Japan

Highlights The structural message for equities: prefer equities over bonds. As long as the global 10-year bond yield remains below 2 percent, the equity market’s rich valuation is underpinned, albeit the long-term return from equities is likely to be a feeble low single-digit. The structural message for bonds: overweight the higher yielding versus the lower yielding quality sovereigns, most notably overweight U.S. T-bonds versus German bunds. 10-year yields cannot rise much – maybe only 50-100 basis points – before the rise destabilises equity and other risk-asset valuations. But 10-year yields that are deeply in negative territory can fall even less. The structural message for currencies: tilt towards lower yielding currencies, with a preference for the yen. Once monetary policy is already ultra-accommodative, a central bank’s ability to devalue its currency becomes more and more constrained. Feature Japanification: Bring It On! I have always been bemused and perplexed by people using ‘Japanification’ as a pejorative for the European economy (Chart of the Week). In the west, the received wisdom is that Japan is a ‘basket case’, a fate to be avoided at all costs. Yet nothing could be further from the truth: Japan is, in many ways, an economic role model to which Europe and the rest of the western world should aspire. Chart of the WeekEmbrace 'Japanification' Over the past twenty years, Japan’s productivity growth has outperformed all the other major economies (Chart I-2). To be clear, this is based on real GDP per head of working age (15-64) population, the cohort of people who generate economic output. Still, some people counter that this definition flatters Japan’s productivity growth by omitting the significant number of over 65s who work, and that a fairer definition should divide by the total population. Yet even on this alternative definition, Japan has been doing just fine, performing better than France and broadly in line with Canada (Chart I-3). Chart I-2Japan Is Not A 'Basket Case' Chart I-3Japan Is Doing Just Fine Japan’s real output per head has improved while consumers have enjoyed genuine price stability (Chart I-4). Meaning zero inflation, and not the ‘fake price stability’ of 2 percent inflation that central banks are trying – and failing – to reach. ‘Japanification’ is a state that Europe should not eschew; it is a state that Europe should espouse. Moreover, contrary to what the Philips Curve would have you believe, the absence of inflation does not mean there is a reserve army of the unemployed. Japan’s unemployment rate, at 2.2 percent, is one of the lowest in the world. As is income inequality (Chart I-5). While life expectancy is one of the highest in the world. Chart I-4Japan Has Enjoyed Genuine ##br##Price Stability... Chart I-5...And The Absence Of Extreme Income Inequality This combination of rising productivity, genuine price stability, absence of extreme income inequality, and rising life expectancy means that, in Japan, living standards have been rising for the many, and not just for the few. In turn this has meant that while populist backlashes have erupted elsewhere in the world, Japan has remained a paragon of political stability. In all of these important regards, ‘Japanification’ is a state that Europe should not eschew; it is a state that Europe should espouse. Countering The Counterarguments Nevertheless, in the interests of a balanced debate, we must address the main counterarguments: First, isn’t Japan’s declining population evidence of a national malaise? No. Japan lacks living space. Its mountainous islands are habitable on only tiny slivers along the coasts, and these are among the most densely populated regions in the world. Therefore, as the journalist and Japan specialist Eamonn Fingleton explains, Japan’s low birth rate is a fundamental national policy that can be traced back to the late 1940s. Japan lacks living space. Shorn of empire, Japan faced a major food security problem. At a stroke, Japanese officials stopped dead in its track a huge baby boom which took hold between 1946 and 1948. Ever afterwards Japan has enjoyed – yes, that is the appropriate word – a low birth rate. Although the program’s rationale is not recognized in the West, it is fully understood in the East and both Singapore and China went on to formulate similar policies. Chart I-6Japan's Rising Public Indebtedness Counterbalanced A Plunge In Private Indebtedness Clearly, a nation whose working population is shrinking will produce less than it otherwise might have, but this doesn’t mean the economy is a basket case. Far from it. On a per head basis, as we have shown, Japan is doing just fine, and the imbalance between workers and retirees will gradually work out as people adjust their retirement ages (just as they will have to in the west). A second counterargument is that Japan’s government indebtedness has skyrocketed to over 200 percent of GDP, the highest among any major economy. But this increase in public debt was needed as a crucial counterbalance to a sharp decline in private indebtedness, and thereby prevent a deep slump (Chart I-6). Japan’s total indebtedness has remained broadly flat for decades. Third, the Nikkei 225, at 21,500 today, is barely at half of its 39,000 peak value in 1989. The simple explanation is that the main determinant of any long-term return is the starting valuation. The 1989 peak bubble valuation was so extreme – a price to sales of 2.2 compared to 0.75 today – that the subsequent dire returns were baked in the cake (Chart I-7). Chart I-7Japan's Bubble Was So Extreme That Subsequent Dire Returns Were Inevitable Fourth, Japanese bond yields have been near-zero or negative for almost two decades, which some commentators claim is a classic sign of an economy in ‘secular stagnation’. But as we have shown, these ultra-low yields have coexisted with a Japanese economy that is doing just fine. More recently, the residents of Switzerland and Sweden will vouch for the same thing – that negative bond yields categorically do not mean that their economies are ‘basket cases’. But have these economies progressed only because they have these ultra-low bond yields? No, the charts in this report show no (inverse) relationship between bond yields and long-term productivity growth. Which begs the question: if ultra-low bond yields are not a sign of an economy stuck in a funk, what are they a sign of? The Real Reason For Ultra-Low Bond Yields Chart I-8Inflation Is Stuck Well Short Of The 2 Percent Target Today, like a stuck record, the ECB will repeat again that inflation remains well short of its 2 percent target (Chart I-8), but that its resolve to reach the target is unwavering. Just as it was at the last meeting… last year… the year before that… and five years before that! Instead of loosening even further, the ECB should be explaining why, in spite of years of negative interest rates and trillions of euros of QE, inflation expectations have barely budged. As the ECB will not provide the explanation, we will. The public’s expected inflation – a fundamental input into economists’ models during the past half-century – is not well defined when an economy has reached price stability, as it has now. Chart I-9Unemployment Rates Are At Multi-Decade Lows Confirming what this publication has previously argued, Professor Jeffrey Frankel of Harvard University explains “most people pay little attention to the inflation rate when price growth is as low as it has been in recent years.” As a result, argues a paper from the NBER, large policy change announcements in the U.S., the U.K., and the euro area seem to have only limited effects on the inflation expectations of households and firms.1 However, as most economists and central banks fear that their credibility is at stake, they remain fixated on the need to reach the 2 percent inflation target. This requires them to double down, triple down, and then quadruple down on extreme accommodation, even though prices are stable, the economy is progressing, and unemployment rates have declined to multi-decade lows (Chart I-9). So in answer to our previous question, ultra-low bond yields are not a sign of an economy stuck in a funk; they are a sign of central banks that are chasing the wrong inflation target, and that are too scared to change the target for the damage it would do to their credibility. What Does This Mean For Stocks, Bonds, And Currencies?   Ultra-low bond yields are coexisting with economies that are doing fine, as we have seen in Japan, Switzerland, and Sweden. But at such low yields, the unattractive asymmetry of limited bond price upside with unlimited downside justifies exponentially higher valuations for equities and other risk-assets. Chart I-1010-Year Bond Yields Can Rise By Only 50-100 Basis Points So the structural message for equities is: as long as the global 10-year bond yield remains below 2 percent, the equity market’s rich valuation is underpinned. And on anything other than a trading horizon, equities are to be preferred over bonds – albeit the long-term return from equities is likely to be a feeble low single-digit. The structural message for bonds is: 10-year yields cannot rise much – maybe only 50-100 basis points – before the rise destabilises equity and other risk-asset valuations, thereby acting as a limiter (Chart I-10). But given that there is a lower bound to policy interest rates, 10-year yields that are deeply in negative territory can fall even less. Hence, the risk-reward dynamic suggests going overweight the higher yielding versus the lower yielding quality sovereigns: most notably, overweight U.S. T-bonds versus German bunds. On a structural horizon, prefer equities over bonds. The structural message for currencies is essentially the opposite to that for bonds: tilt towards lower yielding currencies because in a ‘race to the bottom’, a central bank’s ability to devalue its currency becomes more and more constrained. But which low yielding currency? As Japan has already undergone its ‘Japanification’, we like the yen. Fractal Trading System* With geopolitical risks having ebbed somewhat, a good tactical trade would be to lean against the technically overbought conditions in high-quality government bonds. Hence, this week’s recommended trade is to short the U.S. 10-year T-bond setting a profit target of 1.5 percent with a symmetrical stop-loss. In yield terms, this broadly equates to a target yield of 1.9% and stop-loss at 1.5%. Chart I-11U.S. 10-year T-Bond Price For any investment, excessive trend following and groupthink can reach a natural point of instability, at which point the established trend is highly likely to break down with or without an external catalyst. An early warning sign is the investment’s fractal dimension approaching its natural lower bound. Encouragingly, this trigger has consistently identified countertrend moves of various magnitudes across all asset classes. The post-June 9, 2016 fractal trading model rules are: When the fractal dimension approaches the lower limit after an investment has been in an established trend it is a potential trigger for a liquidity-triggered trend reversal. Therefore, open a countertrend position. The profit target is a one-third reversal of the preceding 13-week move. Apply a symmetrical stop-loss. Close the position at the profit target or stop-loss. Otherwise close the position after 13 weeks. Use the position size multiple to control risk. The position size will be smaller for more risky positions. * For more details please see the European Investment Strategy Special Report “Fractals, Liquidity & A Trading Model,” dated December 11, 2014, available at eis.bcaresearch.com. Footnotes 1 Please see http://conference.nber.org/conf_papers/f117592.pdf and the European Investment Strategy Special Report ‘The Case Against Secular Stagnation’ August 29, 2019 available at eis.bcaresearch.com. Dhaval Joshi, Chief European Investment Strategist dhaval@bcaresearch.com Fractal Trading System Cyclical Recommendations Structural Recommendations Closed Fractal Trades Trades Closed Trades Asset Performance Currency & Bond Equity Sector Country Equity Indicators Bond Yields Chart II-1Indicators To Watch - Bond Yields Chart II-2Indicators To Watch - Bond Yields Chart II-3Indicators To Watch - Bond Yields Chart II-4Indicators To Watch - Bond Yields   Interest Rate Chart II-5Indicators To Watch - Interest Rate Expectations Chart II-6Indicators To Watch - Interest Rate Expectations Content Chart II-7Indicators To Watch - Interest Rate Expectations Chart II-8Indicators To Watch - Interest Rate Expectations    

Special Report

Feature Feature ChartNo 'Secular Stagnation' In Japan! Bond yields have plummeted to all-time lows and inflation has continued to undershoot the 2 percent target which central bankers tell us is ‘price stability’. This configuration has led to renewed fears that the European and global economies are entering a so-called ‘secular stagnation’. We strongly disagree with this line of thinking. Near-zero bond yields and inflation are categorically not portents of a long-term drought in economic progress. Quite the opposite. Chart I-2Japan Has Experienced Near-Zero Inflation For Decades Japan has experienced near-zero bond yields and inflation for decades (Chart I-2). Yet since the late 1990s, the growth in Japan’s real GDP per head has outperformed every other major economy1 (Feature Chart). Granted, the Japanese government has been running persistent deficits, but this is to counterbalance private sector de-levering. Total indebtedness as a share of GDP has not been rising. In the post credit boom era, Japan’s economic progress has come entirely from productivity improvements. The ability to learn, experiment, and innovate boosts the quality and/or quantity of output from a fixed set of inputs. Unlike the unsustainable growth that is fuelled by credit booms and asset bubbles, real growth that comes from productivity improvements marks genuine and sustainable economic progress. In Europe, Switzerland tells a similar tale. Swiss bond yields and inflation have been near zero for decades, but they have not defined a secular stagnation. Real GDP per head and living standards have steadily advanced, even from an already high base. In the post credit boom era, Japan’s economic progress has come entirely from productivity improvements.  But the best counterexample comes from economic history. At the height of the British Empire in 1914, British consumer prices were little different to where they stood at the end of the English Civil War in 1651 – meaning that Britain experienced near-zero inflation and low bond yields for almost three centuries (Chart I-3). Did these define a secular stagnation? No, quite the opposite. For Britain, this was a golden epoch in which it emerged as the world’s preeminent economy. Chart I-3Britain Experienced Near-Zero Inflation For Centuries The Real Reason For Near-Zero Inflation And Bond Yields The fear-mongering about a secular stagnation misses the real reason for today’s sub-2 percent inflation and record low bond yields. Central banks have wrongly defined price stability. Central banks have wrongly defined price stability because they think of it in terms of the economics and mathematics in which they have expertise. Their models tell them that they can nail inflation to one decimal place – two point zero. But price stability has as much to do with biology and psychology. Biologists will tell you that the human brain cannot distinguish inflation rates between -1 and 2 percent, a range we indistinguishably perceive as ‘price stability’. If biology teaches us that we cannot distinguish between -1 and 2 percent inflation, then central banks have a huge problem. It is impossible for a central bank to change our inflation expectations within that range, because the entire range just feels the same to us. Therefore, our behaviour in terms of wage demands and willingness to borrow will also stay unchanged. And if our economic behaviour is unchanged, what is the transmission mechanism to fine tune inflation within the -1 to 2 percent range? Central banks have wrongly defined price stability.  Therefore, price stability is actually like a ‘quantum state’. You’re in the state or you’re out of the state, but once you’re in the state you cannot then fine tune inflation to an arbitrary number like two point zero. In fact, average inflation over, say, five years will gravitate to the mid-point of the price stability state, 0.5 percent, which is a long way below the central bank’s arbitrary target of 2 percent (Chart I-4). This forces the central bank into drastic and prolonged monetary policy easing – which depresses bond yields (Chart I-5). Chart I-4Central Banks Have Wrongly Defined Price Stability... Chart I-5...Forcing Them To Depress Bond Yields Monetary Policies Will Ultimately Converge As structural credit booms have sequentially ended, economies have one by one entered the state of price stability. First it was Japan; then it was Switzerland; more recently it has been the euro area and the United States. It follows that the 5-year annualised inflation rates have also sequentially tumbled to the mid-point of the price stability state, around 0.5 percent. By which point, inflation is so far below the misplaced 2 percent target, that the central bank’s drastic and prolonged monetary policy easing has depressed the 5-year bond yield to near zero. Japan reached this point in the late 1990s, Switzerland in the early 2010s, and the euro area in the late 2010s. Begging the question: why has the 5-year inflation rate in the U.S. not tumbled towards 0.5 percent too?  The answer is that actually, it has. On a like-for-like basis, 5-year inflation rates are way below the 2 percent target in all the major jurisdictions. You see, the Americans measure inflation differently to the Europeans. In the U.S., the consumer price basket includes owner-occupied housing costs at a substantial weighting, while in Europe it is completely excluded. Using the same definition of inflation as in Europe, the U.S. 5-year inflation rate is not at 1.5 percent, it is at a feeble 0.6 percent (Chart I-6). Chart I-6On a Like-For-Like Basis, U.S. 5-Year Inflation Is A Feeble 0.6 Percent Crucially, on a like-for-like basis, 5-year inflation rates are way below the 2 percent target in all the major jurisdictions: the U.S., euro area, and Japan. This leads us to believe that the current chasm in monetary policies is unsustainable. Even including owner-occupied housing in the consumer price basket, as the U.S. does, the long run boost to annual inflation is only about 0.2 percent (Chart I-7). Meaning that it is only a matter of time before U.S. structural inflation and bond yields converge with those in the euro area. Chart I-7Owner-Occupied Housing Boosts Inflation, But In The Long Run By Only 0.2 Percent In the meantime, the chasm between monetary policies has become a major geopolitical risk. This is because it has depressed the euro versus the dollar by at least 10 percent – based on the ECB’s own competitiveness indicators. The exchange rate distortion stemming from polarised monetary policies is the culprit for the euro area’s huge trade surplus with the United States (Chart I-8). On this point, President Trump is spot on to complain that the Fed’s policy stance relative to other central banks is severely handicapping U.S. manufacturers. As the president tries to counter this handicap with tariffs, real or threatened, the Fed is being forced to lean against the risks to growth and inflation. Chart I-8Blame Polarised Monetary Policies For The Euro Area’s Huge Trade Surplus With The U.S. What Does All Of This Mean? One way or another, the dollar will come under structural pressure in the coming years as the current chasm in monetary policies proves to be unjustified. However, in the near term, we prefer to express this not via the euro, but via the yen. The corollary is that U.S bond yields will eventually converge with their European counterparts. But to reiterate, a world with near-zero inflation is categorically not a portent of secular stagnation. It is just the true state of price stability as the human brain perceives it, rather than the over-precise two point zero that central banks have arbitrarily picked. In turn, ultra-low bond yields stem from the monetary policy response to this massive undershoot of true price stability from central bank defined price stability.   All of this raises a fascinating question: if bond yields are lower than is truly required, why hasn’t it created a new inflation? The answer is that it has, but the new inflation is not in the real economy. The reason is that the world has just been through a structural credit boom, remains heavily indebted, and is still unwinding some of the credit excesses. In this world, as Japan has illustrated in recent decades, productivity growth must drive economic progress. Instead, the new inflation is in equity and other risk-asset prices. At ultra-low bond yields the prospect of bond capital gains diminishes versus potential losses, making bonds as risky as equities. This removes the need for an excess return on equities and other risk-assets versus bonds, meaning that the valuation of risk-assets inflates exponentially (Chart I-9). So long as bond yields remain depressed, this new inflation in risk-asset valuations is well justified and supported.2  But be very careful if the global 10-year bond yield rises above 2 percent.3   Dhaval Joshi, Chief European Investment Strategist dhaval@bcaresearch.com Footnotes 1 Based on real GDP per working age (15-64) population, but also broadly true for real GDP per total population. 2 Please see the European Investment Strategy Weekly Report “Risk: The Great Misunderstanding Of Finance”, October 25 2018, available at eis.bcaresearch.com. 3 The global 10-year bond yield is the simple average of 10-year government bond yields in the U.S., euro area (or France as a proxy), and China.

Special Report

Dear Client, Please note that there will be no regular Weekly Report next week, as we take a summer break. Our regular publication will resume September 6th. Best regards, Chester Ntonifor, Vice President Foreign Exchange Strategy Highlights Our PPP models show the DXY index to be overvalued by 10-15%. Within the G10 universe, the cheapest currencies are the Swedish krona, the British pound, the Japanese yen and the Norwegian krone. Look to go short CHF/GBP on valuation grounds. Feature Regular readers of our publication will notice that we tend to adhere to very simple and time-tested ideas. One such is the concept of purchasing power parity (PPP). The beauty comes from its simplicity. If the price of a good in Sweden is rising faster than in South Africa, then the krona should depreciate versus the rand to equalize prices across both borders. Otherwise, the krona becomes incrementally expensive, relative to the rand. In practice, various models have shown PPP to be a very poor tool for managing currencies. One roadblock comes from measurement issues, since consumer price baskets tend to differ in composition from one country to the next. Second, there is less price discovery for services, than there is for tradable goods. For example, it is rather difficult to import a haircut from Mumbai into the U.S., and so arbitraging those prices away tends to be impractical. Tariffs, trade restrictions and transport costs also tend to dampen the explanatory power of PPP models, though those have had diminishing importance over time. In order to get closer to an apples-to-apples comparison across countries, we make two adjustments. First, we divide the consumer price index (CPI) baskets into five major groups. In most cases, this breakdown captures 90% of the national CPI basket: Food, restaurants and hotels Shelter Health, culture and recreation Energy and transportation Household goods The second adjustment is to run two regressions with the exchange rate as the dependent variable. The first regression (call it REG1) uses the relative price ratios of the five groups as independent variables. This allows us to observe the most influential price ratios that help explain variations in the exchange rate. The second regression (call it REG2) uses a weighted average combination of the five groups to form a synthetic relative price ratio. If for example, shelter is 33% in the U.S. CPI basket, but 19% in the Swedish CPI basket, relative shelter prices will represent 26% of the combined price ratio. This allows for a uniform cross-sectional comparison, compared to using the national CPI weights. The results were largely consistent: Both regressions were statistically significant, but more so for REG1. This makes intuitive sense, as the number of variables were higher in the first regression. The sign for household goods was negative for some countries. This could be due to some specter of multicollinearity, if the tradable goods price effect is captured in other categories. There is also the low value-to-weight ratio for many household goods such as refrigerators or air conditioners, which could make currency deviations from PPP persistent. The shelter sign was also negative for some countries, meaning rising shelter prices tended to be associated with an incrementally cheaper currency. This could be due to the Balassa-Samuelson effect. Rising incomes (one key determinant of rising house prices) usually reflect rising productivity levels, which tend to lift the fair value of the exchange rate. The results showed the U.S. dollar as overvalued, especially versus the Swedish krona, British pound and Norwegian krone. Commodity currencies were closer to fair value, and within the safe haven complex, the Japanese yen was more attractive than the Swiss franc. The euro was less undervalued than implied by the overvaluation in the DXY index. As a final note, PPP models are just an additional kit to our currency toolbox, and so should never be used in isolation, but in conjunction with other currency signals. This is just a first iteration in our PPP modelling work, which we intend to improve in the months to come. U.S. Dollar We reverse-engineered the fair value for the DXY index by aggregating the model results from its six constituents. This includes the euro, the Japanese yen, the British pound, the Canadian dollar, the Swedish krona, and the Swiss franc, using the corresponding DXY weights. The message from the synthetic model is clear: the U.S dollar is above its fair value, in line with our fundamental view (Chart 1). Chart 1The Dollar is Slightly Expensive Americans spent 35% of their income in 2018 on goods and 65% on services. Shelter remains the single largest consumption item for American households, which makes up 33% of the consumption basket. However, the relative importance of shelter is dwarfed by much more rampant rent and house price increases in other developed countries. Medical care accounts for 8.7% of the CPI basket, and is the highest in the developed world on a per capita basis. Total spending on health care accounts for almost 20% of nominal GDP. Since the 1980s, the CPI for medical care has risen fivefold, far outpacing many developed countries. This makes the dollar incrementally expensive.  Core CPI edged higher to 2.2% in July, driven by medical care and shelter. While above the Federal Reserve’s 2% target, the risks to inflation remain asymmetric to the downside. That will keep the Fed on a dovish path near-term, which should help close overvaluation in the dollar. Euro We had limited data for the euro area, and so our regression results were less robust. REG1 shows the euro as cheap, while REG2 is more ambiguous (Chart 2). In short, a PPP model for the euro had one of lowest explanatory powers within the G10 universe. Food, restaurants and hotels are the largest consumption item in the euro CPI basket. Looking at the details, food and non-alcoholic beverages account for 14%, alcohol and tobacco make up 4%, and restaurant and hotels account for about 10% (Table). Relative price trends have moved to undermine the fair value of the euro. Chart 2The Euro Is Slightly Cheap Euro Area CPI Weights Shelter’s weight in the euro area CPI basket currently stands at 16.7%, the smallest among G10 countries. Since 2012, relative house and rent prices in the euro area have been decreasing compared with that in the U.S. Rampant rent controls, especially in places like Germany have subdued housing CPI, and tempered the fair value of the euro. This makes sense to the extent that it represents a concomitant rise in the welfare state. It is well-known that the euro area is relatively open and so tradable goods prices are important for the fair value of the euro. Given that the epicenter of trade tensions is between the U.S. and China, this will act to boost the relative attractiveness of European goods, which will be a bullish underpinning for the euro. Inflation expectations have collapsed in the euro area. However, compared to the Federal Reserve, there is little the European Central Bank can do to boost inflation. This is relatively euro bullish. Once global growth eventually picks up, improved competitiveness in the periphery will allow for non-inflationary growth. Japanese Yen The yen benefits from being cheap, as well as being a safe-haven currency (Chart 3). The overarching theme for Japan is a falling (and rapidly aging) population, which means that deficient demand and falling prices are the norm. This makes the yen relatively attractive on a recurring basis. Most of the Heisei era in Japan has been characterized by deflation. Importantly, all categories in Japan have been in a relative price downtrend during this period (Table). Domestically, an aging population (that tends to be a large voting base), prefer falling prices. Meanwhile, the bursting of the asset bubble in the late 80s/early 90s led to a powerful deleveraging wave that undermined prices. Chart 3The Yen Is Quite Cheap Japan CPI Weights The relative prices for most items have been decreasing, but culture and recreation inflation have experienced a meaningful rebound since 2013, largely due to a booming tourism industry in Japan.1 According to tourism statistics, the number of international visitors to Japan reached 31 million in 2018, almost five times the number ten years ago. But as long as the younger generation in Japan continues to save more and consume less, prices will remain under pressure. BoJ Governor Haruhiko Kuroda remains committed to achieving a 2% inflation target, but inflation expectations are falling to historical lows at a time when the BoJ is running out of policy bullets.2  That means inflation will likely lag that of other developed countries, lifting the fair value of the yen. British Pound Both regressions show the pound as undervalued. This supports our view that over the long term, the pound is a categorical buy (Chart 4). The consumption baskets in both the U.K. and the U.S. are roughly similar, which means traditional PPP models do a good job at capturing the true underlying picture of price differentials (Table). For example, OECD PPP models, based on national expenditure, show the pound as 15% undervalued. Chart 4The Pound Is Cheap U.K. CPI Weights Housing is the largest item in the consumption basket, with a total weight close to 30% (including housing electricity and water supply). The shelter consumer price index in the U.K. started to fall relative to the U.S. in 2016, which has lowered the fair-value of the pound (in the Balassa-Samuelson framework). That said, the fall in the pound has been much more deep and violent than suggested by domestic price fundamentals. For example, food restaurants and hotels are 10% cheaper in the U.K. compared to the U.S. over the last half decade. However, rather than appreciating 10%, the pound has plummeted by about 30%. Brexit will continue to dictate the ebb and flow of sterling gyrations, but the reality is that the pound should be higher on a fundamental basis. Meanwhile, a pick up in the global economy will benefit the pound. Going short CHF/GBP on valuation grounds is an attractive bet today. Australian Dollar As a commodity currency, PPP models are less useful for the Australian dollar than terms of trade, or even interest rate differentials. That said, the Aussie dollar is still relatively cheap versus the USD on a PPP basis (Chart 5). The key driver for value in the AUD has been a drop in the currency, relative to what price differentials will dictate. Food, restaurants and hotels comprise 23% of the Australian CPI basket, with the alcohol and tobacco category alone making up 7.4% (Table). Given food price differentials have been stable versus the U.S. in over a decade, Aussie citizens have not been particularly worse off. Chart 5The Aussie Is Slightly Cheap Australia CPI Weights Shelter accounts for almost a quarter percent of the basket. Relative shelter prices in Australia have been rising since the late 1990s, but started to soften in the past few years, on the back of macro prudential measures. Meanwhile, holiday travel and accommodation have a total weight of 6%, of which domestic travel makes up 2.9%, and international travel 3.1%. The overall cost of tourism in Australia has been falling relative to the U.S., boosting the fair value of the Aussie. In the 1980s, inflation in Australia averaged around 8.3% year-on-year. This made the Aussie incrementally expensive, creating grounds for a subsequent 50% devaluation from 1980 to 1986. Inflation targeting was finally introduced and has realigned Aussie prices with the rest of the world. Our bias is that the Aussie will be less driven by price differentials going forward, but more by RBA policy and terms of trade. New Zealand Dollar The New Zealand dollar is at fair value according to both models (Chart 6).  Like the aussie, the kiwi is less driven by price differentials and more by terms of trade. Food and shelter account for the largest share of the consumption basket, and relative prices have not been moving in favor of the kiwi (Table). So, while the kiwi was overvalued earlier this decade, the overvaluation gap has been mostly closed via a higher dollar. Chart 6The Kiwi Is At Fair Value New Zealand CPI Weights Relative shelter prices in New Zealand have been soaring in recent decades compared to the U.S. Higher immigration, foreign purchases and a commodity boom helped. However, in August 2018, the ban on foreign property purchases came into effect, which helped cool down the housing market. Like in Australia, the inflation rate in New Zealand reached 18% year-on-year in the early 1980s, and was subsequently addressed via inflation targeting. This has realigned New Zealand prices somewhat with the rest of the world. Our bias is that going forward, the kiwi will underperform the aussie, mainly because of a negative terms of trade shock. Canadian Dollar The loonie is currently trading below its fair value, according to both of our models (Chart 7).  Shelter remains the largest budget item for Canadian households (Table). The average Canadian household spent C$18,637 on shelter per year, around 29.2% of the total consumption in 2017.3 Interestingly, the shelter consumer price index does not fully capture skyrocketing house prices in Canada over the last decade. Since 2005, Canadian house prices relative to U.S. have doubled, according to OECD. On the contrary, the relative shelter CPI has trended downwards. These crosscurrents have dampened the explanatory power of the exchange rate. Chart 7The Loonie Is Slightly Cheap Canada CPI Weights Canadians are avid users of private transportation. The average spending on transportation accounted for 20% of total consumption, the second-largest expenditure item. Relative prices in this category have been rising, which has lowered the fair value of the exchange rate. Canada stands as the sixth largest energy producer in the world, but due to heavy taxation, Canadian consumers are paying more for gas prices than their U.S. neighbors. That said, terms of trade have been more important than PPP concerns for the loonie. In the near term, we believe energy prices (and the Western Canadian Select price spread) will continue to be important for the loonie. Swiss Franc USD/CHF is trading slightly below fair value, despite structural appreciation in the franc in recent years (Chart 8). The largest consumption item in Switzerland is the food, restaurants and hotels category (Table). The second item is shelter. Social services have a higher weight in the CPI basket, compared to other developed nations. This has been a huge driver of relative prices between Switzerland and the rest of the world, with falling relative prices boosting the fair value of the franc. Chart 8The Swiss Franc Is At Fair Value Switzerland CPI Weights Healthcare notably accounts for 15.5% in the total CPI basket, of which patient services makes up 11.5%. The Swiss healthcare system is a combination of public, subsidized private, and entirely private systems. It is mandatory for a Swiss resident to purchase basic health insurance, which covers a range of treatments. The insured person then pays the insurance premium plus part of the treatment costs. Finally, as a small open economy, tradable goods prices are important for Switzerland. Given high levels of specialization, terms-of-trade in Switzerland are soaring to record highs. This makes the franc a core holding in a currency portfolio. Norwegian Krone The Norwegian krone is undervalued according to both models (Chart 9). Food and shelter account for the largest share of the Norwegian CPI basket (Table). While the share of shelter is lower than in the U.S., other categories share similar weights, allowing traditional PPP models to be adequate for USD/NOK. One difference is that in terms of social services, only 0.2% of the expenditures are allocated to education, since all schools are free in Norway, including universities. Chart 9The Norwegian Krone Is Cheap Norway CPI Weights As a large energy producer, Norwegians pay less for electricity, gas, and other fuels. Norway is also a heavy producer of renewable energy, notably hydropower. This makes the domestic energy basket less susceptible to the ebbs and flows of energy prices. Going forward, the path of energy prices will continue to dictate ebbs and flows in the krone. Meanwhile, long NOK positions also benefit from an attractive valuation starting point.  Swedish Krona The krona is the cheapest currency in our universe by a wide margin (Chart 10). This stems less from fluctuations in relative prices and more from negative rates that have hammered the exchange rate. Like many countries, food and shelter is the largest component of the consumption basket (Table). Transportation is also important. However, an important driver for undervaluation in the currency has been a drop in the relative price of social services. Chart 10The Swedish Krona Is Very Cheap Sweden CPI Weights Sweden experienced very high inflation rates in the 1980s, and since then, has been in a disinflationary regime. More recently, the inflation rate has edged down below the Riksbank’s target, mostly dragged down by recreation, culture, and healthcare. This makes Swedish real rates relatively attractive. We remain positive on the Swedish krona and believe that it will be one of the first to benefit, should global growth pick up.   Chester Ntonifor, Foreign Exchange Strategist chestern@bcaresearch.com   Kelly Zhong, Research Analyst kellyz@bcaresearch.com Footnotes 1 We removed the shelter component in regression 1, since it was distorting results. 2 Please see Foreign Exchange Strategy Weekly Report, titled “Short USD/JPY: Heads I Win, Tails I Don’t Lose Too Much”, dated May 31, 2019, available at fes.bcaresearch.com 3 Please see “Survey of Household Spending, 2017,” Statistic Canada, December 12, 2018. Trades & Forecasts Forecast Summary Core Portfolio Tactical Trades Limit Orders Closed Trades

Highlights Duration: Global manufacturing growth will rebound near the end of this year. Much like in 2016, this will result in higher global bond yields on a 12-month horizon. Investors should keep portfolio duration close to benchmark for now, but be prepared to shift to below-benchmark when our global growth indicators show signs of improvement. Country Allocation: Countries with yield curves furthest away from the effective lower bound also have the most cyclical bond markets. At present, this means that U.S. and Canadian bond markets will perform best if global growth continues to weaken. They will also perform worst in the event of an economic turnaround. Japanese bonds will perform best in a bond bear market, with German debt a close second. Relative Value In Global Government Debt: Changes in the level and shape of global yield curves have altered the relative value opportunities in the global government bond space. We find that the most positive carry (including both yield income and rolldown) in global government bond markets is earned in 30-year German, Japanese and Australian bonds, and in 10-year U.K. and Japanese bonds. Feature Reflexivity Chart 1A Brief Inversion The decline in global bond yields has been unrelenting, and it took on a life of its own last week when the U.S. 2-year/10-year slope briefly inverted (Chart 1). After the inversion, the 30-year U.S. Treasury yield broke below 2% and the 10-year yield broke below 1.50%. The average yield on the 7-10 year Global Treasury Index closed at 0.49% last Thursday, just above its all-time low of 0.48% (Chart 1, bottom panel). There’s an interesting self-fulfilling prophesy that can take hold when the yield curve inverts. Investors interpret the inversion as a signal of weaker economic growth ahead. They then bid up long-dated bond prices causing the curve to invert even more. This sort of circular reasoning can cause bond yields to disconnect from the trends in global economic data, often severely. While recession fears have benefited government bonds, risky assets – equities and corporate bonds – have experienced relatively minor pain. The S&P 500’s recent sell-off pales in comparison to the one seen late last year (Chart 2). Meanwhile, corporate bond spreads remain well below early-2019 peaks. Risky assets have clearly benefited from the drop in bond yields, as markets price-in a future where central banks ease monetary policy in response to weaker economic growth, and where that easing is sufficient to keep equities and credit well supported. Chart 2Low Yields Support Risk Assets I Chart 3Low Yields Support Risk Assets II Further evidence of this dynamic is presented in Chart 3. The chart shows the sensitivity of daily changes in the U.S. 10-year Treasury yield to changes in the S&P 500 for each year since 2010. The sample is split into days when the S&P 500 rose and when it fell. For example, in 2010 the sensitivity on “up days” was 2.6, meaning that on days when the S&P 500 rose, the 10-year yield rose 2.6 basis points for every 1% increase in the S&P 500. Similarly, the sensitivity in 2010 on “down days” was 3.2. This means that the 10-year yield fell 3.2 bps for every 1% drop in the equity index. The main takeaway from Chart 3 is how dramatically the sensitivities have shifted in 2019. The yield sensitivity on “up days” has fallen sharply – down to 0.8. This means that yields barely rise on days when equities move up. Meanwhile, the sensitivity on “down days” has shot higher, to just under 4. This means that yields fall a lot on days when equities sell off. The perception of easier monetary policy has been the main support for risk assets this year.  The logical interpretation of these trends is that the perception of easier monetary policy has been the main support for risk assets this year. Global Growth Needed At present, we are stuck in an environment where aggressively easy monetary policy and low bond yields are the sole supports for risky assets. In turn, falling bond yields are stoking concerns about the economy, leading to even easier monetary policy. Only one thing can bust us out of this pattern, and that’s a resurgence of global manufacturing growth. Unfortunately, there is little evidence that this is taking place (Chart 4). The Global Manufacturing PMI is now down to 49.3, below the 2016 trough of 49.9 (Chart 4, top panel). U.S. Industrial Production growth remains weak, but is showing signs of stabilization above the 2016 trough (Chart 4, panel 2). European Industrial Production, on the other hand, continues to contract (Chart 4, panel 3). The downtrend in our favorite real-time indicator of global manufacturing – the CRB Raw Industrials index – remains unbroken (Chart 4, bottom panel). However, even though evidence of a turnaround in global manufacturing is scant, we expect a rebound near the end of this year, for the following reasons: Global financial conditions have eased this year, the result of aggressive central bank stimulus. Financial conditions are easier now than they were in 2018, and much easier than they were prior to the 2015/16 global growth slowdown (Chart 5, top panel). China has started to ease credit conditions in response to U.S. tariffs and the slowdown in growth. So far, stimulus has been tepid relative to 2015/16 levels, but it should ramp up in the coming months.1 Many large important segments of the global economy remain unaffected by the global manufacturing slowdown. The U.S. consumer continues to spend: Core retail sales are growing at a robust 5% year-over-year rate, and consumer sentiment remains elevated (Chart 5, panels 2 & 3). Even in the Eurozone, the service sector has not experienced the same pain as manufacturing (Chart 5, bottom panel). Fiscal policy will remain a tailwind for economic growth this year and next. Last week, there were even rumors of increased fiscal thrust from Germany if the growth slowdown persists.2 Strong inflation readings only increased market worries that the Fed might not be as accommodative as necessary.  On the whole, we expect that the above 4 factors will lead to a rebound in global manufacturing growth near the end of this year. Much like in 2016, this will result in higher global bond yields on a 12-month horizon, but the global growth indicators shown in Chart 4 will need to rebound first. Chart 4Global Growth Indicators Chart 5Catalysts For Economic Recovery Inflation Puts Pressure On Powell Chart 6Strong Inflation Could Complicate The Fed's Message Strong U.S. inflation prints during the past two months add an interesting wrinkle to the macro landscape. Core U.S. inflation grew at an annualized rate of 3.55% in July, following an annualized rate of 3.59% in June (Chart 6). However, these strong inflation readings only increased market worries that the Fed might not be as accommodative as necessary. This exacerbated the flattening of the yield curve and sent long-dated TIPS breakeven inflation rates lower. Our sense is that the Fed is chiefly concerned with re-anchoring inflation expectations (Chart 6, bottom panel). This probably means that another rate cut is coming in September, and that Chairman Powell will do his best to sound accommodative in his Jackson Hole address on Friday. However, recent strong inflation data could prompt Powell to sound more hawkish than the market would like, causing yield curves to flatten and risky assets to fall. Bottom Line: Global manufacturing growth will rebound near the end of this year. Much like in 2016, this will result in higher global bond yields on a 12-month horizon. Investors should keep portfolio duration close to benchmark for now, but be prepared to shift to below-benchmark when our global growth indicators show signs of improvement. Country Allocation & The Zero Lower Bound Perhaps the most straightforward way to think about country allocation within a portfolio of developed market government bonds is to classify the different markets as either “high beta” or “low beta”. Chart 7 shows the trailing 3-year sensitivity of major countries’ 7-10 year bond yields relative to the global 7-10 year yield.3 The U.S. and Canada have the highest betas, followed by the U.K. and Australia. Germany has a beta close to one, and Japan’s beta is the lowest. Chart 7Global Yield Beta In other words, if global growth falters and global bond yields decline, U.S. and Canadian bond markets should perform best, followed by the U.K. and Australia. German bonds should perform in line with the global index, and Japanese bonds should underperform the global benchmark. What makes this approach to portfolio allocation even better is that the calculation of trailing betas is not really necessary. A very similar ordering of countries – from “high beta” to “low beta” – is achieved by simply ranking the markets from highest yielding to lowest yielding. High yielding countries, like the U.S. and Canada, have the most room to ease monetary policy in response to a negative growth shock. This means that yields in those countries will respond most to global growth fluctuations. On the other hand, the entire Japanese yield curve is already pinned near the effective lower bound. Even in the event of a negative growth shock, there is little scope for easier Japanese monetary policy, and JGB yields will be relatively unaffected. Chart 8High Beta Countries Are Most Sensitive To Economic Growth It’s interesting to note in Chart 7 that while German yields are actually below JGB yields, bunds remain somewhat less defensive than the Japanese market. This is because the German term structure has only recently moved to the effective lower bound, and investors likely still retain some hope that an improvement in global growth could lead to European policy tightening at some point in the future. This belief is largely absent in Japan, where the term structure has been pinned at the lower bound for many years.   Chart 8 provides some further evidence of the split between “high beta” and “low beta” bond markets. It shows that the bond markets with the highest yields are also the most sensitive to trends in global growth, as proxied by the Global Manufacturing PMI. U.S. bond yields are highly correlated with the Global PMI, while Japanese bond yields are hardly correlated at all. It follows that if the slowdown in global growth continues and all nations’ yield curves converge to Japanese levels, then the overall economic sensitivity of global bond yields will decline. Bottom Line: Countries with yield curves furthest away from the effective lower bound also have the most cyclical bond markets. At present, this means that U.S. and Canadian bond markets will perform best if global growth continues to weaken. They will also perform worst in the event of an economic turnaround. Japanese bonds will perform best in a bond bear market, with German debt a close second. Looking For Positive Carry Yield curves have undergone dramatic shifts in recent months, in terms of both level and shape. Not only have curves for the major government bond markets shifted down since the beginning of the year, they also now exhibit varying degrees of a ‘U’ shape (Charts 9A-9F). With that in mind, in this week’s report we look for the best “positive carry” opportunities in global government bond markets. Yield curves for the major government bond markets have shifted down since the beginning of the year, they also now exhibit varying degrees of a ‘U’ shape. We use the term carry to mean the expected return from a given bond assuming an unchanged yield curve. This is essentially the combination of yield income (i.e. coupon return) and the price impact of rolling down (or up) the yield curve. For the purposes of this report, we assume a 12-month investment horizon and incorporate the impact of currency hedging into each security’s yield income. Rolldown ‘U’ shaped yield curves mean that bonds near the base of the ‘U’ currently suffer from negative rolldown, while the rolldown for long maturities is often highly positive. Table 1 shows that rolldown is currently negative for all 2-year bonds, but especially for U.S. and Canadian debt. The U.S. and Canada have the highest policy rates within developed markets, so it’s not surprising that the front-end of their yield curves are also the most steeply inverted. In other words, their yield curves are pricing-in that they have more room to cut rates than other countries. Table 112-Month Rolldown* (%) For A Long Position In Government Bonds In general, rolldown is relatively modest for most 5-year and 7-year maturities. The exceptions being German 5-year debt and Aussie 7-year debt, which benefit from 31 bps and 45 bps of positive rolldown, respectively. As mentioned above, rolldown is currently very positive for long maturity debt. In fact, a 10-year U.K. bond offers a whopping 85 bps of rolldown on a 12-month horizon. Yield Income & Overall Carry As mentioned above, rolldown is only one part of a bond’s carry. The other is the yield an investor earns over the course of the investment horizon – the yield income. Because we assume that investors hedge the currency impact of their bond positions, this yield income also depends on the native currency of the investor. Therefore, we show yield income and overall carry below from the perspective of investors in each of the major currency blocs (USD, EUR, JPY, GBP, CAD, AUD). USD Investors Being the global high yielder, USD investors benefit the most from currency hedging. That is, USD investors earn a lot of additional income on their currency hedges, making non-U.S. bonds look more attractive. Unsurprisingly, carry is most positive at the long-end of yield curves (Tables 2 & 3). Table 2In USD: 12-Month Yield Income* (%) For A Long Position In Government Bonds Table 3In USD: 12-Month Carry (%) For A Long Position In Government Bonds EUR Investors The polar opposite of USD investors, EUR-based investors give up a lot of return through currency hedging. This makes the potential for positive carry much less. In any case, the best positive carry opportunities still lie in German, Japanese and Australian 30-year bonds. U.K. and Japanese 10-year bonds are also attractive (Tables 4 & 5). Table 4In EUR: 12-Month Yield Income* (%) For A Long Position In Government Bonds Table 5In EUR: 12-Month Carry (%) For A Long Position In Government Bonds JPY Investors Yen-based investors currently have more opportunities to earn positive carry than those based in euros. But these opportunities remain confined to long-maturity debt. Once again, the standouts are Japanese, German and Australian 30-year bonds, and also U.K. and Japanese 10-year debt  (Tables 6 & 7). Table 6In JPY: 12-Month Yield Income* (%) For A Long Position In Government Bonds Table 7In JPY: 12-Month Carry (%) For A Long Position In Government Bonds GBP Investors Currency hedges work more in favor of GBP than EUR or JPY. As a result, GBP-based investors see more opportunities to earn positive carry (Tables 8 & 9). Table 8In GBP: 12-Month Yield Income* (%) For A Long Position In Government Bonds Table 9In GBP: 12-Month Carry (%) For A Long Position In Government Bonds CAD Investors As with USD-based investors, CAD-based investors also benefit from currency hedging. All securities continue to offer positive carry when hedged into CAD (Tables 10 & 11). Table 10In CAD: 12-Month Yield Income* (%) For A Long Position In Government Bonds Table 11In CAD: 12-Month Carry (%) For A Long Position In Government Bonds AUD Investors AUD-based investors also see positive carry across the entire global bond space, after factoring-in the impact of currency hedging (Tables 12 & 13). Table 12In AUD: 12-Month Yield Income* (%) For A Long Position In Government Bond Table 13In AUD: 12-Month Carry (%) For A Long Position In Government Bonds Bottom Line: Changes in the level and shape of global yield curves have altered the relative value opportunities in the global government bond space. We find that the most positive carry (including both yield income and rolldown) in global government bond markets is earned in 30-year German, Japanese and Australian bonds, and in 10-year U.K. and Japanese bonds.   Ryan Swift, U.S. Bond Strategist rswift@bcaresearch.com Footnotes 1 Please see U.S. Bond Strategy Weekly Report, “The Trump Interruption”, dated August 13, 2019, available at usbs.bcaresearch.com 2 https://www.bloomberg.com/news/articles/2019-08-16/germany-ready-to-raise-debt-if-recession-hits-spiegel-reports  3 We calculate betas using average yields from the Bloomberg Barclays Global Treasury Master index. Fixed Income Sector Performance Recommended Portfolio Specification