-4Select Austrian Companies Outperforming##br## MSCI EM Index Chart I-5Valuations Are Diverging##br## Depending On Sector Nevertheless, Austrian companies display better bottom-line growth dynamics, helped by recovering performance on an operational level, translating into slightly higher profitability (Chart I-6A, Chart I-6B, Chart I-6C). Chart I-6AA Recovery In Operating Margins Of ##br##Austrian Companies In Late 2015... Chart I-6B...Has Helped EPS Growth To Outstrip EM ##br##Companies Since The End Of 2015... Chart I-6C...While Profitability Is Close ##br##To The EM Average Chart I-7ACash Flow Generation Is Subdued##br## Among Austrian Companies... Furthermore, despite negative cash flow generation for the select basket, Austrian companies have comfortable debt levels, and are paying out higher dividends than EM companies (Chart I-7A, Chart I-7B, Chart I-7C). Chart I-7B...With Debt Levels Close To The EM Average... Chart I-7C...And Dividend Yields Higher Than EM Peers The Overweight Basket Erste Group Bank (EBS AV) Erste Group Bank (EBS AV) (Chart I-8). Chart I-8Performance Since October 2016: ##br##Erste Group Bank vs. MSCI EM Erste Group Bank (EBS AV) reported better-than-expected third-quarter 2017 financial results on November 3. Net interest income stabilized, ticking up 1% year over year, mainly driven by the integration of Citigroup's consumer banking business in Hungary. Net interest margin was still under pressure, down 4 basis points year over year to 2.39%. Net fee and commission income expanded by 4%, supported by fee income, but was offset by trading income deterioration. Operating expenses grew by 3% year over year due to regulatory and IT project costs. With the decrease in provisions offsetting declining operating results, the bottom line rose by 8% year over year. Asset quality showed improvement, with the NPL ratio shrinking by a significant 111 basis points year over year to 4.3%. The company's tier-1 ratio grew by 2 basis points year over year to 13.4%. The market is estimating a 0.2% EPS CAGR over the next four years. We believe operating expenses should grow at a slower pace in the coming quarters, positively affected by decelerating regulatory and IT project investments. At the same time, we expect net interest income to continue to expand, driven by strong macro performance in the CEE region and countercyclical measures by the corresponding central banks. Raiffeisen Bank (RBI AV) Raiffeisen Bank (RBI AV) (Chart I-9). Chart I-9Performance Since October 2016:##br## Raiffeisen Bank vs. MSCI EM Raiffeisen Bank International (RBI AV) reported remarkable third-quarter 2017 financial results on November 14, solidly beating market expectations. Net interest income advanced by 4% year over year, with net interest margin up 4 basis points to 2.47%. Net fee and commission income climbed by 8% year over year, boosted by the bank's payment transfer business but offset by sluggish trading income as well as a one-off litigation cost in Slovakia. However, pre-provisional profit surged by 35% thanks to disciplined cost management. As a result, net income soared 46% year over year, substantially beating market expectations. Asset quality improvement was another positive. The NPL ratio came in at 6.7%, down 200 basis points year over year, aided by slower NPL formation and write-offs. The tier-1 capital ratio expanded by 100 basis points year over year to 13.4%. The market is estimating an 18% EPS CAGR over the next four years. We welcome the bank's digital transformation strategy in Romania. We believe the new version of the banking platform to be launched in 2018 will better support customers' needs and optimize the bank's transaction business. Andritz AG (ANDR AV) Andritz AG (ANDR AV) (Chart I-10). Chart I-10Performance Since October 2016:##br## Andritz vs. MSCI EM Andritz AG (ANDR AV) reported weak third-quarter 2017 financial results on November 3. Revenue contracted by 8% year over year, weaker across all business segments, especially in pulp and paper (-13%). This was reflected by a shrinkage in overall order intakes, down 9% year over year. In terms of geographic exposure, Andritz continues its sales expansion in Europe (+6%) and China (+25%). EBITDA fell 9% year over year, mainly dragged down by the materials business, despite this being moderately compensated by the separation business segment. EBITDA margin was also disappointing across the board, down 20 basis points year over year to 7.2%, except for the hydro segment (+154%). As a result, the bottom line declined by 20% year over year, missing market expectations. Andritz is trading at a forward P/E of 16.5x, while the market is estimating a 4.7% EPS CAGR over the next four years. Despite lower-than-expected third-quarter earnings, we remain bullish on the company, given its strong track record of business growth in difficult environments. Earlier this month, the company won a contract from SaskPower to refurbish a hydroelectric power station in Canada, with a total contract value of more than US$104 million. CA Immobilien Anlagen (CAI AV) CA Immobilien Anlagen (CAI AV) (Chart I-11). Chart I-11Performance Since October 2016: ##br##CA Immobilien Anlagen vs. MSCI EM CA Immobilien Anlagen AG (CAI AV) reported better-than-expected third-quarter 2017 financial results on November 22. Revenue increased by 5.6% year over year, helped by a 10% increase in rental income, as occupancy rates increased in all three major regions (Germany, Austria and CEE). On the operating side, expenses fell by 5% year over year, while income jumped by 21.4% year over year, pushing operating margin higher to 45.8% from 39.8% for the same period last year. The EBITDA grew 11% year over year. As a result of strong top line performance and a further decline in costs, bottom line expanded by 25% year over year on adjusted basis. CA Immo is trading at a forward P/E of 19.5x, while the market is estimating a 6% EPS CAGR over the next three years. Among some of the highlights of this quarter was the successful reduction in financing cost (-22% compared to the first quarter 2017). The new property additions in Budapest and Prague have already positively contributed to the results, and focus is now shifting to the future pipeline, which is heavily tilted towards Germany (in terms of projects). We expect the positive earnings momentum to continue in 2018. UBM Development (UBS AV) UBM Development (UBS AV) (Chart I-12). Chart I-12Performance Since October 2016:##br## UBM Development vs. MSCI EM UBM Development reported better-than-expected third quarter 2017 financial results on November 28. Quarterly revenue fell by 66.5% year over year, but nine-month output volume stood 18% higher, while operating expenses contracted by 66.7% year over year, helped by lower material costs. Nevertheless, operating income decreased by 70% compared to the same period last year, while operating margin finished 70 basis points lower at 7.9%. Pretax income was helped by a one off gain from affiliates, as a result, net profit climbed 10% compared to last year, and 24% for the first three quarters. On adjusted basis bottom line finished the quarter in negative territory. UBM Development is currently trading at a forward P/E of 10x, while the market is forecasting an EPS CAGR of 6.5% over the next three years. The company came close to reaching its debt reduction target of EUR 550 million, despite EUR 164 million of investments in the first half of the year. Improvements on the balance sheet should provide the company with cheaper financing in 2018. Furthermore, sales are on track, with another EUR 120 million of cash sales secured after the third quarter reporting period, bringing UBM close to its full achieving its full-year guidance. DO & CO (DOC AV) DO & CO (DOC AV) (Chart I-13). Chart I-13Performance Since October 2016: ##br##DO & CO vs. MSCI EM DO & CO (DOC AV) announced first-half year financial results on November 16. Revenues dropped by 10% year over year, primarily dragged down by the international event catering segment. EBITDA contracted accordingly, down 13% year over year. However, EBITDA margin remained stable in the international event catering as well as the restaurants and lounges segments. The bottom line came in shy of expectations, shrinking by 18% year over year. We believe the inclusion of a new customer - Juventus soccer club - will boost the margin further in the second-half of the year. DO & CO is trading at a forward P/E of 17.5x, while the market is estimating a 7.2% EPS CAGR over the next four years. The company is fairly valued compared to its five-year average, but trades at up to a 30% discount to its international peers. We believe that DO & CO should be able to crystalize the effects of a strong 2018 pipeline, with new clients in the airline segment (e.g. Lufthansa, and Air China) and the opening of new locations in Los Angeles and Paris (and expansions in London and New York). On a longer-term perspective, the positive outcome on possible construction of a third airport in Turkey would also boost performance. How To Trade? The EMES team recommends gaining exposure to this theme through a basket of listed equities consisting of six overweight recommendations. The main goal is active alpha generation by excluding laggards and including out-of-benchmark plays, to avoid passive index-hugging via an ETF. Direct: Equity access through the tickers (Bloomberg): Erste Group Bank (EBS AV); Raiffeisen Bank (RBI AV); Andritz AG (ANDR AV); CA Immobilien Anlagen (CAI AV); UBM Development (UBS AV); DO & CO (DOC AV). ETFs: iShares Austria Capped ETF (EWO US) provides exposure to all described companies. Funds: Pioneer Funds Austria (VIENTPF AV); 3 Banken Osterrrech-Fonds (3BKOESI AV); Raiffeisen-Oesterreich-Aktien (OSTAKTT AV). Please note this trade recommendation is long term (1Y+) and based on an overweight trade. We do not see a need for specific market timing for this call (for technical indicators please refer to our website link). For convenience, the performance of both market cap-weighted and equal-weighted equity baskets will be tracked (please see upcoming updates as well as the website link to follow performance). Risks To Our Investment Case On a macro level, we see the main risks stemming from possible asset-purchase tapering by the European Central Bank, which could slow GDP growth in Eastern Europe as well as trigger FX weakness and a slowdown in property markets. Taking into account that exposure to this region is high, such a scenario would most likely cause earnings headwinds for the selected companies, especially in the banking sector. Separately, some of the companies have high exposure to Russia and Turkey. Both countries are prone to geopolitical turbulence, as seen in the past, which in turn can negatively affect economic development and negatively affect companies. Company specific risks include higher rates of projects under construction in the real estate sector, with risks of delays and higher input costs inflating budgets. For Andritz, we see the main risk in the slowdown of capex in the European auto segment (which it seems already happened in the second quarter), and the possible need for additional restructuring in the auto division. We also see some regulatory risk for the banking segment from adverse regulations, such as the bank tax introduction already seen in Hungary, or possible increases in bank taxes in Austria. Oleg Babanov, Associate Vice President obabanov@bcaresearch.co.uk Billy Zicheng Huang, Research Analyst billyh@bcaresearch.com 1 BCA Estimates and Bloomberg Data

Highlights Chart 12017 Bond Returns Treasuries sold off for the third consecutive month in November (Chart 1), and with Congress about to deliver tax cuts and core inflation showing signs of bottoming, the bond bear market is poised to shift into a higher gear. At the moment, the biggest upside risk for bonds is that the Fed continues its hawkish posturing but inflation refuses to comply. That combination would put downward pressure on TIPS breakeven inflation rates and cause the yield curve to flatten further. A flat yield curve increases the odds of a risk-off episode in equities and credit spreads, with a consequent flight into the safety of Treasuries. We do not think the Fed will get it wrong and expect TIPS breakevens to widen alongside rising inflation, easing the flattening pressure on the yield curve. Investors should maintain a below-benchmark duration stance and an overweight allocation to spread product on a 6-12 month investment horizon.   Feature Investment Grade: Overweight Chart 2Investment Grade Market Overview Investment grade corporate bonds underperformed the duration-equivalent Treasury index by 3 basis points in November, dragging year-to-date excess returns down to 285 bps. The average index option-adjusted spread widened 2 bps on the month and now sits at 97 bps. Spreads gapped wider early in the month but then reversed course, ending November not far from where they began. In other words, investment grade corporate bonds remain extremely expensive. We calculate that Baa-rated spreads can only tighten another 39 bps before reaching the most expensive levels since 1989. This represents 3 months of historical average spread tightening. Corporate bonds are essentially a carry trade at this stage of the cycle, but should continue to deliver positive excess returns to Treasuries until inflation pressures mount and the credit cycle comes to an end. We expect the credit cycle will end sometime in 2018.1 Last week's profit data showed that our measure of EBITD increased at an annualized rate of 4% in Q3 (Chart 2), solidly above zero but significantly slower than the 12% registered in Q2. If corporate debt grows by more than 4% in the third quarter, our measure of gross leverage will tick higher (panel 4). As we have shown in prior reports, this would bring the end of the credit cycle closer.2 Quarterly corporate debt growth has averaged just under 6% (annualized) since 2012, so higher leverage in Q3 is likely (Table 3). Table 3ACorporate Sector Relative Valuation And Recommended Allocation* Table 3BCorporate Sector Risk Vs. Reward* High-Yield: Overweight Chart 3High-Yield Market Overview High-Yield underperformed the duration-equivalent Treasury index by 2 basis points in November, dragging year-to-date excess returns down to 578 bps. The index option-adjusted spread widened 6 bps on the month, and currently sits at 349 bps. Excess returns were negative in November for only the fourth month since spreads peaked in February 2016. In a recent Special Report we argued that last month's sell-off would prove fleeting, but also cautioned that excess returns are likely to be low between now and the end of the credit cycle.3 The report flagged five reasons why investors might be nervous about their high-yield allocations. The two most important being that spreads are very tight and the yield curve is very flat. Tight spreads imply that investors should not expect much in the way of further capital gains, insofar as much further spread tightening would lead to historically expensive valuations. In a baseline scenario where spreads remain flat, we forecast excess returns to junk of 246 bps (annualized) (Chart 3). An inverted yield curve signals that investors believe the Fed will be forced to cut rates in the future. This makes it an excellent indicator for the end of the credit cycle. When the yield curve is very flat investors are more inclined to view any negative development as a signal that the cycle is about to turn. This leads to more frequent sell-offs. A period of curve steepening led by higher inflation would mitigate the risk. MBS: Neutral Chart 4MBS Market Overview Mortgage-Backed Securities outperformed the duration-equivalent Treasury index by 4 basis points in November, bringing year-to-date excess returns up to 35 bps. The conventional 30-year zero-volatility MBS spread was flat on the month, as a 2 bps widening in the option-adjusted spread (OAS) was offset by a 2 bps decline in the compensation for prepayment risk (option cost). Agency MBS OAS continue to look reasonably attractive, especially relative to Aaa-rated credit. And with the pace of run-off from the Fed's balance sheet already well telegraphed, there is no obvious catalyst for further OAS widening. In addition, mortgage refinancings are unlikely to spike any time soon. This will ensure that nominal MBS spreads remain capped at a low level (Chart 4). If bond yields rise during the next 6-12 months, as we expect, then higher mortgage rates will be a drag on refinancings. However, as we showed in a recent report, even if rates move lower, the coupon and age distribution of outstanding mortgages has made refi activity much less sensitive to rates than in the past.4 All in all, with OAS more attractive than they have been for several years, Agency MBS are an alluring alternative for investors looking to scale back exposure to corporate bonds. We anticipate shifting some of our recommended spread product allocation out of corporate bonds and into MBS once we are closer to the end of the credit cycle, likely sometime in 2018. Government-Related: Underweight Chart 5Government-Related Market Overview The Government-Related index outperformed the duration-equivalent Treasury index by 28 basis points in November, bringing year-to-date excess returns up to 221 bps. Foreign Agencies and Local Authorities outperformed the Treasury benchmark by 39 bps and 34 bps, respectively. Meanwhile, Sovereign bonds delivered a stellar 93 bps of outperformance. Domestic Agency bonds outperformed by 4 bps, while Supranationals underperformed by 1 bp. We continue to hold a negative view of USD-denominated Sovereign debt. Not only is valuation unattractive compared to similarly-rated U.S. corporate bonds (Chart 5), but historically, periods of sovereign bond outperformance have coincided with falling U.S. rate hike expectations.5 Our Global Fixed Income Strategy team flagged similar concerns in a recent Special Report on the merits of USD-denominated EM debt (both corporate and sovereign).6 The recent moderation in Chinese money and credit growth also heightens the risk of near-term Sovereign underperformance.7 We remain overweight Local Authorities and Foreign Agencies. Year-to-date, those sectors have delivered 256 bps and 402 bps of excess return, respectively, and continue to offer attractive spreads after adjusting for credit rating, duration and spread volatility. Municipal Bonds: Underweight Chart 6Municipal Market Overview Municipal bonds underperformed the duration-equivalent Treasury index by 19 basis points in November (before adjusting for the tax advantage). The average Municipal / Treasury (M/T) yield ratio moved sharply higher in November, with short maturities bearing the brunt of the sell-off. But even after November's weakness, the average M/T yield ratio remains below its average post-crisis level, and long maturities continue to offer a significant yield advantage over short maturities. Both the Senate and House have already passed their own versions of a tax bill, which now just need to be reconciled before new tax legislation is signed into law. Judging from the two versions of the bill, the following will likely occur: The Muni tax exemption will be maintained, the top marginal tax rate will remain close to its current level, the corporate tax rate will be reduced substantially, the state & local income tax deduction will be at least partially eliminated, the tax exemption for private activity bonds might be removed, and advance refunding of municipal bonds will be outlawed or severely restricted. Last month's poor Muni performance was driven by a surge in supply (Chart 6), almost certainly issuers trying to get their advance refundings done before the passage of the final bill. Given that the other provisions in the bill should not have a major impact on yield ratios (any negative impact from lower corporate tax rates should be mitigated by stronger household demand stemming from the removal of the state & local tax deduction), this back-up in yield ratios could present a tactical buying opportunity in Munis once the bill is passed. Stay tuned.   Treasury Curve: Favor 5-Year Bullet Over 2/10 Barbell Chart 7Treasury Yield Curve Overview The Treasury curve bear-flattened in November, as investors significantly bid up the expected pace of Fed rate hikes but did not correspondingly increase their long-dated inflation expectations. The sharp upward adjustment in rate hike expectations means that investors are now positioned for 69 bps of rate hikes during the next 12 months (Chart 7). Similarly, the July 2018 fed funds futures contract is now priced for 52 bps of rate hikes between now and next July. Even if the Fed lifts rates in line with its dots, we would only see 75 bps of rate hikes between now and next July. Since there are strong odds that the Fed will proceed more gradually, this week we close our short July 2018 fed funds futures position for an un-levered profit of 21 bps. In a Special Report published last week, we presented several scenarios for the slope of the 2/10 yield curve based on different combinations of Fed rate hikes and future rate hike expectations.8 We also noted that the positive correlation between long-maturity TIPS breakeven inflation rates and the slope of the nominal 2/10 yield curve has remained intact this cycle. We conclude that the 2/10 slope will steepen modestly in the first half of 2018, before transitioning to flattening once TIPS breakevens level-off at a higher level. With the 2/5/10 butterfly spread now discounting some mild curve flattening (panel 4), investors should remain long the 5-year bullet versus the duration-matched 2/10 barbell.   TIPS: Overweight Chart 8TIPS Market Overview TIPS outperformed the duration-equivalent nominal Treasury index by 15 basis points in November, bringing year-to-date excess returns up to -84 bps. The 10-year TIPS breakeven inflation rate fell 2 bps on the month and, at 1.86%, it remains well below its pre-crisis trading range of 2.4% to 2.5%. As was detailed in last week's Special Report, one of our key views for 2018 is that core inflation will resume its gradual cyclical uptrend, causing long-maturity TIPS breakeven inflation rates to return to their pre-crisis trading range between 2.4% and 2.5%.9 A wide range of indicators, such as our own Pipeline Inflation Indicator and the New York Fed's Underlying Inflation Gauge, already suggest that TIPS breakevens are biased wider (Chart 8). Even more encouragingly, both year-over-year core CPI and core PCE inflation have printed higher in each of the last two months. But even if inflation remains stubbornly low, we think any downside in long-maturity breakevens will prove fleeting. We are quickly approaching an inflection point where if inflation does not rise, the Fed will have to adopt a more dovish policy stance. A sufficiently dovish policy response would limit any downside in breakevens. According to our model, the 10-year TIPS breakeven inflation rate is currently trading in-line with other financial market variables - oil, the trade-weighted dollar and the stock-to-bond total return ratio (panel 2). ABS: Neutral Chart 9ABS Market Overview Asset-Backed Securities outperformed the duration-equivalent Treasury index by 11 basis points in November, bringing year-to-date excess returns up to 92 bps. Aaa-rated ABS outperformed the Treasury benchmark by 10 bps and non-Aaa ABS outperformed by 30 bps. The index option-adjusted spread (OAS) for Aaa-rated ABS tightened 3 bps on the month and, at 31 bps, it remains well below its average pre-crisis trading range. The value proposition in Aaa-rated ABS is not what it once was. At 31 bps, the average index OAS is only 1 bp greater than the average OAS for a conventional 30-year Agency MBS. Agency CMBS are even more attractive, offering an index OAS of 44 bps. Further, the credit cycle is slowly turning against consumer debt. Delinquency rates are rising, albeit off a very low base, but this has caused banks to start tightening lending standards on consumer credit (Chart 9). Tight bank lending standards typically coincide with wider spreads. Importantly, while lending standards are tightening they are not yet very restrictive in absolute terms. In response to a special question from the July 2017 Fed Senior Loan Officer's Survey, banks reported (on net) that lending standards are tighter than the midpoint since 2005 for subprime auto and credit card loans, but are still easier than the midpoint since 2005 for credit card and auto loans to prime borrowers. Non-Agency CMBS: Underweight Chart 10CMBS Market Overview Non-agency Commercial Mortgage-Backed Securities underperformed the duration-equivalent Treasury index by 1 basis point in November, dragging year-to-date excess returns down to 180 bps. The index option-adjusted spread (OAS) for non-agency Aaa-rated CMBS widened 3 bps in November, but is still about one standard deviation below its pre-crisis average (Chart 10). With spreads at such low levels in an environment of tightening commercial real estate (CRE) lending standards and falling CRE loan demand, we continue to view the risk/reward trade-off in non-Agency CMBS as quite unfavorable. Agency CMBS: Overweight Agency CMBS outperformed the duration-equivalent Treasury index by 15 basis points in November, bringing year-to-date excess returns up to 112 bps. The index OAS for Agency CMBS tightened 2 bps on the month but, at 44 bps, the sector continues to offer an attractive spread pick-up relative to other low-risk spread product. The Aaa-rated consumer ABS OAS is only 31 bps, and the OAS on conventional 30-year Agency MBS is a mere 30 bps. Such an attractive spread pick-up in a sector that benefits from Agency backing is surely worth grabbing. Treasury Valuation Chart 11Treasury Fair Value Models The current reading from our 2-factor Treasury model (based on Global PMI and dollar sentiment) pegs fair value for the 10-year Treasury yield at 2.81% (Chart 11). Our 3-factor version of the model (not shown), which also incorporates the Global Economic Policy Uncertainty Index, places fair value at 2.79%. The Global Manufacturing PMI edged higher once more in November, up to 54 from 53.5 in October. It is now at its highest level since March 2011. Meanwhile, sentiment toward the dollar remains significantly less bullish than it was in 2015 and 2016 (bottom panel). A higher PMI reading and less bullish dollar sentiment both lead to a higher fair value in our model. At the country level, both the Eurozone and Japanese PMIs ticked higher in November. The Eurozone PMI broke above 60 for the first time since April 2000. The U.S. and Chinese PMIs both moved modestly lower. For further details on our Treasury models please refer to U.S. Bond Strategy Weekly Report, "The Message From Our Treasury Models", dated October 11, 2016, available at usbs.bcaresearch.com. At the time of publication the 10-year Treasury yield was 2.39%. Ryan Swift, Vice President U.S. Bond Strategy rswift@bcaresearch.com 1 Please see U.S. Bond Strategy Special Report, "2018 Key Views: Implications For U.S. Fixed Income", dated November 28, 2017, available at usbs.bcaresearch.com 2 Please see U.S. Bond Strategy Weekly Report, "Won't Back Down", dated September 26, 2017, available at usbs.bcaresearch.com 3 Please see U.S. Bond Strategy Special Report, "Junk Bond Jitters", dated November 21, 2017, available at usbs.bcaresearch.com 4 Please see U.S. Bond Strategy Weekly Report, "Dollar Watching: Yet Another Update", dated October 10, 2017, available at usbs.bcaresearch.com 5 Please see U.S. Bond Strategy Weekly Report, "Living With The Carry Trade", dated October 17, 2017, available at usbs.bcaresearch.com 6 Please see Global Fixed Income Strategy Special Report, "Examining The Role Of EM Hard Currency Debt In Global Bond Portfolios", dated October 31, 2017, available at gfis.bcaresearch.com 7 Please see China Investment Strategy Special Report, "The Data Lab: Testing The Predictability Of China's Business Cycle", dated November 30, 2017, available at cis.bcaresearch.com 8 Please see U.S. Bond Strategy Special Report, "2018 Key Views: Implications For U.S. Fixed Income", dated November 28, 2017, available at usbs.bcaresearch.com 9 Please see U.S. Bond Strategy Special Report, "2018 Key Views: Implications For U.S. Fixed Income", dated November 28, 2017, available at usbs.bcaresearch.com Fixed Income Sector Performance Recommended Portfolio Specification Corporate Sector Relative Valuation And Recommended Allocation Total Return Comparison: 7-Year Bullet Versus 2-20 Barbell (6-Month Investment Horizon)

Highlights A more bearish backdrop for bonds, led by the U.S.: Faster global growth, with rebounding inflation expectations, will trigger tighter overall global monetary policy. This will be led by Fed rate hikes and, later in 2018, ECB tapering. Global bond yields will rise in response, primarily due to higher inflation expectations. Growth & policy divergences will create cross-market bond investment opportunities: Global growth in 2018 will become less synchronized compared to 2016 & 2017, as will individual country monetary policies. Government bonds in the U.S. and Canada, where rate hikes will happen, will underperform, while bonds in the U.K. and Australia, where rates will likely be held steady, will outperform. The most dovish central banks will be forced to turn less dovish: The ECB and BoJ will both slow the pace of their asset purchases in 2018, in response to strong domestic economies and rising inflation. This will lead to bear-steepening of yield curves in Europe, mostly in the latter half of 2018. The BoJ could raise its target on JGB yields, but only modestly, in response to an overall higher level of global bond yields. The low market volatility backdrop will end through higher bond volatility: Incremental tightening by central banks, in response to faster inflation, will raise the volatility of global interest rates. This will eventually weigh on global growth expectations over the course of 2018, and create a more volatile backdrop for risk assets in the latter half of the year. Feature BCA's annual Outlook report, outlining the main investment themes that will drive global asset markets in 2018, was sent to all clients in late November.1 In this Weekly Report, we drill down into the specific implications of those themes for global bond markets over the next year. In a follow-up report to be published in two weeks, we will discuss how to piece together those implications into an effective fixed income portfolio for 2018. A More Bearish Backdrop For Bonds, Led First By The U.S., Then By Europe The first major takeaway for bond investors from the BCA Outlook is that the current bullish global backdrop of easy monetary policy, solid growth and low inflation is going to change in the coming year. A robust global economy with broadening inflation pressures will force the major central banks to continue incrementally moving away from extraordinarily accommodative monetary policy settings. This will set up an eventual collision between policy and the markets, the latter of which have benefitted so much from the support of the former during the current bull run for risk assets. The changing monetary backdrop will essentially split 2018 into two halves. The current pro-risk backdrop will be maintained in the first half of the year, with continued above-potential global growth and higher realized inflation in the major developed economies at a time when monetary policy is still too accommodative (Chart 1). This will put upward pressure on global bond yields. There is potential for a significant move higher, as real yields now are too low relative to robust global growth and market-based inflation expectations remain well below central bank inflation targets (Chart 2). Chart 1Central Banks Are##BR##Lagging The Cycle Chart 2Both Global Real Yields AND Inflation##BR##Expectations Are Too Low The trend of rising bond yields will be most acute in the U.S., at least in the first half of 2018. The economy is already operating above potential (Chart 3), and this is before factoring in any impact from the tax cut plan currently being finalized in the U.S. Congress. This fiscal stimulus risks overheating the U.S. economy and will likely encourage the Fed to hike interest rates in 2018 by at least as much as it is currently projecting (75bps after the almost certain rate hike later this month). A faster growth trajectory, combined with a rebound in realized inflation after the 2017 slump, will restore investors' belief that U.S. inflation can move back to the Fed's 2% target. The latter can boost the inflation expectations component of the benchmark 10-year U.S. Treasury yield by as much as 60bps next year. The Fed will feel more emboldened to continue delivering rate hikes if inflation expectations are closer to the central bank's target, thus providing an additional boost to Treasury yields. We project that the 10-year Treasury yield can rise up into the 2.9-3% range, well above the current market forwards. The pressure on global bond yields will not only come from the U.S., according to the BCA Outlook. The booming European economy, freed from the years of fiscal austerity after the Euro Debt Crisis and supported by hyper-easy monetary policy from the European Central Bank (ECB), will continue to grow at an above-trend pace in 2018. Japan is enjoying a very powerful cyclical move (by its own modest post-bubble standards) that should continue given very easy monetary policy, robust profit growth and a historically tight labor market. While China is expected to slow on the back of tighter monetary policy and less fiscal stimulus, growth is still expected to be above 6% in 2018. For all of these economies, inflation is expected to rise alongside growth (to varying degrees) given tight labor markets and diminished levels of global spare capacity. Higher oil prices will also boost global inflation and raise the inflation expectations component of global bond yields, given BCA's above-consensus view on oil prices in 2018 (Chart 4). This will also put bear-steepening pressure on many developed market government bond yield curves as inflation expectations increase, particularly with so many countries operating without much economic slack. This argues for being long inflation protection (i.e. inflation-linked bonds vs. nominals or CPI swaps) in 2018, particularly in the U.S., Euro Area and Japan where inflation expectations are well below central bank targets. Chart 3The Global Output Gap Is Closed Chart 4Rising Oil Will Boost Inflation Expectations The BCA Outlook noted that government bond valuations are poor in most countries, with inflation-adjusted (real) yields well below long-run historical averages (Chart 5). We see higher inflation expectations translating directly into higher global bond yields next year, with little room for real yields to decline as an offset. Chart 5Valuation Ranking Of Developed Bond Markets The latter half of 2018 will see increased worries about future U.S. growth after the Fed has delivered a few more rate hikes and U.S. monetary policy potentially shifts into restrictive territory. At the same time, the strength in global growth and, especially, inflation will cast doubts on the need for continued aggressive bond buying by the ECB and the Bank of Japan (BoJ). Unlike last year, the ECB will be unable to wiggle its way out of the politically difficult decision to begin tapering its asset purchases when the latest program ends in September. Even the BoJ may be forced to alter its current "yield curve control" strategy by raising the target on longer-term JGB yields in response to pressures from better domestic growth and rising global bond yields. Thus, the pressures for higher bond yields will rotate away from the U.S. in the latter half of 2018 towards Europe and possibly Japan. Other developed economy central banks, like the Bank of England (BoE), the Bank of Canada (BoC), the Reserve Bank of Australia (RBA) and the Swedish Riksbank will also be faced with decisions on dialing back monetary accommodation in 2018. Although we anticipate that only the BoC and the Riksbank could credibly deliver on monetary tightening given robust growth and, in the case of Sweden, rapidly rising inflation. Which leads to the second major takeaway from the BCA 2018 Outlook ..... Growth & Policy Divergences Will Create Cross-Market Bond Investment Opportunities The BCA Outlook noted that growth expectations for 2018 still look too cautious in many countries. For example, the IMF is forecasting growth in the developed economies will slow from 2.2% to 2% next year, led by decelerations in the Euro Area, Japan, the U.K., Canada and Sweden (Table 1). At the same time, growth in the emerging economies is optimistically projected to accelerate to a 4.9% pace in 2018, even as China's economy cools to 6.5%. Inflation is expected to modestly increase across most of the world, but remain below central bank targets in many countries. So upside growth surprises, particularly in the U.S. and Europe, will continue to be a major investment theme in 2018. Table 1IMF Global Growth & Inflation Forecasts For 2018 Are Too Pessimistic The growth trends, however, may be more divergent than seen in 2017. This leads to potential cross-market bond trading opportunities by playing relative central bank expectations. The OECD's leading economic indicators are accelerating in the U.S., Europe and Japan; potentially peaking at a very high level in Canada; and outright slowing in the U.K. and Australia (Chart 6). When looking at our central bank discounters, which measure the amount of interest rate changes that are currently priced into money market curves, there are some notable discrepancies with the leading indicators (Chart 7). Chart 6More Divergent##BR##Growth... Chart 7...Will Lead To More Divergent##BR##Monetary Policies The market is now pricing in multiple rate hikes in 2018 from the Fed and BoC, modest increases from the BoE and RBA, and no move from the ECB and BoJ. Given the trends in the leading indicators, rate hikes from the Fed and the BoC are likely, while the BoE and RBA will be hard pressed to raise rates at all next year. Thus, U.S. Treasuries and Canadian government bonds are likely to underperform in 2018, while U.K. Gilts and Australian government bonds can be relative outperformers against a backdrop of rising global bond yields. The outlook for the ECB and BoJ, and the implications for bond yields in Europe and Japan, are a special case that represents the third major takeaway from the BCA Outlook ... The Most Dovish Central Banks Will Be Forced To Turn Less Dovish Chart 8ECB Will Fully Taper By The End Of 2018 The BCA Outlook noted that growth in both the Euro Area and Japan has done very well versus the U.S. over the past four years, essentially matching U.S. growth on a per capital basis (i.e. adjusting for faster population growth in the U.S.). In the Euro Area, an end to the painful fiscal austerity after the 2011-13 sovereign debt crisis was a big driver of the economic strength. The BCA Outlook noted that the drag from tighter fiscal policy during the crisis years was equivalent to around 10% of GDP in Greece and Portugal and 7% of GDP in Ireland and Spain. There has been little fiscal tightening in the following three years, which allowed growth in those economies to catch up rapidly. Add in extremely easy financial conditions - low borrowing rates, a cheap euro, and booming European equity and credit markets - and it is no surprise that the Euro Area economy has enjoyed robust growth over the past couple of years. Looking ahead to 2018, the outlook for Euro Area growth still looks very positive. The OECD leading indicator is rising steadily (Chart 8, top panel). The stock of non-performing loans that has clogged up banking systems in the Peripheral European economies is being whittled down - even in Italy where efforts to fix the many problems of its banks are starting to bear fruit (second panel). At the same time, there will be continued upward pressure on Euro Area inflation in 2018. This will mostly come from higher headline inflation related to higher oil prices (third panel), but also from a grind higher in core inflation and wage growth with the Euro Area unemployment rate already at the OECD's estimate of full employment (bottom panel). The Euro Area economy is likely to expand at an above-potential pace over 2% in the first half of 2018, while headline inflation is set to accelerate back towards the ECB's 2% target. This means that the ECB will have to go through another long conversation with the markets about the future of the asset purchase program. Only the outcome will be different than in 2017 as the economic and inflation arguments for continuing with ECB bond buying will be much harder to justify - especially to the hard money core of the ECB led by Germany. Already, the reduced pace of ECB bond buying set for next year, with the monthly purchases cut in half to €30bn/month, implies a significant slowing of Euro Area monetary liquidity (Chart 9). This will put upward pressure on German Bund yields, but with the move being more concentrated in the latter half of the year as the talk of a true ECB taper, perhaps as soon as the end of 2018, builds. Thus, we see Euro Area government debt being an outperformer in the first half of 2018 and an underperformer in the second half. A move in the benchmark 10-year German Bund yield to the 0.8-1.0% range by year-end is a reasonable target. This would reflect the rise in global bond yields that we expect (i.e. the 10-year U.S. Treasury pushing close to 3%), more normalization in Euro Area inflation expectations and the market pulling forward the timing of future ECB rate hikes. Our base case is still that the ECB will not hike policy interest rates until late 2019, however, which will limit the upside for Euro Area yields next year to some degree. In Japan, the BoJ will continue with its current yield curve targeting regime, aiming to cap 10-year JGBs yields through its bond purchases. This is the most effective way to try and boost Japanese inflation through a weaker yen (Chart 10). The BoJ hopes that this will then lead to rising wage growth as workers demand more pay in response to higher realized inflation. Only if there is a pickup in core/wage inflation in Japan can the BoJ have any chance of reaching its 2% inflation target. Chart 9ECB Tapering Will Put European Yields##BR##Under Upward Pressure Chart 10BoJ Will Keep Rates Low To Boost Inflation##BR##Through A Weaker Yen The current BoJ yield target is around 0% on the 10-year JGB. There has been talk of late from some BoJ officials that the yield target could be raised in response to the strengthening Japanese economy. This is likely just talk to placate BoJ board members who were against the yield curve targeting regime in the first place (it was a very close 5-4 vote to implement the new policy framework in September 2016). Yet the BoJ could conceivable raise the yield target by a modest amount in the context of a bigger move higher in global bond yields. According to a simple econometric model of the 10-year JGB yield unveiled by the BoJ in 2016, a 10bp move higher in the 10-year U.S. Treasury yield would raise the fair value of the JGB yield by 2.7bps (Table 2).2 That model currently shows that JGB yields are about 8bps above fair value (around 0%) at the moment. If the 10yr U.S. Treasury yield were to rise to 3%, however, the current level of the JGB yield would be 7bps too low, which would represent the limit of "overvaluation" on this model since 2013 (Chart 11). Under such a scenario, the BoJ raising the yield target to 0.2%, for example, would not be an unusual response - and it would still be consistent with keeping yield differentials wide enough to generate a weaker yen. Table 2Bank Of Japan 10-Year##BR##JGB Yield Model Chart 11BoJ Could Face Pressure To Raise##BR##The Yield Target If UST Yields Rise In any event, the boost to global monetary liquidity from the asset purchases of the ECB and BoJ will fade next year as both central banks will buy a smaller number of bonds than in 2017. Which brings us to the final main takeaway from the 2018 BCA Outlook .... The Low Market Volatility Backdrop Will End Through Higher Bond Volatility The Outlook noted that the conditions underpinning the growth and liquidity driven bull markets for risk assets will start to turn more negative by mid-2018. Tightening financial conditions, especially as the Fed delivers more rate hikes, will eventually start to weigh on global growth expectations. There is even a very real possibility that the Fed will engineer a U.S. recession in 2019 through tighter monetary policy. At the same time, the Fed will be in the process of its balance sheet runoff, while the ECB and BoJ will be buying smaller amounts of bonds. As we have noted many times this year in Global Fixed Income Strategy reports, a slower growth rate of central bank balance sheets will weigh on the performance of risk assets in 2018 (Chart 12). Add in the risk of growth expectations starting to deteriorate in response to tighter monetary policy in the U.S. (and in China, as well), and markets may become increasingly more volatile later next year - starting with more volatile government bond yields (Chart 13). Chart 12Central Bank Liquidity Tailwind To##BR##Risk Assets Will Fade In 2018 Chart 13The Low Market Vol Backdrop Will End##BR##Through Rising Bond Vol A higher volatility backdrop raises the risk for so many global fixed income markets that have benefitted from investors stretching for yield in order to try and achieve adequate returns. In Chart 14, we show the historical range of yields for global government bonds and spread product (using the benchmark indices for each country or sector) dating back to 2000. The gray dots in the chart represent the current yield for each fixed income category and shows how yields are at historic lows in all markets. Chart 14Historical Range Of Bond Yields For Various Fixed Income Markets, 2000-2017 In Chart 15, we present the historic range of volatility-adjusted yields (the same yields from the previous chart, divided by the trailing 12-month realized index total return volatility of each sector). In this chart, the gray dots again represent the current readings. The blue squares show how volatility-adjusted yields would look if the median volatility of each asset class since 2000 was used in the denominator instead of the latest low level of volatility. Chart 15Historical Range Of VOLATILITY-ADJUSTED Bond Yields##BR##For Various Fixed Income Markets, 2000-2017 As can be seen in the chart, many of the sectors that currently have reasonably attractive volatility-adjusted yields, like U.S. Investment Grade, U.S. High-Yield, and hard-currency Emerging Market debt, will look much less compelling if volatility were to increase to more "normal" levels. The market response will be typical in such a higher volatility environment, as yields would increase to compensate for the greater volatility of returns. The current low volatility regime will end when higher inflation and less accommodative central banks raise interest rate volatility and, eventually, future growth uncertainty. We see that inflection point occurring sometime next year, leading to a more challenging environment for global fixed income "carry trades" that are also focused on global growth, like developed market corporate bonds and emerging market debt. In terms of the investment strategy implications, we end this report with a quote taken directly from the 2018 BCA Outlook: "Given our economic and policy views, there is a good chance that we will move to an underweight position in risk assets during the second half of 2018." Robert Robis, Senior Vice President Global Fixed Income Strategy rrobis@bcaresearch.com 1 Please see the December 2017 edition of The Bank Credit Analyst, "Outlook 2018 - Policy And The Markets: On A Collision Course", available at bca.bcaresearch.com and gfis.bcaresearch.com. 2 The model can be found in this report: https://www.boj.or.jp/en/announcements/release_2016/rel160930d.pdf The GFIS Recommended Portfolio Vs. The Custom Benchmark Index Recommendations Duration Regional Allocation Spread Product Tactical Trades Yields & Returns Global Bond Yields Historical Returns

GAA DM Equity Country Allocation Model Update One thing worth noting is that the model now is neutral on Canada, after a long-standing underweight. Canada's valuation ranking had been improving, but the signal was only confirmed this month by the technical ranking. There are no significant changes among other countries, as shown in Table 1. As shown in Table 2 and Charts 1, 2 and 3, the overall model underperformed its benchmark by 68 bps in November as the model was underweight both the U.S. and Japan, which were the only two countries to outperform the MSCI World benchmark in November! The underweight in the U.K. and Australia worked well, but not enough to offset the loss from the overweight of the euro area. Since going live in January 2016, the overall model has outperformed the benchmark by 157 bps, largely from the allocation among the 11 non-U.S. countries, which has outperformed its benchmark by 489 bps. Table 1Model Allocation Vs. Benchmark Weights Table 2Performance (Total Returns In USD) Chart 1GAA DM Model Vs. MSCI World Chart 2GAA U.S. Vs. Non U.S. Model (Level1) Chart 3GAA Non U.S. Model (Level 2) Please see also on the website http://gaa.bcaresearch.com/trades/allocation_performance. For more details on the models, please see the January 29, 2016 Special Report, "Global Equity Allocation: Introducing the Developed Markets Country Allocation Model." http://gaa.bcaresearch.com/articles/view_report/18850. Please note that the overall country and sector recommendations published in our Monthly Portfolio Update and Quarterly Portfolio Outlook use the results of these quantitative models as one input, but do not stick slavishly to them. We believe that models are a useful check, but structural changes and unquantifiable factors need to be considered too in making overall recommendations. GAA Equity Sector Selection Model The GAA Equity Sector Selection Model (Chart 4) is updated as of November 30, 2017. Chart 4Overall Model Performance Table 3Allocations Table 4Performance Since Going Live The model has turned more bullish on global growth and consequently increased the aggregate cyclical overweight. However, within the cyclical basket there has be re-shuffle from resources-based sectors to consumer discretionary and technology stocks. This was driven by improving momentum in these two sectors. Finally, utilities stocks have been downgraded to underweight on the back of the bullish growth outlook. For more details on the model, please see the Special Report "Introducing The GAA Equity Sector Selection Model," July 27, 2016 available at https://gaa.bcaresearch.com. Xiaoli Tang, Associate Vice President xiaoli@bcaresearch.com Aditya Kurian, Research Analyst adityak@bcaresearch.com

Watching The Warning Signals Recommended Allocation Two of the three indicators we have focused on all year as reliable signals of recession (and, therefore, of the timing for reducing exposure to risk assets) have wobbled in the past month. But, for now, we are not too concerned about this, and continue to argue that the current bull market has maybe another year to run, until a possible 2019 recession starts to get priced in. Global growth indicators are showing no signs of slowdown, with the Global Manufacturing PMI at 53.5, and 26 of the 29 markets for which Markit runs its survey returning a PMI above 50 - close to the highest percentage on record (Chart 1). However, the flattening yield curve in the U.S. has raised concerns: the gap between the yield on two-year and 10-year Treasuries has fallen to less than 60 bps (Chart 2). But a flattening yield curve is not unusual when the Fed is tightening policy, and historically the curve has needed to invert before it became a recession signal. Also of concern was a jump in early November in high-yield spreads, which have also been a good lead indicator for recession (Chart 3). The rise was caused by poor earnings from lowly-rated telecoms companies, which triggered a sell-off in junk bond ETFs. But the rise in spreads remains insignificant, and has mostly reversed since. Chart 1Global Growth Looks Fine... Chart 2But Should We Worry About The Yield Curve... Chart 3...And Rising Credit Spreads? BCA's macro view, as laid out in detail in our recent 2018 Outlook,1 is that the strong growth that has been a positive for risk assets this year will slowly become a negative next year as it is increasingly accompanied by rising inflation. Two-thirds of countries globally now have unemployment below the NAIRU (Chart 4). In the U.S., employment has reached a level at which the Philips Curve has historically been "kinky", associated with an acceleration in wage growth (Chart 5). Upside surprises in inflation will mean that the Fed will hike three or four times next year (compared to the market's expectation of only 1½ hikes), 10-year bond yields will rise to above 3%, and the dollar will appreciate. Chart 4Unemployment Is Below Nairu In Most Places Chart 5The 'Kinky' U.S. Philips Curve What are the implications of this scenario for portfolio construction? We continue to recommend an overweight on risk assets on the 12-month time horizon, as we would expect equities to outperform bonds until Fed policy tightens above the neutral level (which is still about five rate hikes away, as long as core PCE inflation picks up to 2%, as we expect - Chart 6). However, the risks to this scenario are rising. The Fed could stubbornly push ahead with rate hikes even if inflation remains subdued. Chinese growth could slow if the authorities misjudge the timing of structural reforms. Our geopolitical strategists argue that, while investors overestimated political risks at the start of 2017, now they are underestimating the risks (North Korea, NAFTA renegotiation, China trade issues, Italian elections).2 With valuations stretched, small shocks could trigger a disproportionate negative market reaction. More risk-averse investors, therefore, might choose to reduce exposure now, at the risk of leaving some money on the table. Equities: If global equities have further upside, as we believe, higher beta markets such as the euro zone (average beta to global equities over the past 20 years: 1.2) and Japan (beta: 0.9) are likely to continue to outperform. Both have central banks that remain accommodative, our models suggest further upside for earnings growth into next year (Chart 7), and valuations are less stretched than in the U.S. While EM equities are also high beta, we think they are likely to lag next year: higher U.S. interest rates, a stronger U.S. dollar, potential slowdown in China, and sluggish domestic demand in most major emerging economies all represent significant headwinds. Chart 6How Long Until Rates Above Neutral? Chart 7Euro and Japan Earnings Have Upside Fixed Income: A combination of higher inflation and a more aggressive Fed is not a positive environment for government bonds. We expect the yield curve to steepen over the next six months, as the market prices in higher inflation and fiscal deficits (after the U.S. tax cut), but to resume flattening mid next year, as the Fed pushes ahead with rates hikes, and worries about the risk of a policy error emerge. For now, we remain underweight duration, and prefer inflation-linked over nominal bonds. For spread product, while valuations are stretched, we see some attractiveness. As long as the global expansion continues, U.S. investment grade bonds should see a carry pickup over Treasuries of around 100 bps, and high-yield bonds one of around 250 bps (adjusting for likely defaults) - even if we don't assume further spread contraction. In a world of continuing low rates, that remains alluring. Currencies will continue to be driven by relative monetary policy. While we see the Fed tightening more than the market expects, the ECB will not raise rates until late 2019, since underlying inflationary pressures in the euro zone are much weaker. This is largely in line with what the futures market is pricing in. Interest rate differentials (and an unwind of the current large speculative long-euro positions) should cause some weakness of the euro versus the dollar. We expect the Bank of Japan to stick to its 0% target for 10-year JGBs, which means that the yen will also weaken, to below 120 to the dollar, if U.S. interest rates rise in line with our forecasts (Chart 8). Emerging market currencies have already fallen by 1.3% since early September as U.S. rates rose, and amid signs of economic weakness in some emerging economies. We expect this to continue. Chart 8Yen Is Driven By U.S. Rates Chart 9China Is What Matter For Metals Commodities: Our energy strategists recently raised their target for Brent and WTI crude to an average over the next two years of $65 and $63 respectively, with risk of upside surprises in the event of geopolitical disruptions (Venezuela, Kurdistan etc.). They see the OPEC agreement being extended possibly to December 2018, and argue that backwardation of the oil curve (futures prices lower than spot) and rising extraction costs will delay the response of shale oil producers to the higher price. The outlook for industrial commodities depends, as always, on China, which now comprises greater demand for base metals than the rest of the world put together (Chart 9). The risk of a slowdown in Chinese infrastructure spending next year makes us wary on metals such as iron ore, and markets such as Australia and Brazil. Garry Evans, Senior Vice President Global Asset Allocation garry@bcaresearch.com 1 Please see The Bank Credit Analyst Special Report, "2018 Outlook - Policy And The Markets On A Collision Course," dated 20 November 2017, available at bca.bcaresearch.com 2 Please see Geopolitical Strategy Weekly Report, "From Overstated To Understated Risks," dated 22 November 2017, available at gps.bcaresearch.com GAA Asset Allocation