the big picture43b029b3-81cc... · the big picture 2019 outlook: war, inflation, recession? in...

The Big Picture2019 outlook: war, inflation, recession?Quarterly update For professional/qualified/accredited investors only

18 November 2018

Data as of 31/10/18 unless stated otherwise

November 2018 For professional/qualified/accredited investors only 1

The Big Picture 2019 outlook: war, inflation, recession?

In considering the outlook for 2019 we take a step back and analyse the long-term backdrop. This reaffirms an ongoing preference for equity-like assets despite diminished long-term growth potential. Events that typically cause a divergence from that path are: war, inflation and/or recession. Though all are possible during 2019 (we assign a 25% probability to recession), our central scenario remains one of modest global growth and inflation. This leaves us expecting better returns on equity-like assets but we examine alternative scenarios. Model asset allocation In our view:

Equities offer good returns but are volatile and correlated to other assets. We remain slightly Underweight.

Real estate has the potential to produce the best returns. We stay at Maximum.

Corporate high-yield (HY) now looks more interesting. We remain Overweight.

Corporate investment-grade (IG) preferred to government debt. We remain Overweight.

Government debt better than it was but still unattractive. We remain Underweight.

Emerging markets (EM) is still the sovereign space with the best potential. We stay at Maximum.

Cash returns are low but stable and de-correlated. We stay at Maximum.

Commodities have not bottomed. We remain Zero-weighted.

No currency hedges. Assets where we expect the best returns

Japanese equities – estimated 15% 5yr annualised total return in USD

EM real estate – estimated 14% 5yr annualised total return in USD

US IG – estimated 4% 5yr annualised total return in USD

USD cash – estimated 3% 5yr annualised total return in USD and good in a crisis Figure 1 – Projected 5-year returns for global assets and neutral portfolio

Returns are annualised in local currency. Risk-adjusted returns are projected returns divided by historical 5y standard deviations. See appendices for definitions, methodology and disclaimers. There is no guarantee that these views will come to pass. Source: Invesco

-0.6

-0.4

-0.2

0.0

0.2

0.4

0.6

0.8

1.0

1.2

-6%

-4%

-2%

0%

2%

4%

6%

8%

10%

12%

Returns Risk-Adjusted (RHS)


Table of contents Summary and conclusions: war, inflation, recession? ....................................................................................... 3

Model asset allocation ............................................................................................................................................ 5

The long-term outlook ............................................................................................................................................. 6

Medium term issues ............................................................................................................................................... 8

The outlook for 2019: war, inflation, recession? ................................................................................................ 10

A year in politics – hard to imagine war ............................................................................................................... 10

The risk of higher inflation .................................................................................................................................... 12

No recession in 2019 ........................................................................................................................................... 13

We need to talk about Jerome ............................................................................................................................. 17

Farewell Mario (and Mark); welcome Yi .............................................................................................................. 19

Behind the projections -- valuations remain a limiting factor ............................................................................... 20

The US equity conundrum ................................................................................................................................... 21

Projections for 2019 and beyond ......................................................................................................................... 23

Optimisation favours cash and real estate ........................................................................................................... 24

Hope versus reality .............................................................................................................................................. 26

Sectors ................................................................................................................................................................. 27

What if we are wrong? Five scenarios for 2019 ................................................................................................. 28

Model Asset Allocation: adding to UK; reducing EM ........................................................................................ 30

Where do we expect the best returns? ............................................................................................................... 32

Appendices ............................................................................................................................................................ 33

Appendix 1: Consensus economic forecasts ....................................................................................................... 33

Appendix 2: Global valuations vs history ............................................................................................................. 34

Appendix 3: Asset class total returns ................................................................................................................... 35

Appendix 4: Expected returns (%) ....................................................................................................................... 36

Appendix 5: Key assumptions ............................................................................................................................. 37

Appendix 6: Optimised allocations for global assets for different currency bases .............................................. 38

Appendix 7: Methodology for asset allocation, expected returns and optimal portfolios ..................................... 40

Appendix 8: Sector multiple regression model methodology ............................................................................... 41

Appendix 9: Definitions of data and benchmarks ................................................................................................ 42

Important information ........................................................................................................................................... 44

Multi-asset research The Big Picture


Another tough year ahead but we expect equity-like assets to continue outperforming Will long-term equity risk premium continue? Growth potential is diminished but equity-like assets still expected to offer best returns 2019 threats War? Ciao Mario, hello Yi Inflation? Recession? We think not What if we are wrong?

Summary and conclusions: war, inflation, recession? 2018 has been a difficult year and we expect more of the same in 2019. However, we expect neither war, rapid inflation nor recession and therefore believe equity-like assets will stay on their outperforming trend. Among such assets we enter the new year Overweight real estate and high-yield (HY) but slightly Underweight equities (entirely because of the US). We are also Overweight investment-grade credit (IG), which we expect to produce better medium term returns than government debt (Underweight). We remain fully allocated to cash and zero-weighted to commodities (including gold). Over the very long-term our analysis suggests that equities have outperformed other assets by quite a large margin (we suspect the same applies to real estate) and we see no evidence that they are out of line with those long-term trends. Reasons to expect a different outcome could include a belief that long-term growth potential is lower than it was or that a short-term shock could provoke an equity bear market. We do believe that long-term growth potential is diminished (demographics, debt and impotent central banks) and allow for this in our forecasts. Nevertheless, our five-year projections still suggest the best returns will accrue to real estate and equities, with those on fixed income assets broadly close to zero (especially in the eurozone and Japan). We expect negative returns on commodities (except agriculture). See Figure 1. Shortening the time horizon to 2019, our research suggests that equity bear markets are often associated with war, inflation and/or recession. The good news is that we expect neither but the risks are growing. War could be provoked by a worsening of geopolitical tensions and potential flashpoints include the Ukrainian presidential election at end-March (given the tensions with Russia and the potential for Western involvement if Russia intervenes). Though we believe the belligerent attitude of President Trump has doubled the risk of a serious military conflict somewhere in the world, we put the probability at only 20% (versus the normal 10%) ECB President Mario Draghi will be replaced in November and Mark Carney’s successor at the BOE should be known by year-end. But, PBOC Governor Yi Gang could become increasingly important to global markets: can the PBOC offset the negative effect of declining QE activity among developed world central banks? Indeed, we think it probable that core inflation, interest rates and bond yields will rise during 2019. This will reduce fixed income returns (we expect negative returns on many eurozone and Japanese bonds over one and five-year horizons). However, we do not expect this to-spill over into equity markets: the correlation between US equities and bond yields remains largely positive, the tipping point is often a 10-year yield around 5%. Recession could change all of that. The risk is growing by the day and we put the probability at 25%, versus 15% a year ago. However, we are encouraged by US profit and investment spending trends and do not believe that Fed policy is yet restrictive. We suspect that global GDP growth will ease to 3% during 2019 (from 3.6% in the year to mid-2018). If this is correct, we believe it will be enough to keep equity-like assets in outperforming mode. Our US equity bear market indicator is currently 65%, the wrong side of normal (50%) but below the 75%-80% that would scare us (see Figure 28). Of course, we may be wrong. As suggested by Figure 2 we assign a probability of 60% to that central scenario, 25% to recession and 5% to each of boom, benign deflation and stagflation. The growing risk of recession is one reason to listen when our optimiser tells us to be Underweight equities (we implement this by being Underweight US equities – see Figure 3). We also hold the most cash that we allow ourselves but continue to give the maximum allocation to real estate and are Overweight HY (though only in the US). Among less volatile fixed income assets, we prefer IG (Overweight) to government debt (Underweight) and remain zero-weighted in commodities (since they are broadly well above historical norms in real terms).



No big asset allocation changes but adding to UK and slightly reducing EM equities

We make no changes to the broad asset class allocations but are making some adjustments within the various groups. We are adding to UK equities and UK real estate (taking them to Neutral on the belief that a lot of Brexit bad news is in the price). We are also adding to Japanese equities on the back of strong earnings and dividend growth, taking them to the maximum allowed, financed by reducing Japanese real estate to zero. The financing of UK additions is achieved by reducing the extent of the Overweight in EM equities (EPS growth has stalled). Our diversified selection of four assets where we expect good returns during 2019 is: EM real estate, Japanese equities, US IG and USD cash.

Figure 2 – Five scenarios for 2019 and our favoured assets (percentages reflect assigned probabilities)

See appendices for definitions, methodology and disclaimers. Source: Invesco



Model asset allocation* Figure 3 – Model asset allocation (18/11/2018)

*This is a theoretical portfolio and is for illustrative purposes only. It does not represent an actual portfolio and is not a recommendation of any investment or trading strategy. Cash is an equally weighted mix of USD, EUR, GBP and JPY. Currency exposure calculations exclude cash. Arrows show direction of change in allocations. See appendices for definitions, methodology and disclaimers. Source: Invesco

Neutral Policy Range Allocation Position vs Neutral Hedged CurrencyCash 5% 0-10% 10%Cash 2.5% 10%Gold 2.5% 0%

Bonds 45% 10-80% 44%Government 30% 10-50% 20%US 10% 14%Europe ex-UK (Eurozone) 8% 0%UK 2% 2%Japan 8% 0%Emerging Markets 2% 4%Corporate IG 10% 0-20% 16%US Dollar 5% 10%Euro 3% 2%Sterling 1% 2%Japanese Yen 1% 2%Corporate HY 5% 0-10% 8%US Dollar 4% 8%Euro 1% 0%

Equities 45% 20-70% 40%US 25% 8%Europe ex-UK 7% 13%UK 4% ↑ 4%Japan 4% ↑ 8%Emerging Markets 5% ↓ 7%

Real Estate 3% 0-6% 6%US 1% 2%Europe ex-UK 1% 1%UK 0.5% ↑ 1%Japan 0.5% ↓ 0%Emerging Markets 0% 2%

Commodities 2% 0-4% 0%Energy 1% 0%Industrial Metals 0.3% 0%Precious Metals 0.3% 0%Agriculture 0.3% 0%

Total 100% 100%

USD 49% 47%EUR 21% 18%GBP 8% ↑ 10%JPY 14% 11%EM 7% ↓ 14%

Total 100% 100%

Currency Exposure (including effect of hedging)



Equities outperform over the long-term. Can we find reasons why that may not be the case in the future?

Surprisingly, US equities do not seem out of line with long-term trends.

The long-term outlook Investment is a simple process if we don’t try to predict every twist and turn in markets. Over the very long-term equities (stocks) and real estate tend to outperform fixed income assets, while commodities give the same return as bonds but with the same volatility as stocks (see Figure 4). If we were to draw an efficient frontier it would run from cash to stocks, with investment-grade grade credit (IG) the closest of the other assets shown. Not surprisingly, and as we outlined in Asset allocation in pictures (November 2017), optimal portfolios based on this data set would be dominated by stocks, IG and cash (the exact balance depending on one’s appetite for risk). Figure 4 – Risk and reward on US assets 1915-2017 (CPI adjusted, %) *

*Based on calendar year data from 1915 to 2017. Area of bubbles is in proportion to average correlation with other assets. Calculated using: spot price of gold, Global Financial Data (GFD) US Treasury Bill total return index for cash, our own calculation of government bond total returns (Govt) using 10-year treasury yield, GFD US AAA Corporate Bond total return index (IG), Reuters CRB total return index until November 1969 and then the S&P GSCI total return index for commodities (CTY) and Robert Shiller’s US equity index and dividend data for stocks. Indices are deflated by US consumer prices. Past performance is no guarantee of future results. Source: Datastream, Global Financial Data, Reuters CRB, S&P GSCI, Robert Shiller, Invesco The problem with such an analysis is the critical importance of the start and end-dates. Equities may look so good in Figure 4 simply because they are at the end of an unsustainable bull-run. Figure 5 offers some evidence that US equities are not out of line with long-term trends, while the commodities bubble seems to have deflated. Figure 5 – US real total return indices (Sep 1914 = 100, logarithmic scale) *

*30/09/1914 to 31/10/2018. Calculated using: spot price of gold, Global Financial Data (GFD) US Treasury Bill total return index for cash, our own calculation of government bond total returns (Govt) using 10-year treasury yield, GFD US AAA Corporate Bond total return index (IG), Reuters CRB total return index until November 1969 and then the S&P GSCI total return index for commodities (CTY) and Robert Shiller’s US equity index and dividend data for stocks. Indices are deflated by US consumer prices. Past performance is no guarantee of future results. Source: Datastream, Global Financial Data, Reuters CRB, S&P GSCI, Robert Shiller, Invesco



But there are good reasons to suppose long-term growth potential is not what it was

Looking for reasons why our long-term projections could differ from that historical precedent, there are some structural issues that we believe could depress growth and inflation, thereby weakening the prospects for equities and real estate: Global demographics were extremely favourable in the fifty years that followed

WW2, with population growth at levels not seen in the post-1000 era (see Figure 6). As we discussed in Pictures of distress, we believe this supported global economic growth and boosted inflation (as hinted at in Figure 6). Now that population is decelerating in most countries and regions, we believe economic growth and inflation will be lower than during the post-war era. Hence, we assume that dividend and rental growth will be lower than in recent decades but also that central bank interest rates and bond yields will be lower.

Debt is now so high that we suspect it will be a constraining factor, whereas prior to the financial crisis the accumulation of debt supported growth in many parts of the world (for instance, the BIS estimates that US debt/GDP doubled from 125% in 1953 to 251% in 2017). As we have explained on many occasions, we believe that debt is largely a developed world problem, with most emerging economies enjoying much lower debt ratios (China and South Korea being obvious exceptions). See The stabilisation of global debt.

Full normalisation of central bank polices will not have occurred until interest rates are more in line with economic fundamentals and balance sheets have returned to a more normal level in relation to GDP. History suggests this will be a multi-decade process and we see no reason to expect otherwise this time around (with retreat away from normalisation during periods of recession). We believe that such normalisation will represent an ongoing drag on economies and financial markets.

Most major central banks have been running extreme policies since the financial crisis and would appear to have exhausted much of their ability to offset the next recession. The PBOC is a notable exception, with plenty of scope to loosen policy in an economy where there is abundant demand for credit. Among developed world central banks that have used extreme policies, the US Fed is perhaps the only one that has built up some cushion to loosen policy when the next recession arrives, though it is still operating with a bloated balance sheet. The BOE, BOJ, ECB and SNB risk impotence during the next recession. This, added to the high level of government debt in many countries suggests that policy makers will find it difficult to offset a big recession, which should dampen our long-term growth forecasts.

Figure 6 – World population growth and consumer prices in selected countries (annualised rolling 50-year changes, %)*

*Annual data from 1260 to 2100. Historical world population data comes from Global Financial Data. Forecast population data (from 2015) is an interpolation of United Nations forecasts (which is in five year intervals). Consumer price indices supplied by Global Financial Data and the rolling 50-year changes end in 2018 and start in 1260 (UK), 1340 (Sweden) and 1500 (Netherlands). Source: Global Financial Data, United Nations and Invesco



Our long-term growth assumptions are reduced accordingly, as are our inflation and policy rate forecasts We assume there will be recession in the next two to three years and it could be deep. History suggests the US political backdrop will not favour equities over the coming years (after the mid-terms)

Our long-term interest rate and growth assumptions take account of the above factors but we otherwise assume a large degree of convergence to historical norms (credit spreads, default rates, real commodity prices, real exchange rates etc.). Another long-term factor will be the impact of technology such as robotics and automation. We do not believe that unemployment will increase as result of such innovations: after all, there have been many technological revolutions over the centuries (agricultural and industrial) and despite concerns at the time, labour was always re-deployed to other uses. It is not hard to imagine periods during which there could be disruption in the labour market and a short-term rise in unemployment but history suggests that new jobs will eventually appear. The greater question for us is whether these technological advances will bring a leap in total factor productivity. For the moment, any such enhancement is hard to detect but, should it arrive, it could take long-term growth above our assumptions. To be seen.

Medium term issues Shortening the time horizon, we assume there will be a recession in the next two to three years. When that recession comes it is easy to think of factors that could aggravate it: First, the above mentioned high levels of debt could render economies more

vulnerable, with recession provoking abnormal levels of default. Second, as already mentioned, central banks may find themselves unable to play

their usual stabilising role. The same could be said for governments Third, If the common perception about the fragility of the Chinese economy proves

correct, this would have a negative effect on the global economy. Finally, in the absence of fiscal union, we believe eurozone countries are vulnerable

to speculative attack during recession, which could endanger the eurozone and EU. It is hard to know the medium-term stock market implications of the US mid-term elections. Markets do not usually like congressional gridlock (Figure 7 shows that Republican presidents confronted by a mixed Congress are associated with poor stock market returns – note that the first two years of President Trump’s term followed the usual pattern of strong returns for a president with a friendly Congress).

Figure 7 – Annualised US equity gains during US presidencies since 1853 by type of Congress (%)*

* Based on the S&P 500 index since 1957 and comparable indices as derived by Robert Shiller prior to that (see details in Appendix 8). The analysis starts at the beginning of the presidency of Franklin Pierce on 04 March 1853 and ends on 06 November 2018 (the date of the recent mid-term elections). “Friendly Congress” is when both houses are of the same party as the president; “Weak Friendly Congress” is when both houses support the President for most of his full term; “Mixed Congress” is when both parties have an equal stake in Congress; Weak Hostile Congress” is when both houses are predominantly against the president and “Hostile Congress” is when both houses are against the president throughout his term. Past performance is no guarantee of future results. Source: 270towin, Robert Shiller, Global Financial Data, Bloomberg, Datastream, Wikipedia and Invesco



Our five-year return projections suggest real estate and equities offer the best medium-term potential, though cash and credit categories look good on a risk-adjusted basis.

We believe that commodities are expensive in real terms

The initial market reaction after the elections was positive, perhaps in the belief that there may now be less action from President Trump. Time will tell whether this is a good or a bad thing. Taking the above into account, we formulate five-year projections for interest rates, yields and the growth of dividends and rentals. We then generate five-year expected returns for each asset class. Figure 8 shows a summary of the results, with the full regional detail shown in Appendix 4 and assumptions in Appendix 5.

Figure 8 – Projected 5-year total returns for global assets and neutral portfolio*

Returns are annualised total returns and expressed in local currency. Risk-adjusted returns are projected returns divided by historical 5y standard deviations. See appendices for definitions, methodology and disclaimers. There is no guarantee that these views will come to pass. Source: Invesco Overall, the results are as expected – more volatility is associated with better expected returns (except for gold and broad commodities). Hence, the return advantage of real estate and equities is diminished if we take account of volatility (projected return divided by historical standard deviation of returns). The attraction of cash now becomes apparent and investment-grade credit (IG) also receives a leg-up from a risk-adjusted approach (IG has similar volatility to government debt but higher returns, we believe). The low return that we project for commodities is because many appear expensive compared to historical norms (in real terms). We assume that over the medium/long-term they will return to those norms (as they have done over long cycles – see Figure 9, which shows that US oil tends to return to the $20-$60 range in today’s prices).

Figure 9 – Real US oil price since 1870 (US$ CPI adjusted)

Monthly data since January 1870. As of 31 October 2018. Source: Global Financial Data, Datastream, Invesco



What could upset the apple cart in 2019? The presidential election in Ukraine promises to be the most interesting set-piece political event of 2019

The outlook for 2019: war, inflation, recession? Shortening the horizon to 2019, our previous work suggests that three outcomes could provoke an equity bear market, thus disturbing the long-term uptrend: war, a sharp rise in inflation and/or recession (see Figure 10). Figure 10 – What causes equity bear markets (1915-2016)? *

*Items in bold are conditions that were true at the start of the calendar year. Others are conditions met during the year. “-ve EPS momentum” refers to negative earnings per share momentum. “Yld” is yield and “Yld gap” is inverse of Shiller PE minus 10-year yield. Horizontal axis measures proportion of equity bear markets for which stated condition was present. Hit rate is the proportion of times the stated condition occurred and was associated with negative equity returns (or equities being in lowest third of assets). Based on the 27 years from 1915 to 2016 when US equity total returns were negative or when equities ranked among the bottom third of assets. See appendices for definitions. Source: Robert Shiller, Global Financial Data, St. Louis Fed, Datastream and Invesco

A year in politics – hard to imagine war The calendar of elections for 2019 lacks a real blockbuster but there will still be plenty to keep us occupied (see Figure 11 for our selection of the most important). Figure 11: Selected elections during 2019

16/02/2019 Nigeria President, Senate, House of Representatives 31/03/2019 Ukraine President

14/04/2019 Finland Parliament

17/04/2019 Indonesia President, House of Representatives

20/04/2019 Afghanistan President

02/05/2019 UK Local

13/05/2019 Philippines Senate, House of Representatives

18/05/2019 Australia Federal (on or before this date)

26/05/2019 Spain Local and regional

26/05/2019 Belgium Federal

31/05/2019 India General (in April or May)

17/06/2019 Denmark General (on or before this date)

28/07/2019 Japan Upper House (probably in July)

04/08/2019 South Africa General (must be held by this date)

20/10/2019 Switzerland Federal

21/10/2019 Canada Federal

27/10/2019 Argentina President, National Congress

05/11/2019 Israel Knesset (on or before this date) Source: International Foundation for Electoral Systems, Wikipedia, Invesco Perhaps the most interesting on the global geo-political scale will be the presidential election in Ukraine, given the strained relationship with Russia. Opinion polls are



Who will dare to take on President Trump in the 2020 elections? Brexit still has the potential to surprise/disappoint Will Italy consider leaving the EU?

currently suggesting that former Prime-Minister Yulia Tymoshenko will become president, although things could change before the election on 31 March. She has antagonised Russia in the past and her desire that Ukraine join the EU and NATO is unlikely to play well in Moscow. Depending on the reaction of Vladimir Putin, there is the possibility of more tension between Russia and the West. Otherwise, perhaps the most intriguing election-related activity could be the announcement of candidates for the 2020 US presidential race. We know that President Trump intends to seek re-election but there have been no firm declarations of intent from anybody who could mount a serious challenge. Though President Trump’s popularity ratings are low, he defied the opinion polls in 2016 and anybody who confronts him is likely to face a bruising battle. This may deter potential candidates, though if the economy deteriorates over the coming 12 months they may be encouraged to raise their heads above the parapet. India remains the largest democracy and opinion polls over recent months suggest it is heading for a hung-parliament. The NDA alliance of Narendra Modi looks as though it will remain the biggest parliamentary grouping but without the current degree of control. Argentina’s elections had looked as though they would result in the re-election of President Macri but recent economic turmoil (recession, inflation at 40% and currency depreciation) render the outcome uncertain. Likewise, recession in South Africa may render the ruling ANC party less popular but for now opinion polls suggest it will be the largest party in parliament, with close to, if not more than 50% of seats. This suggests that President Ramaphosa will be selected by parliament to continue in his post. We will be keeping an eye on Spain’s local and regional elections, just in case there is a replay of Catalonian dissent. If so, this could add to volatility in European markets. In Europe, there are also three non-election political developments that we shall be following closely: the conclusion of the Brexit process; the battle between Italy and the EU and the durability of Angela Merkel’s Chancellorship. The UK and EU have now published the text of their Brexit agreement and, not surprisingly, the EU got its way on most points. The political reaction in the UK has been furious and it is not clear the deal will be acceptable to the UK parliament (it must also be accepted by all other EU parliaments). If not, finding an alternative in time for the 29 March 2019 exit now looks impossible. Although we think this was the only deal available, the political reaction leads us to believe the probability of both extreme outcomes has increased: a no-deal Brexit or no-exit (via a second referendum). We must also consider the possibility of a leadership challenge to Theresa May and, if parliament rejects the deal, a general election (and the risk of a Labour government). We still believe the most likely outcome is that the proposed deal will be accepted by the UK parliament (it is detested but the alternatives are collectively unpalatable to even bigger numbers). However, we need to balance the negative risk of no-deal and Labour government scenarios against the positive risk of no-exit. We think it is finely balanced. As previously noted, the populist government in Italy has been received with suspicion by financial markets (despite its policy programme bearing a striking resemblance to that of President Trump -- see The good, the bad and the ugly). Fiscal orthodoxy reigns in the markets (and at the credit rating agencies), whereby it is believed that fiscal deficits are bad. However, as originally suggested by Keynes, we believe there are circumstances in which fiscal expansion makes sense, especially when central banks find themselves in liquidity traps (as was the case after the financial crisis) and when fiscal multipliers are high (as they proved to be in the case of Greece in the last 10 years). If any country needs fiscal expansion and deregulation, it is Italy (in our opinion) and if the markets were not anchored on the EU’s fiscal rules, they wouldn’t be worried about Italy’s budget proposal. Unfortunately, the EU (and importantly Germany) does not have the same opinion and this battle could rumble on and become very serious for eurozone markets if Italy is pushed to reconsider its EU membership.



Germany risks becoming a source of instability What will he do next? Tightening labour markets in many countries suggest wage pressures should be building

On the topic of Germany, Angela Merkel’s announcement that she will not seek re-election as Chairperson of the CDU party at the December conference and will not seek any political role when her chancellorship ends in 2021 raises uncertainty in Germany, the EU and the world. Seen as a leader within the EU and a foil to President Trump, the question is who will replace her in not only Germany but also on the European and world stages. Though she wants to stay until 2021, there is a real possibility that she will be gone before then (traditionally, the leader of the CDU takes the role of chancellor when they are in power). Once the new CDU leader is in place, there will be constant pressure for change and, if the coalition crumbles, there could be new elections. Given that Germany has been a force for stability, this could bring unwanted uncertainty (though a change of German government could bring a more flexible EU approach to Italy). Finally, the presidency of Donald Trump brings the risk of new geo-political surprises. Though settlement was reached with Canada and Mexico, the US continues to wage trade disputes with China and the European Union and we wouldn’t be surprised to see Japan targeted. At some stage, even the biggest school bully will suffer if he picks too many fights: there is a risk that the US and global economies will be further weakened. Trade disputes are one thing but history suggests that military conflict would have a more dramatic effect on markets. Our analysis of important conflicts over the last 100 years or so (starting with WW1) suggests that war is often associated with losses on US equities, though the bottom is usually felt within the first 12 months and indices tend to have recovered lost ground within 18 months (on average). It is hard to imagine the outbreak of a military conflict serious enough to shake global markets during 2019 but we reckon that the belligerence of President Trump has doubled the probability to 20% (we identified 10 such conflicts over a 100-year period). Iran and North Korea are likely to continue to be recipients of US attention, especially as the US presidential race gets underway and a stand-off is always possible between the US and Russia and/or China as a result of tensions centred on Crimea, Syria and North Korea, for example.

The risk of higher inflation Figure 10 shows that many of the periods of poor US equity market performance have also been periods when inflation has been on the rise (see horizontal axis). We have been anticipating a rise in global inflation for some time due to tightening labour markets in many parts of the world, especially Japan and the US. Hence, the concern at the apparent acceleration of wages in the US (see Figure 12). Figure 12 – US unemployment and average hourly earnings *

*Monthly data from January 1985 to October 2018. Hourly earnings are for private sector, non-farm, non-supervisory staff. Source: Datastream and Invesco



Which could be a problem for equity markets But so far it has not translated to a pick-up in core inflation We expect at least three Fed hikes between now and end-2019 But, don’t panic – equities tend to continue doing well while the Fed is tightening Recession is what we fear the most And most economies are in the late-expansion phase with more deceleration to come, we think, but not recession

A rise in wage inflation poses a dual problem for equity markets: first, if it is not matched by a rise in selling prices, it will squeeze profit margins and, second, if it does cause a rise in selling prices it will boost core inflation and could provoke higher central bank interest rates and bond yields. It is also possible that we experience a mix of the two (depressed margins and higher rates). Either way, it is a warning signal for equity investors. That sounds scary but there has so far been no noticeable uptick in core inflation (see Figure 13), perhaps because broad monetary growth has remained subdued. Either way, there is no sign yet that core inflation is moving to a sustainably higher path.

Figure 13 – Core CPI by region (% yoy) *

*Monthly data from January 2000 to September 2018. Source: Datastream and Invesco Nevertheless, we doubt the Fed will wait for definitive signs of inflation before continuing with its policy normalisation. We expect a further three or four rate hikes between now and the end of 2019 (one next month and two or three during 2019), taking policy rates to the 3.00%-3.25% range (see Figure 30 for the full set of forecasts). Despite the fact we expect global core inflation to rise gently during 2019 (to 3%), there are reasons for equity investors to remain calm: first, the vertical axis of Figure 10 suggests that rising inflation does not always provoke poor equity market performance (far from it); second, as written here, stocks tend to continue outperforming bonds throughout the entirety of Fed tightening cycles and, third, the correlation between bond yields and equity prices has been largely positive throughout this century (rising yields associated with rising equity prices) and our analysis suggests that a switch to a negative correlation typically happens when US 10-year yields are closer to 5% than 3% (see What’s in a correlation published on 27 February 2018). We do not expect inflation to be a problem for equity markets during 2019.

No recession in 2019 We still believe the most likely cause of an equity bear market will be recession, particularly a US recession. This is the one phase of the economic cycle when our historical analysis suggests it is better to be in defensive assets such as gold, cash and government debt. Figure 14 shows our view of where the world’s 10 largest economies are within their economic cycles (it also shows our interpretation of what history tells us about which assets tend to perform the best at each stage of the cycle). We believe that the large economies are spread across the mid to late-cycle phase but we do not think any of them are in recession and we do not expect them to be during 2019. This said, we do believe there will be some deceleration, in both the US and farther afield, with global growth slowing to more like 3% (from around 3.6% in the year to mid-2018).



Equity-like assets tend to continue doing well at this stage of the cycle The UK is at the most risk of recession, we think And the eurozone has lost momentum

The consensus economic forecasts shown in Appendix 1 suggest an expectation of slight global GDP deceleration in 2019 and 2020 (from 3.8% in 2018 to 3.6% and 3.2%, respectively) with headline inflation relatively stable at just above 3%. After an acceleration in the US economy in 2018, the consensus expects a deceleration there and elsewhere, partially balanced by acceleration in Brazil and India. Not surprisingly, given the information in Figure 14, we suspect the best returns will still be on equity-like assets (equities, real estate and high-yield credit) and this is borne out by the return projections discussed earlier. Implicit in those projections is the assumption that a global recession will occur in the next five years but, importantly, we do not expect it during 2019. This said, recession in 2020 could still damage equity-like assets during 2019 if anticipated by financial markets. We suspect the UK economy is the most susceptible to recession during 2019, largely because of the potential negative effect of the Brexit process (rise in uncertainty and a loss of investment and jobs). The UK economy has underperformed since the Brexit vote (despite the depreciation of sterling) and we suspect that will continue. It is also apparent that the eurozone economy has been among the most disappointing during 2018 (economic surprise indices make it very clear). That may be a spill-over from the UK’s Brexit woes and could also reflect the deceleration of global exports, in part due to President Trump’s actions (the eurozone is a very open economy and therefore vulnerable to any such trade slow-down). More recently, the volatility in Italian markets may have dampened confidence. We were not expecting this weakness in the eurozone and still believe that it is too early in its economic cycle for recession to occur but the signs have not been good. We hope to see better data over the coming months and quarters. If not, we may have to revisit our views on the region.

Figure 14 – The economic and asset class roller coaster

Chart shows our view of the cyclical positioning of the world’s largest economies. The selection of preferred assets is based on our research published in “Asset allocation in pictures” in November 2017. See appendices for definitions, methodology and disclaimers. Source: Invesco


Are there any worrying signs from the US economy? We think the Fed remains accommodative

Figure 15 – Fed policy rates versus nominal GDP growth (%)

Note: based on monthly data from July 1954 to October 2018 (except for GDP data which is quarterly and is up to 2018 Q3). Synthetic policy rate is the actual policy rate adjusted for Fed asset purchases using the “Bernanke rule of thumb” whereby each $150bn-$200bn of asset purchases is equivalent to a 25bp rate cut (the same now applies in reverse). Source: Datastream and Invesco Playing devil’s advocate, what signs should alert us about US recession? Many commentators focus on an inversion of the yield curve but we view that as a symptom, rather than a cause of recession. In more normal times, the yield curve inverts when the Fed has raised rates enough to depress the economy and inflation expectations. The curve may now be inverting because of the lingering impact of Fed QE on long rates, rather than the fact that policy rates are high enough to staunch growth. Yes, new home and auto sales have flattened but they no longer make big contributions to GDP. Based on the evidence in Figure 15 we doubt that Fed policy is yet choking the US economy. At 2.25%, Fed policy rates are well below nominal GDP growth (5.5% in Q3). Even if the Fed raises rates in December (18-19) and nominal GDP growth settles back down to around 4.5%, there would still be a two-percentage point gap between the two, which is hardly indicative of a tight policy stance (in our opinion). Even better, if we share the Fed’s opinion that it is the stock of its assets that counts (rather than the flow), then Fed policy remains super accommodative. Using the Fed’s own rule for calculating a synthetic rate that adjusts policy rates for what has happened to the balance sheet, we derive a rate that remains negative (-2.4% in October). We do not know at what point Fed rates will dampen activity but we suspect it will not be until they are closer to 4%. Of course, recent dollar strength may dampen net exports (which were very weak during 2018 Q3) but, even then, the dollar is hardly at an extreme valuation (see Figure 38).

Figure 16 – Components of US GDP before and after the onset of recessions*

*Note: components of GDP (in constant prices) are shown in the four quarters before and five quarters after that in which recession starts (all components are indexed to 100 in that quarter). The chart shows the average path of each component across the eight GDP recessions since 1950. Source: Datastream and Invesco



Investment and profits play a key role in economic downturns and they are still trending upward

As are profits in the rest of the world

We have previously written on the topic of economic downturns (see The anatomy of a US recession) and Figure 16 shows why we focus on profits -- investment seems to play a key role in the onset of recession in the US. We identified three indicators that are almost always present either before or during a US recession: falling profits, falling investment and a falling equity market. President Trump’s tax cuts could help to prolong the profit and investment cycles and Figure 17 shows that all three of those indicators are still trending upward, which is a good sign (though be aware that the data is quarterly and thus excludes the S&P 500 dip during October and that investment was down a touch in Q3 if we exclude inventories, having been strong in the previous three quarters). Figure 17 – US Profits, investment and the S&P 500 during this economic upswing

*Note: the chart shows the path of investment, profits and the S&P 500 since the US economic cycle bottomed in 2009 Q2 (all components indexed to 100 in that quarter). Data is quarterly. Horizontal axis shows number of quarters since 2009 Q2. Investment is gross private domestic investment (in constant prices). Profits are from the national accounts (profits after taxes with inventory valuation adjustment and capital consumption adjustment) and are in current prices. As of 2018 Q3. Source: Datastream and Invesco. When it comes to the global picture, Figure 18 is relatively comforting. Global profit growth compares well to that of recent years, with impressive growth in Japan but deceleration in the eurozone and EM. Chinese industrial profits have decelerated in recent months (to 4% y-o-y in September), along with the economy, though GDP growth in the 6.50%-7.00% range is not too alarming. It is hard to disentangle the structural and cyclical elements of the Chinese deceleration and we remain relatively optimistic about China’s ability to deal with a slowdown (see More pictures of China). Hence, though we can see why the US and global economies should decelerate during 2019, we do not believe recession will occur. We suspect that can wait until 2020/21. Figure 18 – Earnings per share growth (% yoy)

Note: monthly data from January 2012 to October 2018. Based on Datastream indices. Source: Datastream and Invesco



But the lack of central bank asset purchases may limit portfolio returns Things could get worse in 2019 We expect fixed income assets to feel the most pain We don’t fear Fed rate hikes

We need to talk about Jerome Even without war, inflation or recession during 2019, financial markets will be challenged by the behaviour of central banks. Fed tightening will be the main cause for concern but other developed world central banks will play a role. Figure 19 shows how we expect the aggregate balance sheet of the QE5 to develop through to the end of 2019 (the QE5 is the group of central banks that have used quantitative easing in a meaningful way: The Fed, ECB, BOE, SNB and BOJ). Unfortunately, the combination of Fed balance sheet reductions and tapering by the ECB should bring the aggregate balance sheet into negative growth territory over the coming months (if the BOJ continues buying at the same pace as during the last 12 months, which is about half of its stated plan). 2018 has already proven more difficult for investors than the previous year and Figure 19 suggests 2019 could be even worse. In recent years there has been a reasonable correlation between the growth of that aggregate balance sheet and the returns provided by a global multi-asset portfolio (represented by our Neutral allocation, as shown in Figure 3). Based on that historical relationship, and given our assumptions about the behaviour of these central banks, we fear that portfolio returns could be limited and perhaps negative over the next 12-18 months. Figure 19 – QE5 balance sheet* and asset returns

* Aggregate balance sheet of Fed, ECB, BOE, BOJ and SNB, in USD and rebased to 100 in May 2006. The forecast considers the plans of the Fed and ECB (assuming ECB makes no further asset purchases after December 2018). It is also assumed SNB and BOE make no further purchases and that BOJ continues buying in line with the rate over the last 12 months (around half the rate of its stated plan). The multi-asset benchmark is a fixed weighted index based on our Neutral asset allocation (see Figure 3). From January 2010 to December 2019. Past performance is no guarantee of future results. Source: BOE, Datastream and Invesco Volatility around a rising trend is one thing but volatility around a flat-to-declining trend is quite another -- we are preparing ourselves for many periods of doubt during 2019. Our gut instinct would push us away from equity-like assets (to dampen volatility) but we continue to believe the major pain caused by the retreat from QE should be felt in those assets that were the major beneficiaries of central bank purchases, namely developed world fixed income markets. It is only natural to believe that equity-like assets will suffer along with bonds but we do not think that will be the case. We have written extensively on the topic over the years, including this published in October. Put simply, our research suggests that US equities usually continue to outperform bonds until the Fed stops raising rates, which is usually at about the time that unemployment bottoms (see Figure 20). Also, though the yield curve continues to flatten over that same period, short-maturity bonds tend to continue outperforming long-maturity counterparts until the Fed has finished its work.



Surprising things happen when the Fed tightens We suspect (hope?) the worst is over for EM currencies

Figure 20 – Fed rate hikes, the yield curve and unemployment

Note: monthly data from June 1976 to October 2018. Past performance is no guarantee of future results. Source: Datastream and Invesco. Other conclusions from our work on Fed hiking periods include: Volatility remains overall quite low (it rises once recession comes). The dollar may strengthen but the evidence is patchy (it has not yet recovered all the

losses endured since the Fed started tightening in December 2015). Fed rate hikes increase bond yields elsewhere, even if other central banks ease. There is no systematic evidence that Fed rate hikes weaken EM currencies, in fact

quite the opposite (see Figure 21 which shows a tendency for our EM currency index to strengthen during Fed rate hike periods).

On the topic of emerging markets, we wrote in September (see The Big Picture) that the biggest problem so far during 2018 was currency weakness (underlying asset performance was not too bad in their local currencies). Further, we concluded that currency markets had been quite surgical in their approach – the main currency weakness was related to countries that had already shown signs of financing stress (rising external debt servicing cost to export ratio) and that most other countries did not have such problems (except for Indonesia). Now that our EM FX index has returned to its historical norm, we suspect the worst is over and believe that EM assets will prove rewarding over the medium term (and during 2019). Figure 21 – Real EM FX and commodities

Emerging currency indices are trade weighted averages of national currencies versus US dollar (trade weights are based on total trade flows for each country). There are 18 currencies in the EM basket – those of South Korea, Mexico, India, Russia, Singapore, Malaysia, Brazil, Thailand, Poland, Turkey, Indonesia, Czech Republic, South Africa, Hungary, Nigeria, Chile and Philippines (ordered by size of trade flows). Real adjustments use national CPI indices versus that of the US. “Real Commodity Prices” is the GSCI Spot Index in USD deflated by US CPI. All indices rebased to 100 as of June 1976. As of October 2018. Source: IMF, OECD, Oxford Economics, GSCI, Bloomberg L.P., Datastream and Invesco.



BOJ is the last man standing

Who will take over at the BOE and will rates rise? Will Mario raise rates before he leaves and will we finally get a German as head of the ECB? Our new best friend, Yi Gang

Farewell Mario (and Mark); welcome Yi We believe the BOJ will be the only major central bank still buying assets during 2019 (it is only buying at half its stated rate) but expect very little in terms of interest rate hikes outside of the Fed. Indeed, we fear the US central bank will be alone in having returned policy rates to something like normal by the time of the next recession. The Bank of Japan has yet to discuss formal tapering, let alone talk about rate hikes (we feel the 2% inflation target is unrealistic). We expect no rate increases from Tokyo during 2019 (see the full set of forecasts in Figure 30). The Bank of England wants to raise rates and has done so twice since the Brexit emergency cut to 0.25% in August 2016 (most recently in August). However, it is now constrained by anaemic growth and Brexit uncertainty. As it is hard to know the shape of Brexit, we assume no change in BOE rates during 2019. Also on the agenda next year will be the replacement of Governor Mark Carney, who serves until 31 January 2020. Front-runners include BOE insiders Andrew Bailey (now CEO of the Financial Conduct Authority) and Dave Ramsden (Deputy Governor, Markets and Banking). The choice lies with the Chancellor of the Exchequer. Departing even sooner will be Mario Draghi, whose term as President of the ECB ends on 31 October 2019. The architect of the ECB’s “whatever it takes” approach may want to complete the process by putting ECB rates on the path to normalisation. Hence, we expect one small ECB rate hike before Draghi leaves. Perhaps more important will be the identity of his successor. It was originally thought that Bundesbank President Jens Weidmann would be favourite and that he would adopt a more orthodox Bundesbank-type approach (higher interest rates and smaller balance sheet). However, during the summer it was rumoured that Angela Merkel was more interested in having a German appointed as President of the European Commission. If Weidmann is not chosen, potential candidates include Erkki Liikanen (Finland), Francois Villeroy de Galhau (France) and Philip Lane (Ireland), along with the IMF’s Christine Lagarde (France). Apart from the Fed, the PBOC may be the most influential central bank during 2019 (and beyond). First, it is easing and we suspect it will continue to do so (Figure 22 shows the potential for further reductions in the reserve requirement ratio, which could balance to some extent the reduction of global liquidity due to the ending of developed world QE). Second, if China’s current account continues recent trends, it will soon be in deficit, which could render China’s debt situation more problematic, hence the importance of Chinese bonds attaining full representation in global bond indices. Third, as countries seek non-US dollar settlement systems (to avoid US sanctions etc), the reserve currency status of the CNY can only grow. PBOC Governor Yi Gang may be our new best friend. Figure 22 – China’s monetary policy settings (%)

Monthly data, from January 1980 to October 2018. As of 31 October 2018. RRR is the reserve requirement ratio applied to bank deposits. Source: Peoples Bank of China, Datastream and Invesco



Bond yields are up but are still very low, except in EM

Credit spreads are also close to cyclical tights and default rates are low (neither of which can endure) Equity and real estate yields are more in line with norms

Behind the projections -- valuations remain a limiting factor Bond yields may have risen but remain low (in both nominal and real terms). For instance, Figure 23 shows that Dutch government yields have never been this low in the last 500 years, a period during which deflation and economic disaster were common. Figure 23 – Dutch 10-year bond yields since 1517 (%)

Annual data. 2018 data is as of 31 October 2018. Source: Global Financial Data, Datastream and Invesco Figure 24 shows that yields across fixed income groups remain close to historical lows, with EM debt an obvious exception (see Appendix 2 for regional detail). This naturally limits the scope for fixed income returns, which we expect to be minimal over the next year (especially in Japan and the eurozone), as yields edge higher. We expect slightly more over a 5-year horizon (see the full set of projections in Appendix 4). Credit spreads are close to cyclical tights, so we assume a normalisation over five years, with a step in that direction during 2019. High-yield default rates are also very low now (2.7% in the US and 1.9% in the eurozone) and, again, we assume normalisation over five years (to 4.30% in the US and 3.75% in the eurozone), with a step in that direction next year. The full set of assumptions are shown in Appendix 5. The yields on equity and real estate assets are better than those available on local fixed income alternatives in some places and are relatively close to historical norms. As we believe that starting valuations are an important determinant of medium term investment outcomes, this yield configuration alone would lead us to a preference for equities and real estate, assuming we are not about to enter recession. Figure 24 – Global yields within historical ranges (%)

Start dates are: cash 1/1/01; govt bonds 31/12/85; corp bonds 31/12/96; corp HY 31/12/97; equities 1/1/73; REITs 18/2/05. See appendices for definitions, methodology and disclaimers. As of 31 October 2018. Source: Datastream and Invesco



US equities seem very expensive to us

This could limit returns over the medium-term But is probably not enough to provoke an immediate bear market

The US equity conundrum The only valuation concern that we have among equity markets is the US. Figure 25 shows that it is the only region with a cyclically-adjusted price-earnings ratio (CAPE) above its historical norm. The US CAPE is also well above that of other regions (29.6 versus 13.1 in the emerging markets, for example). Figure 25 – Historical ranges for CAPEs

Note: CAPE = Cyclically Adjusted Price/Earnings and uses a 10-year moving average of earnings. From 1983 (except for EM from 2005). As of 31 October 2018. Source: Datastream and Invesco The same conclusion applies is we use the Shiller PE (a form of CAPE constructed by academic Robert Shiller). Figure 26 shows the extent of the problem – the Shiller PE is now around 32, a level historically associated with low or negative returns over the following ten years. It is commonly supposed that the Shiller PE will decline dramatically once the financial crisis period drops out of the 10-year moving average of earnings. Our analysis, published in the last Big Picture document, suggests the Shiller PE will fall into 26-28 range by the end of 2021, depending upon what is assumed about earnings and price trends in the meantime. This is still well above the historical average of 16.9. Though we believe that rich valuations limit the potential for long term returns, we also believe they are neither necessary nor sufficient to provoke an immediate bear market (see Figure 10). We find that US equity bear markets are associated with a range of factors and that elevated valuations are not that high on the list. Our assessment of the risks is shown in Figure 27 (the left-hand portion shows our subjective probabilities for 2019, while the right-hand part of the chart shows objective market based measures). Figure 26 – S&P 500 Shiller PE and future returns (%)

Monthly data from January 1881 to October 2018. Past performance is no guide to future returns. See appendices for definitions and disclaimers. Source: Robert Shiller and Invesco



We put the risk of a 2019 US recession at 25% Our US equity bear market indicator is currently at 65%, above the 50% norm but not alarming (in our view)

Figure 27 – Probability of factors associated with US equity bear markets

Lower is better in all cases. Left hand group shows factors often associated with bear markets and our assessment of their probability in the next year. Right hand group shows measurable factors associated with bear markets expressed as cumulative probabilities, assuming a normal distribution and using mean and standard deviation of post-1881 history (post-1914 for yield curve). Earnings yield gap is the inverse of the Shiller PE minus the 10-year Treasury yield. EPS momentum is 3m/3m change. As of 31 October 2018. There is no guarantee that these views will come to pass. Source: Global Financial Data, Robert Shiller, Datastream and Invesco (see Appendix 8 for definitions and methodology). As already mentioned, we view the risk of war as more elevated than normal because of President Trump’s belligerence (normally we would put it at 10%) but we believe inflation and rising bond yields are more probable, though not problematic. Recession would pose more of a risk but we put the probability during 2019 at around 25% (though by the end of the year, markets may sniff a 2020 recession). Figure 28 shows our US equity bear market indicator (a simple average of the four indicators on the right-hand side of Figure 27). As can be seen from Figure 27, the Shiller PE and the related earnings yield gap are higher than the 50% historical average (the Shiller PE is at the 99th percentile of its historical range). The yield curve is flatter than usual, hence the reading above 50%, though hardly extreme. Finally, EPS momentum is strong. Taken together, they add up to a bear market indicator of 65%, which is the wrong side of normal (50%) but not dangerously so (a reading of 75%-80% would have us worried). Figure 28 – US equity bear market indicator since 1900

Notes: The bear market indicator is the average of the US yield curve (10y yield minus Fed rates), earnings yield gap (inverse of the Shiller PE minus 10-year yield), Shiller PE and EPS momentum (3m/3m). Each is expressed with reference to the cumulative distribution of its own history since 1881 (since 1914 for yield curve), assuming a normal distribution. A higher reading suggests more risk of an equity bear market (maximum = 100%). Monthly from 31 January 1900 to 31 October 2018. Source: Global Financial Data, Robert Shiller, Datastream and Invesco. See Appendix 8 for definitions and methodology.



One-year equity and real estate forecasts assume some rebound after the recent sell-off

Projections for 2019 and beyond Figures 29 and 30 show a summary of our projected returns and spot forecasts. The rationale for the five-year projections is that central bank rates and bond yields are expected to rise; IG and HY spreads are expected to normalise upwards; equity and real estate yields are expected to be broadly stable but modest income growth helps produce mid-to-high single-digit returns. Commodity prices are expected to normalise (downward) in real terms. All assumptions and regional detail are in Appendices 4 & 5. Figure 29 – Expected total returns on global assets and neutral portfolio (local currency, annualised)

As of 31 October 2018. There is no guarantee that these views will come to pass. See appendices for definitions, methodology and disclaimers. Source: Invesco The one-year equity and real estate projections may appear aggressive but remember they are based on prices as of 31 October, which were depressed. A more realistic idea is given by the index targets in Figure 30 (especially for the S&P 500). Figure 30 – Market forecasts Current Forecast (31/10/18) 1-year 5-year Central Bank Rates US 2.25 3.00 2.50 Eurozone -0.40 -0.25 0.50 China 4.35 4.00 4.00 Japan -0.10 -0.10 0.50 UK 0.75 0.75 1.00 10yr Bond Yields US 3.16 3.35 3.50 Eurozone 0.38 0.60 1.80 China 3.53 3.75 4.00 Japan 0.13 0.20 1.00 UK 1.44 1.45 1.70 Exchange Rates/US$ EUR/USD 1.13 1.15 1.20 USD/CNY 6.98 7.20 7.50 USD/JPY 112.94 105.00 85.00 GBP/USD 1.28 1.25 1.40 USD/CHF 1.01 1.00 1.00 Equity Indices S&P 500 2712 2925 3000 Euro Stoxx 50 3198 3700 4375 FTSE A50 11086 13250 17800 Nikkei 225 21920 26000 30500 FTSE 100 7128 7700 9700 Commodities (US$) Brent/barrel 75 60 45 Gold/ounce 1216 1100 900 Copper/tonne 6037 6000 5800

Notes: There is no guarantee that these views will come to pass. See Appendices for definitions, methodology and disclaimers. Source: Datastream and Invesco



We expect the best returns in the most volatile assets, for the most part

The optimiser likes the returns expected on real estate but is also attracted to cash, despite the low returns (cash has low volatility and low correlation to other assets)

Optimisation favours cash and real estate An optimisation based on the expected returns shown in Appendix 4 allows a balancing of risk and reward (we optimise for global asset class weights and then manually allocate across the regions within each asset class). The optimiser is a useful tool but judgement is the final ingredient. Figure 31 compares our projected five-year local currency returns with historical volatilities. Extra volatility is largely rewarded by higher expected returns (the commodity outlier is due to our bearish long-term price forecasts). Figure 31 – Return versus risk for global assets over five years

Based on annualised local currency returns. Size of bubbles is in proportion to average pairwise correlation with other assets (the hollow bubble for cash indicates a negative correlation with other assets). Cash is an equally weighted mix of USD, EUR, GBP and JPY. Neutral portfolio weights shown in Figure 3. As of 31 October 2018. There is no guarantee that these views will come to pass. See Appendices for definitions, methodology and disclaimers. Source: BAML, MSCI, GSCI, FTSE, Datastream and Invesco Correlations also play a role: the optimiser favours those assets offering the most diversification (cash and commodities according to the size of the bubbles). The optimised allocations are shown in Figure 32 (based on local currency returns). Some results are clear: cash and real estate positions are maximised and commodity positions (including gold) minimised, no matter whether we use one-year or five-year returns or whether we maximise the Sharpe Ratio or maximise returns (subject to volatility being no higher than for the Neutral portfolio). On the other hand, the outcomes vary for fixed income and equities: the longer the time-frame the more that IG and HY credit are preferred to government debt and equities; the more risk we are prepared to take (“Max Return vs “Sharpe Ratio”), the more that equities are favoured over government bonds.

Figure 32 – Optimised allocations for global assets (using local currency returns) Using 1y Return Using 5y Return Neutral

Portfolio Policy Range

Sharpe Ratio

Max Return

Sharpe Ratio

Max Return

Model Asset Allocation*

Cash & Gold 5% 0-10% 10% 10% 10% 10% 10% Cash 2.5% 0-10% 10% 10% 10% 10% 10% Gold 2.5% 0-10% 0% 0% 0% 0% 0% Government Bonds 30% 10-50% 44% 37% 34% 14% 20% Corporate IG 10% 0-20% 0% 0% 20% 20% 16% Corporate HY 5% 0-10% 0% 0% 10% 10% 8% Equities 45% 20-70% 40% 47% 20% 40% 40% Real Estate 3% 0-6% 6% 6% 6% 6% 6% Commodities 2% 0-4% 0% 0% 0% 0% 0%

Notes: Based on local currency returns (for both the projected returns and historical covariance matrix). Cash is an equally weighted mix of USD, EUR, GBP and JPY. “Sharpe Ratio” shows the results of maximising the Sharpe Ratio. “Max Return” maximises returns while not exceeding the volatility of the Neutral Portfolio. We focus on the version where returns are maximised using the 5-year projections (hence the different colour text. * This is a theoretical portfolio and is for illustrative purposes only. It does not represent an actual portfolio and is not a recommendation of any investment or trading strategy. See appendices for definitions, methodology and disclaimers. Source: Invesco


The volatility of equities counts against that asset class

The optimisation results vary according to which currency base is used

Interestingly, and despite the high projected returns for equities, the optimised allocations for that asset class are mainly below Neutral. The problem is that the high returns come with high volatility and that, according to our estimates, real estate offers a better alternative: real estate is on the efficient frontier, the equity asset class is not. At the other end of the efficient frontier is cash, which is also the best diversifier (it has the lowest cross asset correlation). Hence, the optimiser seeks out cash and real estate but is more reticent about equities. Among fixed income assets, Figure 29 made it clear that our projected return premium on credit versus government debt is higher over five years than over one year (the effect of widening spreads occurs over a longer period), which we think explains why government debt is preferred to credit over one year but not over five years. Given that we have written about the importance of lengthening time horizons and not getting caught up in too many short-term decisions (see All things come to he who waits), we focus on the five-year results when there are such differences. We also put more focus on the “Max Return” column because we believe that maximising the Sharpe Ratio pushes us too far to the low volatility end of the efficient frontier. We use local currency returns in the optimisation process to avoid the bias that comes from our currency forecasts. However, we also run the optimisation from the point of view of investors based in USD, EUR, GBP and CHF (see Appendix 6). Figure 33 – Optimal allocations depending upon currency base

The chart shows optimal allocations for global equities and government bonds, depending upon the currency base of the investor (based on maximising the Sharpe Ratio when using 5yr projected returns). The expected returns and covariance matrix are translated into the relevant currency base (cash returns are those in the relevant currency). The allowable ranges are 10%-50% for government bonds and 20%-70% for equities. See appendices for definitions, methodology and disclaimers. Source: Datastream and Invesco Figure 33 shows how the choice of currency base impacts optimal allocations to equities and government debt (the minimum for the latter is 10%). Interestingly, the translation of returns into another currency does make a difference (the Sharpe Ratios of asset groups are affected in different ways). Focusing on the results when using five-year returns, some conclusions remain the same as for our local currency approach (preference for real estate and cash) but there are also some big differences (in general less IG and more government debt and equities).



Combining value and momentum Real estate, US equities and EM debt seem to be in the sweet-spot We avoid fixed income and metals Non-US equities and EM real estate on the launchpad?

Hope versus reality A process that uses valuations to calculate long term return potential often runs into the reality of what markets thinks right now (value investors often suffer quietly until markets move in their favour). We have previously applied the value-momentum approach to sector selection and Figure 34 is our attempt to apply it to asset class selection. The sweet-spot is the top-right quadrant – assets that we think will produce above average returns over the next five years and that have been outperforming our Neutral benchmark index over the last year. Assets in this quadrant are largely in the real estate group (REITs), along with US equities and EM government debt. The inclusion of the latter may be a surprise but remember the chart is based on local currency returns. We believe that assets should move around the chart in a clockwise fashion but there are few assets in the bottom-right quadrant (assets that have performed well but that we think will produce below average returns: USD cash, UK government debt and energy). The bottom-left quadrant contains those assets that we think will give below-par returns and which have been underperforming in the recent past. This includes a lot of the fixed income categories (especially in Japan and the Eurozone), along with metals. The top-left quadrant is worthy of interest, since it shows assets with strong potential (we think) that have not yet started outperforming. It contains all non-US equity regions and EM real estate, which we think offers the best medium-term potential among all assets. The evidence in Figure 34 would push us to stick with our real estate and EM debt holdings and to think carefully about when to boost non-US equity allocations.

Figure 34 – Hope versus reality (returns relative to Neutral benchmark)

Chart shows a comparison of our projections for 5-year returns versus actual returns over the last 12 months, as of 31 October 2018. All are total returns, expressed in local currency (LC) and relative to the returns on our Neutral benchmark (see Figure 3 for details of the benchmark). “EQ” = Equity, “Govt” = government bonds, “REIT” = real estate investment trust, “Ind Met” = industrial metals, “Prec Met” = precious metals, “Agri” = agricultural products. Arrows show the way in which we think assets should move around the quadrants (except for the arrow under “Energy” to indicate that this asset is off the scale with 1-year relative performance of 25.2%). Past performance is no guarantee of future results. Source: BofAML, FTSE, GSCI, JP Morgan, MSCI, Datastream and Invesco


Late-cyclicals should be interesting but we find them expensive We think travel & leisure is the stand-out late-cyclical European banks could benefit from any Italian settlement, in our opinion Among defensives we find telcos and personal & household goods to be attractive

Sectors Because our central economic scenario is still positive for 2019, which should broadly favour risk assets, we prefer cyclical sectors over defensives. Nevertheless, we are concerned that: 1) the slow grind upwards in equity indices will be replaced by a more volatile ride; 2) if the economy starts weakening in the US and does not meaningfully accelerate in Europe, equities may start pricing in the next downturn. Since most major economies are in the late-expansion phase, we should logically favour late-cyclicals. However, comparing current valuations with those implied by our proprietary multiple regression model (Figure 35), it is not obvious that this is the right time to buy some of those late-cyclical sectors. First, we remain negative on most resource-related sectors (oil & gas, basic resources and chemicals). We believe that a rally in commodities is unlikely from here (we forecast declining oil prices, for example), especially if global growth decelerates next year. The “wildcard” here is the Chinese economy: if the PBOC loosens policy further and manages to reverse the slowdown in growth, commodity prices may rally. Valuations in industrials also remain elevated, despite the correction in October. They are also one of the most likely victims of deteriorating trade relationships. The only late-cyclical sector where our multiple regression model suggests good value is travel & leisure. We believe that wage growth and lower oil prices will provide a tailwind for the sector. Also in the consumer discretionary space, the valuations for US media and European autos look interesting. Recent negative sentiment would suggest that we stay away from European banks, especially until the European Union and the Italian government work out their differences over Italy’s budget. However, we think that the Eurozone economy is more likely to strengthen than weaken from here, which will boost banking sector profitability. European financial services also look good value based on our multiple regression model and could also benefit from more disposable income. US financials on the other hand look mostly overvalued and could be vulnerable to any sign of economic slowdown. Defensives have been useful in reducing the volatility of a sector portfolio and moderating drawdowns. We believe that the extreme low volatility of the QE era will not return and markets will be prone to occasional corrections with hawkish central banks withdrawing liquidity and sovereign debt starting to provide a real alternative to equities in the US. Our preferred defensive sectors in the central scenario are personal & household goods and telecommunications. We see the former as undervalued defensive growth and the latter as relatively cheap exposure to defensive value. Figure 35 – Premium/discount to model-predicted ratio*

*% above/below using dividend yield. See appendices for methodology and disclaimers. As of 31 October 2018. Source: Datastream and Invesco

Oil & Gas Basic Res

ChemicalsCon & Mat

Ind G&S

Autos

Media

Retail

Trav & Leis

Food & Bev

Pers & Hh Gds

Healthcare

BanksFin Serv

Insurance

Real Est

Technology

Telecoms

Utilities

-30%

-20%

-10%

0%

10%

20%

30%

40%

-40% -30% -20% -10% 0% 10% 20% 30% 40%

US

Euro

pe



Our central scenario remains one of 3% global growth and 3% global inflation We think recession is the most probable alternative scenario Preferred central scenario assets are cash, real estate, credit and equities

What if we are wrong? Five scenarios for 2019 It is of course possible that we are wrong. What we have described so far should be thought of as our central scenario and we need to consider other possibilities. Figure 36 shows five scenarios, along with our subjective probabilities for 2019. The central scenario remains one of 3% global GDP growth and 3% global consumer price inflation and we assign a probability of 60%. This is the same central scenario that we have used over the last few years but assumes that both growth and inflation will be lower than the presumed outturn during 2018, when we suspect both GDP growth and inflation will be closer to 3.5%. It should be noted that we expect core inflation to gradually rise during 2019 but believe the ongoing decline (or even stability) in the oil price will cause headline inflation to fall towards core inflation. Figure 36 – Five scenarios for the global economy in 2019 with probabilities*

* Our subjective probabilities are shown as the large percentage figures within each box. Source: Invesco Though we could describe an infinite number of scenarios around the central case, we focus on just four. “Recession” and “boom” are the extremes on what we might call the Phillips Curve continuum, whereby more/less demand/growth is accompanied by more/less inflation. The diagonal that links “stagflation” and “benign deflation”, on the other hand, describes supply side-shocks. Of these alternatives, we assign the highest probability (25%) to “recession”, which in fairness could also be described as “deceleration”, given that we define it as global GDP growth below 2%. In theory, the central scenario should be in the centre of the probability distribution, with as much upside as downside risk. However, our subjective probabilities reflect the maturity of the cycle (recession gets closer by the day) and the fact that we believe the distribution of growth rates around the centre is skewed to the negative side (we find it easy to imagine -2% but not +8%). Compared to a year ago we have boosted the probability assigned to the “recession” scenario (to 25% from 15%) but we have also boosted the probability of the central outcome (from 50% to 60%). This is because we now feel the chances of a “boom” scenario have diminished (from 15% to 5%) – if it was going to happen, 2018 would have been the year due to US fiscal expansion. We have also reduced the probability of the supply-side shocks to 5% from 10%. We show in Figure 37 the assets that we would prefer under each scenario. Not surprisingly, given all that was written in the earlier sections, our central scenario asset preferences are cash, real estate, credit and equities. Within the equity asset class, we put the focus on consumer discretionary, travel & leisure and personal & household goods, along with a preference for price momentum among factors.



Under recession, the preferences switch to gold, government debt and IG credit.

Under recession (our favoured alternative scenario) we would prefer assets such as government debt, IG credit, gold and defensive equities (especially the low volatility factor). We would expect gold to benefit from lower bond yields. The choices under a stagflation outcome would be somewhat similar (gold and the low volatility factor, for example) but among bonds we would only choose the inflation protected variety and would now prefer commodities and resource related equity sectors. If anything, the fact that we place more emphasis on recessionary alternatives should lead us to be more conservative in our allocations, as described in the following section. The boom and benign deflation preferences are variants on the central scenario choices, with some tweaking of sector and factor preferences (we suspect commodities, resource sectors and the value factor would benefit from a reacceleration of the global economy). We do not believe these scenarios are very likely.

Figure 37 – Five scenarios for 2019 and our favoured assets (percentages reflect assigned probabilities)

See appendices for definitions, methodology and disclaimers. Source: Invesco



No major changes since last time EM real estate is now among our favourite assets. We add to UK and reduce Japan real estate. Eurozone fixed income unattractive, in our view Slight global equity underweighting comes from big US underweighting. We add to UK and Japan equities by reducing EM Government debt unattractive except for EM and US, in our view We are neutralising several previous underweightings to UK assets

Model Asset Allocation: adding to UK; reducing EM In arriving at our allocations for the coming period, we are focused largely on the five-year outlook in local currency terms but we can and do allow for shorter term risks/opportunities and for how the currency base changes the conclusions. We are making no changes to the broad asset allocation but rather make some limited changes within the equity and real estate categories. Geographically, we are shrinking our exposure to emerging markets, while adding to the UK. Figure 3 shows the full regional detail of the allocations. Though yields and expected returns are higher, optimal allocations are little changed since we last published in September (relative rankings are unchanged). Hence, it should be no surprise that we make no changes to the broad model asset allocations. Cash remains at the maximum allowed (the projected return on other assets is insufficient to overcome the attractive characteristics of cash i.e. low volatility and low correlation to other assets). Real estate is also given the maximum allocation that we allow ourselves (it is volatile but offers the best medium-term returns, we think). We are making a slight adjustment to the regional allocation by reducing the small allocation to Japan and putting it into the UK, based on higher expected returns for the latter (see our projections in Appendix 4). Within real estate, we believe the EM region offers the best potential (we expect the yield to decline from the current 5.3%) but we already have the maximum allowable allocation. Otherwise, we remain Overweight in both IG and HY credit, with allocations of 16% and 8%, respectively, versus neutral allocations 10% and 5%. The allocations to both would be higher were it not for the near-zero returns that we project in the eurozone (we are underexposed to eurozone fixed income assets in general, versus our neutral benchmark). Despite the attractive returns on offer in some parts of the equity asset class, we follow the conclusions of the optimisation process (and the fact we consider recession to be the most likely alternative scenario), by being slightly Underweight, with an allocation of 40% versus a Neutral 45%. Given our belief that US equities are expensive and that future returns will therefore be low (we predict 4.4% annualised total return over five years), we are very much Underweight US equities with an allocation of 8% versus a Neutral 25%. Despite the evidence in Figure 34 which shows that other regions may have potential but have yet to start performing, we are Overweight all other regions, except the UK where we are Neutral (in fact, we are now boosting the UK allocation from 3% to a Neutral 4%). We are also taking the Japan equity allocation from 7% to a maximum allowed 8%, given the strong returns that we predict for 2019 and over the years that follow (earnings and dividends are growing strongly). Finally, we are financing those UK and Japan additions by reducing the EM allocation (lacklustre profit growth reduces the potential for returns over the next year). The allocation to government debt remains 20% (versus a Neutral 30%), with zero allocations to both Japan and the eurozone (where negative returns are projected over both the one and five-year timeframes – see Appendix 4). The only regions where we expect reasonable medium-term returns are the US and EM (even allowing for some currency weakness in the latter). Commodities (including gold), remain zero-weighted because of the negative returns that we project and the volatility inherent in the asset class. Given the delicate state of the Brexit process, this may seem like a strange moment to be adding to allocations in the UK. However, sterling has weakened again and is not too far from our end-2019 forecast of 1.25 (GBPUSD). We suspect a lot of bad news is in the price and want to take advantage of attractive asset valuations by neutralising some positions where we had previously been Underweight (because of the Brexit process). A move to Overweight must await more clarity on the Brexit outcome.



We do not feel the need to hedge, though the yen remains attractive Sterling going through a long bottoming-out process

Finally, we continue with the policy of not hedging currency exposures. Apart from the Chinese yuan, our analysis suggests that no major currency exhibits an extreme (and therefore risky) overvaluation. We suspect the yuan will drift lower over time but that concerns about the reaction of the US authorities will limit the pace at which the PBOC allows that to happen. Of course, bouts of risk-aversion could lead to so-called “safe-havens” appreciating. Such periods of doubt could be caused by continued trade tensions, a disruptive Brexit outcome, a breakdown of relations between Italy and the EU or a weakening of the US economy, for example. For the moment, the global economy still seems to have decent momentum, so we feel comfortable that any such bouts would be short-lived. Figure 38 suggests the yen is the major currency that remains the furthest below its long term historical norm. For now, the BOJ stands out as the major central bank most likely to continue with aggressive asset purchases, which may depress the Japanese currency in the short term. To the extent that is true, when the BOJ decides to first taper and then stop those asset purchases, the yen could strengthen significantly (Figure 30 shows that we expect USD/JPY to be 85 in five years). Given that it also seems to act as a so-called “safe-haven”, due we think to Japan’s large positive net international investment position, we would hedge into the yen when hedging becomes necessary. Though sterling looks attractive based on Figure 38, we feel that the Brexit process will permanently weaken the pound versus historical norms and doubt that it will return to those norms on an ongoing basis (for example, we forecast that GBPUSD will be 1.40 in five years versus a historical norm in the 1.50-1.60 range). Figure 38 – Real effective exchange rates*

*Currency indices measured against a trade-weighted basket of currencies and adjusted for inflation differentials. As of 31 October 2018. Source: OECD, Datastream and Invesco



USD cash US IG EM real estate Japanese equities

Where do we expect the best returns? Each quarter we highlight a list of four assets based on our five-year projections. When we published the September 2018 edition, we nominated the following list: US real estate, eurozone equities, emerging market sovereign debt and USD cash. Not enough time has passed to make a valid comment about performance, though, given the subsequent market volatility, it was just as well that we included USD cash in the list. Looking ahead, and based on the return projections in Appendix 4, we are choosing assets where we expect healthy five-year returns but where we also see potential during 2019. We are sticking with USD cash. We view cash as a good diversifier and dampener of volatility and US cash rates are higher than any of the other developed market equivalents that we consider. We predict a return of 2.4% over 12 months and 3.1% annualised over five years. That may not seem much but it is close to the 4.4% annualised five-year return that we predict for US equities, with far less volatility. Also with an eye on the risk of recession, we prefer IG credit to government debt (as a low volatility asset) and prefer US IG to other regions. Spreads are narrow but even allowing for a normalisation of spreads over five years, we believe that US IG will generate a return of 2.5% over one year and an annualised 4.1% over five years (the yield is currently 4.3%). Given its limited volatility (and the poor returns expected on many other fixed income alternatives), we think that is attractive. Real estate is our favoured asset class and we are switching our regional preference from the US to emerging markets. EM REITs have performed poorly but rentals/dividends are growing nicely and the yield has risen to 5.3% from around 3.2% at the end of January. We envisage a decline in that yield to 4.5% by end-2019 and a further decline to 4.0% in five years. Along with modest 5% annual dividend growth, we believe that will generate a return of 25.5% over one year and an annualised return of 13.9% over five years (in USD). Within equities, we are switching our preference away from the eurozone to Japan. In the short-term, the Japanese corporate sector seems to be generating better profit growth (see Figure 18) and we expect this to help generate a healthy return during 2019. Over five years we expect modest dividend growth of 4% per year but with a slight decline in yield we think the asset class will generate an annualised total return of 8.9%, which translates to 15.3% in USD (given our view that the yen will strengthen). Hence, the four assets are now: Japanese equities EM real estate US IG credit USD cash



Appendices

Appendix 1: Consensus economic forecasts Figure xx – Consensus economic forecasts GDP Growth (%) 2017 2018 2019 2020 World 3.7 3.8 3.6 3.2 US 2.2 2.9 2.5 1.9 Eurozone 2.4 2.0 1.8 1.6 China 6.9 6.6 6.2 6.0 Japan 1.8 1.1 1.1 0.5 UK 1.7 1.3 1.5 1.7 Brazil 1.0 1.4 2.3 2.6 Russia 1.5 1.8 1.5 1.7 India 7.1 6.2 7.6 7.2 Canada 3.1 2.1 2.1 1.7 Australia 2.2 3.2 2.8 2.7 CPI Change (%) 2017 2018 2019 2020 World 3.2 3.3 3.3 3.2 US 2.1 2.5 2.3 2.3 Eurozone 1.5 1.8 1.7 1.7 China 1.6 2.2 2.4 2.3 Japan 0.5 1.0 1.1 1.5 UK 2.7 2.5 2.1 2.0 Brazil 3.5 3.7 4.4 4.2 Russia 3.7 2.9 4.7 4.0 India 3.3 3.3 4.5 5.0 Canada 1.6 2.4 2.1 2.0 Australia 1.9 2.1 2.2 2.4 Nominal GDP (%) 2017 2018 2019 2020 World 7.0 7.2 7.0 6.5 US 4.3 5.5 4.9 4.2 Eurozone 3.9 3.8 3.5 3.3 China 8.6 8.9 8.7 8.4 Japan 2.3 2.1 2.2 2.0 UK 4.4 3.8 3.6 3.7 Brazil 4.5 5.2 6.8 6.9 Russia 5.3 4.8 6.3 5.8 India 10.6 9.8 12.4 12.5 Canada 4.7 4.6 4.2 3.7 Australia 4.1 5.4 5.1 5.2 Source: Bloomberg L.P., except for India and Brazil CPI (provided by Oxford Economics). There is no guarantee that these views will come to pass.



Appendix 2: Global valuations vs history Figure xx – Regional yields within historical ranges

Notes: Past performance is no guarantee of future results. See appendices for definitions, methodology and disclaimers. Source: BAML, FTSE, Datastream, Invesco

-5

0

5

10

15

20

25

30Average Now



Appendix 3: Asset class total returns

Notes: *Five-year returns are annualised. **The currency section is organised so that in all cases the numbers show the movement in the mentioned currency versus USD (+ve indicates appreciation, -ve indicates depreciation). Past performance is no guarantee of future results. Please see appendix for definitions, methodology and disclaimers. Source: Datastream and Invesco.

Data as at 31/10/2018 CurrentIndex Level/RY 3m YTD 12m 5y* 3m YTD 12m 5y*

Equities

World MSCI 485 -6.3 -3.5 0.0 6.7 -5.3 -1.3 1.3 8.6Emerging Markets MSCI 956 -11.6 -15.4 -12.2 1.1 -9.3 -10.1 -8.5 5.1US MSCI 2580 -3.4 2.9 7.2 11.2 -3.4 2.9 7.2 11.2Europe MSCI 1577 -9.9 -9.4 -7.8 1.8 -7.2 -4.1 -4.8 5.8Europe ex-UK MSCI 1853 -10.2 -9.4 -9.0 2.5 -7.5 -4.2 -6.4 6.2UK MSCI 1099 -9.1 -9.2 -4.6 0.2 -6.7 -3.9 -0.8 4.9Japan MSCI 3139 -5.3 -6.7 -3.2 5.2 -4.6 -6.5 -3.9 8.2Government BondsWorld BofA-ML 1.53 -2.0 -3.2 -1.8 0.0 -0.7 -1.0 -0.9 2.3Emerging Markets JPM 7.09 -4.8 -9.2 -6.4 -2.3 0.0 1.2 2.6 6.6US (10y) Datastream 3.16 -1.0 -4.7 -4.7 1.2 -1.0 -4.7 -4.7 1.2Europe Bofa-ML 1.02 -3.7 -6.1 -3.5 -0.2 -0.6 -0.5 -0.8 3.5Europe ex-UK (EMU, 10y) Datastream 0.38 -2.5 -3.9 -1.4 0.4 0.7 1.9 1.3 4.1UK (10y) Datastream 1.44 -2.0 -5.4 -2.1 0.2 0.7 0.2 1.7 4.9Japan (10y) Datastream 0.13 -1.4 -0.6 0.6 -1.1 -0.6 -0.4 -0.1 1.7IG Corporate BondsGlobal BofA-ML 3.43 -1.8 -4.3 -3.0 1.5 -0.8 -2.4 -2.0 3.0US BofA-ML 4.34 -1.1 -3.5 -2.8 3.0 -1.1 -3.5 -2.8 3.0Europe BofA-ML 1.20 -3.6 -6.3 -3.8 -0.9 -0.4 -0.7 -1.1 2.8UK BofA-ML 2.88 -2.6 -6.8 -3.7 0.4 0.0 -1.4 0.1 5.2Japan BofA-ML 0.38 -0.9 0.0 1.0 -2.0 -0.1 0.2 0.3 0.7HY Corporate BondsGlobal BofA-ML 6.75 -1.1 -1.2 -0.8 3.9 -0.6 -0.2 -0.3 4.8US BofA-ML 7.01 -0.4 0.9 0.9 4.7 -0.4 0.9 0.9 4.7Europe BofA-ML 4.17 -4.3 -6.8 -4.4 0.7 -1.1 -1.3 -1.7 4.4Cash (Overnight LIBOR)US 2.18 0.5 1.5 1.7 0.6 0.5 1.5 1.7 0.6Euro Area -0.46 -3.4 -6.1 -3.3 -3.8 -0.1 -0.4 -0.4 -0.3UK 0.68 -2.6 -5.1 -3.4 -4.1 0.2 0.4 0.5 0.4Japan -0.09 -1.0 -0.3 0.6 -2.7 0.0 0.0 -0.1 0.0Real Estate (REITs)Global FTSE 1742 -5.5 -4.8 -0.7 4.7 -2.5 0.9 2.1 8.6Emerging Markets FTSE 1884 -12.9 -18.4 -15.5 2.6 -10.0 -13.5 -13.1 6.4US FTSE 2842 -2.3 -0.1 2.8 7.4 -2.3 -0.1 2.8 7.4Europe ex-UK FTSE 3307 -8.5 -6.3 0.5 7.7 -5.5 -0.7 3.3 11.7UK FTSE 1122 -8.4 -10.8 -1.6 1.1 -5.9 -5.6 2.3 5.8Japan FTSE 2472 -3.0 3.5 4.8 -2.7 -2.2 3.7 4.1 0.0Commodities

All GSCI 2692 -1.1 5.3 11.5 -10.8 - - - -Energy GSCI 529 0.4 14.3 25.2 -13.5 - - - -Industrial Metals GSCI 1215 -6.3 -16.1 -12.3 -2.4 - - - -Precious Metals GSCI 1439 -1.7 -8.7 -6.1 -3.1 - - - -Agricultural Goods GSCI 355 -6.9 -6.5 -6.7 -10.9 - - - -Currencies (vs USD)**

EUR 1.13 -3.2 -5.7 -2.9 -3.6 - - - -JPY 112.94 -0.9 -0.2 0.6 -2.7 - - - -GBP 1.28 -2.6 -5.5 -3.8 -4.5 - - - -CHF 0.99 -1.8 -3.4 -1.1 -2.1 - - - -CNY 6.98 -2.3 -6.7 -4.9 -2.7 - - - -

Total Return (USD, %) Total Return (Local Currency, %)



Appendix 4: Expected returns (%)

Historical Projected

Total Return (USD) Yield Yield Capital Return Total Return Total Return (USD)

10y Overall Now Average 1y 5y 1y 5y 1y 5y 1y 5y

Cash -1.1 1.9 0.6 1.4 0.8 1.1 0.0 0.0 0.8 1.0 2.4 3.1 USD 0.5 1.7 2.2 1.6 2.9 2.4 0.0 0.0 2.5 2.7 2.5 2.7 EUR -1.2 2.6 -0.5 1.4 -0.3 0.4 0.0 0.0 -0.4 0.0 1.3 1.2 GBP -2.0 1.6 0.7 2.3 0.7 0.9 0.0 0.0 0.7 0.8 -1.4 2.7 JPY -1.4 0.2 -0.1 0.1 -0.1 0.5 0.0 0.0 -0.1 0.1 7.5 6.0 Gov. bonds 2.5 6.4 1.5 4.1 1.7 2.3 -0.9 -1.0 0.6 0.9 3.1 3.1 US 2.7 7.0 3.0 4.6 3.2 3.4 -1.2 -0.5 1.9 2.8 1.9 2.8 Eurozone 3.2 7.2 0.9 4.6 1.2 2.3 -1.6 -1.8 -0.6 -0.2 1.0 1.0 UK 3.3 7.5 1.4 5.6 1.4 1.7 -0.1 -0.3 1.3 1.2 -0.8 3.1 Japan 0.6 5.6 0.2 2.1 0.2 1.0 -0.4 -1.2 -0.2 -0.7 7.4 5.2 EM 8.7 8.6 7.1 7.1 7.0 7.0 0.5 0.1 7.8 7.3 2.4 5.1 Corp bonds 5.6 5.1 3.4 4.3 3.8 4.4 -1.8 -1.0 1.8 3.0 2.0 3.5 US Dollar 6.9 7.6 4.3 6.5 4.7 5.1 -2.0 -0.8 2.5 4.1 2.5 4.1 Euro 4.1 4.0 1.2 3.7 1.5 2.8 -2.0 -1.8 -0.6 0.1 1.0 1.3 Sterling 5.5 5.2 2.9 5.3 3.0 3.6 -0.8 -0.8 2.1 2.5 0.0 4.4 Japanese Yen -0.2 1.5 0.4 0.8 0.4 1.1 0.1 -0.9 0.5 -0.2 8.1 5.6 High-yield 11.1 6.5 6.8 9.2 7.4 9.3 -2.7 -2.0 2.7 3.8 2.7 3.7 US Dollar 11.2 8.2 7.0 10.2 7.7 9.2 -2.9 -1.7 2.7 4.2 2.7 4.2 Euro 11.0 5.7 4.2 8.8 4.7 8.1 -2.3 -3.2 1.0 0.8 2.7 2.0 Equities 10.6 9.5 2.6 2.7 2.5 2.9 10.8 4.2 13.6 7.0 14.0 7.1 US 13.3 10.1 2.0 2.9 2.0 2.5 8.0 2.1 10.1 4.4 10.1 4.4 Europe ex-UK 7.5 9.8 3.3 3.1 3.0 3.2 15.5 6.5 19.1 9.8 21.1 11.1 UK 6.5 9.4 3.9 4.1 3.8 3.8 7.8 6.4 11.9 10.4 9.6 12.5 Japan 7.0 9.1 2.2 1.4 2.0 2.0 18.8 6.8 21.2 8.9 30.4 15.3 EM 8.2 10.4 3.1 2.5 3.0 3.0 7.5 7.1 10.7 10.3 5.2 8.0 Real Estate 9.9 6.0 4.2 3.7 4.0 3.9 12.3 5.9 16.9 10.1 16.7 10.7 US 11.0 12.0 4.3 4.5 4.0 4.0 11.8 5.5 16.4 9.8 16.4 9.8 Europe ex-UK 11.0 6.8 4.0 4.3 4.0 4.0 8.0 4.0 12.3 8.2 14.2 9.4 UK 4.7 4.0 4.1 3.4 4.0 4.0 7.6 4.7 12.0 8.9 9.6 11.0 Japan 5.7 1.0 2.5 1.9 2.5 2.5 5.0 4.2 7.6 6.8 15.8 13.0 EM* 10.0 10.0 5.3 3.2 4.5 4.0 26.0 11.5 32.1 16.4 25.5 13.9 Commodities -6.7 7.0 - - - - -10.7 -5.3 -10.7 -5.3 -10.7 -5.3 Energy -9.3 4.8 - - - - -19.8 -9.7 -19.8 -9.7 -19.8 -9.7 Ind. Metals 0.1 6.2 - - - - -0.6 -0.8 -0.6 -0.8 -0.6 -0.8 Prec. Metals 4.4 6.0 - - - - -9.5 -5.8 -9.5 -5.8 -9.5 -5.8 Agriculture -4.7 2.6 - - - - 10.0 5.0 10.0 5.0 10.0 5.0 Notes: *Less than 10y history for Emerging Market Real Estate. See appendices for definitions, methodology and disclaimers. There is no guarantee that these views will come to pass. Source: BAML, JP Morgan, MSCI, FTSE, GSCI, Datastream and Invesco



Appendix 5: Key assumptions Key assumptions for 1-year projected returns US Eurozone/

Europe ex-UK UK Japan EM China

Central bank rates (%) 3.00 -0.25 0.75 -0.10 - 4.00 Sovereign spreads vs rates (bps) 30 150 75 30 - - Corporate IG spreads vs sovereign (bps) 150 35 160 15 - - Corporate HY spreads vs sovereign (bps) 450 350 - - - - Corporate HY default rates (%) 3.0 2.2 - - - - Corporate HY recovery rates (%) 43 50 - - - - Equities dividend growth (%)* 8.0 5.0 5.0 8.0 4.0 7.0 Equities dividend yield (%)* 2.0 3.0 3.8 2.0 3.0 2.3 Real estate dividend growth (%)* 4.0 8.0 5.0 5.0 7.0 - Real estate dividend yield (%)* 4.0 4 4 2.5 4.5 - Notes: *assumptions for Europe ex-UK. One-year assumptions are based on our analysis of how current values compare to historical norms (assuming some degree of reversion to the mean, except where our analysis suggests historical norms are unlikely to be a guide to the future), adjusted for our view about the development of the economic and financial market cycles over the next year in each region. There is no guarantee that these views will come to pass. Source: Invesco

Key assumptions for 5-year projected returns US Eurozone/

Europe ex-UK UK Japan EM China

Central bank rates (%) 2.50 0.50 1.00 0.50 - 4.00 Sovereign spreads vs rates (bps) 100 190 75 50 - - Corporate IG spreads vs sovereign (bps) 165 50 190 10 - - Corporate HY spreads vs sovereign (bps) 575 575 - - - - Corporate HY default rates (%) 4.3 3.8 - - - - Corporate HY recovery rates (%) 43 50 - - - - Equities dividend growth (%)* 6.5 6.0 6.0 4.0 7.0 7.0 Equities dividend yield (%)* 2.5 3.2 3.8 2.0 3.0 2.3 Real estate dividend growth (%)* 4.0 3.0 4.0 4.0 5.0 - Real estate dividend yield (%)* 4.0 4.0 4.0 2.5 4.0 - Notes: *assumptions for Europe ex-UK. Five-year assumptions are based on our analysis of how current values compare to historical norms (assuming a large degree of reversion to the mean, except where our analysis suggests historical norms are unlikely to be a guide to the future). Economic and financial market cycle considerations may impact assumptions about the early stages of the five-year horizon but, except in extreme circumstances, we assume that projected long-term norms will prevail in five years. There is no guarantee that these views will come to pass. Source: Invesco



Appendix 6: Optimised allocations for global assets for different currency bases Optimised allocations for global assets in USD Using 1y Return Using 5y Return Neutral


Sharpe Ratio

Max Return

Sharpe Ratio

Max Return


Cash & Gold 5% 0-10% 10% 10% 10% 10% 10% Cash 2.5% 0-10% 10% 10% 10% 10% 10% Gold 2.5% 0-10% 0% 0% 0% 0% 0% Government Bonds 30% 10-50% 38% 35% 41% 35% 20% Corporate IG 10% 0-20% 0% 0% 20% 0% 16% Corporate HY 5% 0-10% 0% 0% 0% 0% 8% Equities 45% 20-70% 46% 49% 23% 49% 40% Real Estate 3% 0-6% 6% 6% 6% 6% 6% Commodities 2% 0-4% 0% 0% 0% 0% 0% Based on USD returns (for both the projected returns and historical covariance matrix). Cash is USD cash. “Sharpe Ratio” shows the results of maximising the Sharpe Ratio. “Max Return” maximises returns while not exceeding the volatility of the Neutral Portfolio. *This is a theoretical portfolio and is for illustrative purposes only. It does not represent an actual portfolio and is not a recommendation of any investment or trading strategy. Arrows show direction of change of allocations. See appendices for methodology and disclaimers. Source: Invesco

Optimised allocations for global assets in GBP Using 1y Return Using 5y Return Neutral


Sharpe Ratio

Max Return

Sharpe Ratio

Max Return


Cash & Gold 5% 0-10% 10% 10% 10% 10% 10% Cash 2.5% 0-10% 10% 10% 10% 10% 10% Gold 2.5% 0-10% 0% 0% 0% 0% 0% Government Bonds 30% 10-50% 14% 18% 10% 11% 20% Corporate IG 10% 0-20% 0% 0% 0% 0% 16% Corporate HY 5% 0-10% 0% 0% 10% 10% 8% Equities 45% 20-70% 70% 70% 64% 63% 40% Real Estate 3% 0-6% 6% 2% 6% 6% 6% Commodities 2% 0-4% 0% 0% 0% 0% 0% Based on GBP returns (for both the projected returns and historical covariance matrix). Cash is GBP cash. “Sharpe Ratio” shows the results of maximising the Sharpe Ratio. “Max Return” maximises returns while not exceeding the volatility of the Neutral Portfolio. *This is a theoretical portfolio and is for illustrative purposes only. It does not represent an actual portfolio and is not a recommendation of any investment or trading strategy. Arrows show direction of change of allocations. See appendices for methodology and disclaimers. Source: Invesco



Optimised allocations for global assets in CHF Using 1y Return Using 5y Return Neutral


Sharpe Ratio

Max Return

Sharpe Ratio

Max Return


Cash & Gold 5% 0-10% 0% 10% 10% 2% 10% Cash 2.5% 0-10% 0% 10% 10% 2% 10% Gold 2.5% 0-10% 0% 0% 0% 0% 0% Government Bonds 30% 10-50% 24% 31% 38% 39% 20% Corporate IG 10% 0-20% 0% 0% 0% 0% 16% Corporate HY 5% 0-10% 0% 0% 10% 10% 8% Equities 45% 20-70% 70% 53% 36% 42% 40% Real Estate 3% 0-6% 6% 6% 6% 6% 6% Commodities 2% 0-4% 0% 0% 0% 0% 0% Based on CHF returns (for both the projected returns and historical covariance matrix). Cash is CHF cash. “Sharpe Ratio” shows the results of maximising the Sharpe Ratio. “Max Return” maximises returns while not exceeding the volatility of the Neutral Portfolio. *This is a theoretical portfolio and is for illustrative purposes only. It does not represent an actual portfolio and is not a recommendation of any investment or trading strategy. Arrows show direction of change of allocations. See appendices for methodology and disclaimers. Source: Invesco

Optimised allocations for global assets in EUR Using 1y Return Using 5y Return Neutral


Sharpe Ratio

Max Return

Sharpe Ratio

Max Return


Cash & Gold 5% 0-10% 10% 10% 10% 10% 10% Cash 2.5% 0-10% 10% 10% 10% 10% 10% Gold 2.5% 0-10% 0% 0% 0% 0% 0% Government Bonds 30% 10-50% 14% 34% 27% 28% 20% Corporate IG 10% 0-20% 0% 0% 0% 0% 16% Corporate HY 5% 0-10% 0% 0% 10% 10% 8% Equities 45% 20-70% 70% 50% 47% 46% 40% Real Estate 3% 0-6% 6% 6% 6% 6% 6% Commodities 2% 0-4% 0% 0% 0% 0% 0% Based on EUR returns (for both the projected returns and historical covariance matrix). Cash is EUR cash. “Sharpe Ratio” shows the results of maximising the Sharpe Ratio. “Max Return” maximises returns while not exceeding the volatility of the Neutral Portfolio. *This is a theoretical portfolio and is for illustrative purposes only. It does not represent an actual portfolio and is not a recommendation of any investment or trading strategy. Arrows show direction of change of allocations. See appendices for methodology and disclaimers. Source: Invesco



Appendix 7: Methodology for asset allocation, expected returns and optimal portfolios Portfolio construction process The optimal portfolios are theoretical and not real. We use optimisation processes to guide our allocations around “neutral” and within prescribed policy ranges based on our estimations of expected returns and using historical covariance information. This guides the allocation to global asset groups (equities, government bonds etc.), which is the most important level of decision. For the purposes of this document the optimal portfolios are constructed with a one-year and five-year horizon. Which asset classes? We look for investibility, size and liquidity. With that in mind, we have chosen to include: equities, bonds (government, corporate investment grade and corporate high-yield), REITs to represent real estate, commodities and cash (all across a range of geographies). We use cross-asset correlations to determine which decisions are the most important. Neutral allocations and policy ranges We use market capitalisation in USD for major benchmark indices to calculate neutral allocations. For commodities, we use industry estimates for total ETP market cap + assets under management in hedge funds + direct investments. We use an arbitrary 5% for the combination of cash and gold. We impose diversification by using policy ranges for each asset category (the range is usually symmetric around neutral). Expected/projected returns The process for estimating expected returns is based upon yield (except commodities, of course). After analysing how yields vary with the economic cycle, and where they are situated within historical ranges, we forecast the direction and amplitude of moves over the next one year and five years. Cash returns are calculated assuming a straight-line move in short term rates towards our targets (with, of course, no capital gain or loss). Bond returns assume a straight-line progression in yields, with capital gains/losses predicated upon constant maturity (effectively supposing constant turnover to achieve that). Forecasts of corporate investment-grade and high-yield spreads are based upon our view of the economic cycle. Coupon payments are added to give total returns. Equity and REIT returns are based on dividend growth assumptions. We calculate total returns by applying those growth assumptions and adding the forecast dividend yield. No such metrics exist for commodities; therefore, we base our projections on US CPI-adjusted real prices relative to their long-term averages and views on the economic cycle. All expected returns are first calculated in local currency and then, where necessary, converted into other currency bases using our exchange rate forecasts. Optimising the portfolio Using a covariance matrix based on monthly local currency total returns for the last 5 years and we run an optimisation process that maximises the Sharpe Ratio. The optimiser is based on the Markowitz model. Currency hedging We adopt a cautious approach when it comes to currency hedging as currency movements are notoriously difficult to accurately predict and sometimes hedging can be costly. Also, some of our asset allocation choices are based on currency forecasts. We use an amalgam of central bank rate forecasts, policy expectations and real exchange rates relative to their historical averages to predict the direction and amplitude of currency moves.



Appendix 8: Sector multiple regression model methodology We have run a multiple regression analysis in both regions to examine how macroeconomic factors influence sector valuations. We have used the dividend yield relative to market as the dependent variable and have run the regressions with the following independent variables: US (monthly series since 31/01/1990): 1-year change in: consumer price index, average hourly earnings The level of: real oil price (US CPI adjusted), real copper price (US CPI adjusted),

ISM manufacturing index, consumer confidence index (conference board), real USD index (US CPI adjusted), unemployment rate, yield curve (10y-2y yields), net debt/EBITDA (only for non-financial sectors), EBITDA margin

Europe (monthly series since 31/01/1991): 1-year change in: German and UK consumer price indices (average change across

both) The level of: real oil price (US CPI adjusted), real copper price (US CPI adjusted),

IFO business climate index, European Union consumer sentiment index, average of the German and UK yield curves (10y-2y yields), average of the German and UK unemployment rates, net debt/EBITDA (only for non-financial sectors), return on equity

This analysis shows us which independent variables have a statistically significant relationship with sector valuation ratios. In addition, the regression coefficients tell us how much each independent variable influences those ratios. Finally, we use those coefficients to calculate what the valuation ratios should be, based on the model, and compare them to currently observed valuations. In theory, this allows us to determine whether a sector is undervalued or overvalued based on the macroeconomic factors we have used.



Appendix 9: Definitions of data and benchmarks Sources: we source data from Datastream unless otherwise indicated. Cash: returns are based on a proprietary index calculated using the Intercontinental Exchange Benchmark Administration overnight LIBOR (London Interbank Offer Rate). The global rate is the average of the euro, British pound, US dollar and Japanese yen rates. The series started on 1st January 2001 with a value of 100. Gold: London bullion market spot price in USD/troy ounce. Government bonds: Current values in the market forecast table (figure 33) use Datastream benchmark 10-year yields for the US, Eurozone, Japan and the UK and the Thomson Reuters China benchmark 10-year yield for China. Historical and projected yields and returns (figures 1, 29, 30, 31, 32 and appendix 4) are based on Bank of America Merrill Lynch government bond indices with historical ranges starting on 31st December 1985 for the Global, Europe ex-UK, UK and Japanese indices and 30th January 1978 for the US. The emerging markets yields and returns are based on the JP Morgan emerging markets global composite government bond index with the historical range starting on 31st December 2001. The same indices are used to construct figure 24 and appendix 2. Corporate investment grade (IG) bonds: Bank of America Merrill Lynch investment grade corporate bond indices with historical ranges starting on 31st December 1996 for the Global, 31st January 1973 for the US dollar, 1st January 1996 for the euro, 31st December 1996 for the British pound, and 6th September 2001 for the Japanese yen indices. Corporate high yield (HY) bonds: Bank of America Merrill Lynch high yield indices with historical ranges starting on 29th August 1986 for the US dollar, and 31st December 1997 for the Global and euro indices. Equities: We use MSCI benchmark indices to calculate projected returns and calculate long-term total returns with historical ranges starting on 31st December 1969 for the Global, US, Europe ex-UK, UK and Japanese indices, and 31st December 1987 for the emerging markets index. Equity index valuations (figures 24, 25 and appendix 2) are based on dividend yields and price-earnings ratios using Datastream benchmark indices with historical ranges starting on 1st January 1973 for the Global, US, Europe ex-UK and Japanese indices, on 31st December 1969 for the UK index and 2nd January 1995 for the Emerging Markets index. Real estate: We use FTSE EPRA/NAREIT indices with historical ranges starting on 29th December 1989 for the US, Europe ex-UK, UK and Japanese indices, 18th February 2005 for the Global index, and 31st October 2008 for the Emerging Markets index. Commodities: Goldman Sachs Commodity Index with historical ranges starting on 31st December 1969 for the All Commodities and Agriculture indices, 31st December 1982 for the Energy index, 3rd January 1977 for the Industrial Metals index, and 2nd January 1973 for the Precious Metals index. We refer to oil & gas and industrial metals as industrial commodities. Definitions and sources for Figures 26, 27 and 28 US Federal Reserve (Fed) interest rate: Fed Discount Rate from November 1914 to October 1982, then the Fed Funds Rate is used (source: Global Financial Data, Datastream) US 10-year treasury yield (bond yield): monthly from 1871 (source: Robert Shiller and Datastream)



US Shiller PE and Earnings Per Share (EPS): the Shiller PE is a price to earnings ratio constructed by dividing price by the average earnings per share in the previous 10 years (with both numerator and denominator adjusted for inflation). It is what is commonly known as a cyclically-adjusted PE ratio. It is constructed by US academic Robert Shiller. We also use the raw EPS data from his database to calculate EPS momentum on a 3m/3m basis (the percentage change in the latest three months versus the previous three months). Data is monthly from 1881 (source Robert Shiller – see here). EPS momentum data since June 1973 is derived from S&P 500 index and PE data sourced from Datastream. US stock/equity index: we have calculated a total return index for broad US stocks based on index and dividend data from US academic Robert Shiller and Datastream. The index prior to 1926 is Robert Shiller’s recalculation of data from Common Stock Indexes by Cowles & Associates (see here). From 1926 to 1957, the Shiller data is based on the S&P Composite Index and thereafter is based on the S&P 500 as we know it today. Definitions of data and benchmarks for Appendix 3 Sources: we source data from Datastream unless otherwise indicated. Cash: returns are based on a proprietary index calculated using the Intercontinental Exchange Benchmark Administration overnight LIBOR (London Interbank Offer Rate). The global rate is the average of the euro, British pound, US dollar and Japanese yen rates. The series started on 1st January 2001 with a value of 100. Gold: London bullion market spot price in USD/troy ounce. Government bonds: Current levels, yields and total returns use Datastream benchmark 10-year yields for the US, Eurozone, Japan and the UK, and the Bank of America Merrill Lynch government bond total return index for the World and Europe. The emerging markets yields and returns are based on the JP Morgan emerging markets global composite government bond index. Corporate investment grade (IG) bonds: Bank of America Merrill Lynch investment grade corporate bond total return indices. Corporate high yield (HY) bonds: Bank of America Merrill Lynch high yield total return indices Equities: We use MSCI benchmark gross total return indices for all regions. Commodities: Goldman Sachs Commodity total return indices Real estate: FTSE EPRA/NAREIT total return indices Currencies: Global Trade Information Services spot rates



Important information Your capital is at risk. You may not get back the amount you invested. By accepting this document, you consent to communicating with us in English, unless you inform us otherwise. This document is for informational purposes only and is intended only for Professional Clients and Financial Advisers in Continental Europe (as defined in important information); Qualified Investors in Switzerland; Professional Clients only in Dubai, Ireland, the Isle of Man, Jersey, Guernsey, Malta and the UK; for Qualified Clients in Israel, for Professional/Qualified/Sophisticated Investors in Bahrain, Jordan, Kuwait, Lebanon, Mauritius, Oman, Qatar, Saudi Arabia, South Africa, Tunisia, Turkey, and the United Arab Emirates; for Professional Investors in Hong Kong, for certain specific sovereign wealth funds and/or Qualified Domestic Institutional Investors approved by local regulators only in the People’s Republic of China, for Institutional Investors in Australia, the United States and Singapore; for Wholesale Investors in New Zealand; for certain specific Qualified Institutions and/or Sophisticated Investors only in Taiwan, for Qualified Professional Investors in Korea, for certain specific institutional investors in Brunei, for Qualified Institutional Investors and/or certain specific institutional investors in Thailand and for certain specific institutional investors in Malaysia, upon request, for informational purposes only. This document is only intended for use with Qualified Institutional Investors in Japan; in Canada, this document is restricted to Accredited Investors as defined under National Instrument 45-106. It is not intended for and should not be distributed to, or relied upon by, the public or retail investors. It is not intended for solicitation of any security. Please do not redistribute this document. For the distribution of this document, Continental Europe is defined as Andorra, Austria, Belgium, Czech Republic, Croatia, Denmark, Finland, France, Germany, Gibraltar, Greece, Hungary, Italy, Latvia, Liechtenstein, Luxembourg, Monaco, Netherlands, Norway, Portugal, Romania, Slovakia, Slovenia, Spain, and Sweden. This document is not an offering of a financial product and should not be distributed to retail clients who are resident in jurisdiction where its distribution is not authorized or is unlawful. Circulation, disclosure, or dissemination of all or any part of this document to any unauthorized person is prohibited. This document is only intended for and will be only distributed to persons resident in jurisdictions where such distribution or availability would not be contrary to local laws or regulations. This document is solely for duly registered banks or a duly authorized Monegasque intermediary acting as a professional institutional investor which has such knowledge and experience in financial and business matters as to be capable of evaluating the contents of this document. Consequently, this document may only be communicated to banks duly licensed by the “Autorité de Contrôle Prudentiel et de Résolution” and fully licensed portfolio management companies by virtue of Law n° 1.144 of July 26, 1991 and Law 1.338, of September 7, 2007, duly licensed by the “Commission de Contrôle des Activités Financières. Such regulated intermediaries may in turn communicate this document to potential investors. This document has been prepared only for those persons to whom Invesco has provided it. It should not be relied upon by anyone else. Information contained in this document may not have been prepared or tailored for an Australian audience and does not constitute an offer of a financial product in Australia. You may only reproduce, circulate and use this document (or any part of it) with the consent of Invesco. The information in this document has been prepared without taking into account any investor’s investment objectives, financial situation or particular needs. Before acting on the information the investor should consider its appropriateness having regard to their investment objectives, financial situation and needs. You should note that this information: may contain references to dollar amounts which are not Australian dollars;



may contain financial information which is not prepared in accordance with Australian law or practices;

may not address risks associated with investment in foreign currency denominated investments; and

does not address Australian tax issues. Issued in Australia and New Zealand by Invesco Australia Limited (ABN 48 001 693 232), Level 26, 333 Collins Street, Melbourne, Victoria, 3000, Australia which holds an Australian Financial Services Licence number 239916. This document is issued only to wholesale investors in New Zealand to whom disclosure is not required under Part 3 of the Financial Markets Conduct Act. This document has been prepared only for those persons to whom it has been provided by Invesco. It should not be relied upon by anyone else and must not be distributed to members of the public in New Zealand. Information contained in this document may not have been prepared or tailored for a New Zealand audience. You may only reproduce, circulate and use this document (or any part of it) with the consent of Invesco. This document does not constitute and should not be construed as an offer of, invitation or proposal to make an offer for, recommendation to apply for, an opinion or guidance on Interests to members of the public in New Zealand. Applications or any requests for information from persons who are members of the public in New Zealand will not be accepted. The distribution and offering of this document in certain jurisdictions may be restricted by law. Persons into whose possession this marketing material may come are required to inform them about and to comply with any relevant restrictions. This does not constitute an offer or solicitation by anyone in any jurisdiction in which such an offer is not authorised or to any person to whom it is unlawful to make such an offer or solicitation. This does not constitute a recommendation of any investment strategy or product for a particular investor. Investors should consult a financial professional before making any investment decisions. This overview contains general information only and does not take into account individual objectives, taxation position or financial needs. Nor does this constitute a recommendation of the suitability of any investment strategy for a particular investor. It is not an offer to buy or sell or a solicitation of an offer to buy or sell any security or instrument or to participate in any trading strategy to any person in any jurisdiction in which such an offer or solicitation is not authorized or to any person to whom it would be unlawful to market such an offer or solicitation. It does not form part of any prospectus. All material presented is compiled from sources believed to be reliable and current, but accuracy cannot be guaranteed. Investments have risks and you may lose your principal investment. Please obtain and review all financial material carefully before investing. Asset management services are provided by Invesco in accordance with appropriate local legislation and regulations. The opinions expressed are those of the authors and may differ from the opinions of other Invesco investment professionals. Opinions are based upon current market conditions, and are subject to change without notice. Past performance is no guarantee of future results. This material may contain statements that are not purely historical in nature but are “forward-looking statements.” These include, among other things, projections, forecasts, estimates of income, yield or return or future performance targets. These forward-looking statements are based upon certain assumptions, some of which are described herein. Actual events are difficult to predict and may substantially differ from those assumed. All forward-looking statements included herein are based on information available on the date hereof and Invesco assumes no duty to update any forward-looking statement. Accordingly, there can be no assurance that estimated returns or projections can be realized, that forward-looking statements will materialize or that actual returns or results will not be materially lower than those presented. All information is sourced from Invesco, unless otherwise stated.



Effective 8/18/17, Invesco Ltd completed the acquisition of Source. Links to documents published prior to this date are from Source as a predecessor firm and are provided for historical and informational purposes only. Investment strategies involve numerous risks. The calculations and charts set out herein are indicative only, make certain assumptions and no guarantee is given that future performance or results will reflect the information herein. Past performance is not a guarantee of future performance. The Directors of Invesco do not guarantee the accuracy and/or the completeness of any data included herein and we shall have no liability for any errors, omissions, or interruptions herein. We make no warranty, express or implied, as to the information described herein. All data and performance shown is historical unless otherwise indicated. Investors should consult their own business, tax, legal and accounting advisors with respect to this proposed transaction and they should refrain from entering into a transaction unless they have fully understood the associated risks and have independently determined that the transaction is appropriate for them. In no way should we be deemed to be holding ourselves out as financial advisers or fiduciaries of the recipient hereof and this document is not intended to be "investment research" as defined in the Handbook of the UK Financial Conduct Authority. Invesco, and our shareholders, or employees or our shareholders may from time to time have long or short positions in securities, warrants, futures, options, derivatives or financial instruments referred to in this material. As a result, investors should be aware that we may have a conflict of interest that could affect the objectivity of this report. Investors should consider this report as only a single factor in making their investment decision. This document is provided by Invesco UK Services Limited, Perpetual Park, Perpetual Park Drive, Henley-on-Thames, Oxfordshire, RG9 1HH, authorised and regulated by the Financial Conduct Authority, Invesco Asset Management S.A., 18, rue de Londres, 75009 Paris, France, authorised and regulated by the Autorité des marches financiers, Invesco Asset Management Deutschland GmbH, An der Welle 5, 60322- Frankfurt/M., Germany, Invesco Asset Management (Schweiz) AG, Talacker 34, 8001 Zurich, Switzerland, and Invesco Asset Management Limited, Perpetual Park, Perpetual Park Drive, Henley-on Thames, Oxfordshire RG9 1HH, UK Authorised and regulated by the Financial Conduct Authority. In the US by Invesco Capital Management LLC, 3500 Lacey Road, Suite 700, Downers Grove, IL 60515. In Canada by Invesco Canada Ltd., 5140 Yonge Street, Suite 800, Toronto Ontario, M2N 6X7. Terms and Conditions for Canadian investors can be seen here. This document is issued in the following countries: in Hong Kong by Invesco Hong Kong Limited景順投資管理有限公司, 41/F, Champion

Tower, Three Garden Road, Central, Hong Kong. This document has not been reviewed by the Securities and Futures Commission.

in Singapore by Invesco Asset Management Singapore Ltd, 9 Raffles Place, #18-01 Republic Plaza, Singapore 048619.

in Taiwan by Invesco Taiwan Limited, 22F, No.1, Songzhi Road, Taipei 11047, Taiwan (0800-045-066). Invesco Taiwan Limited is operated and managed independently.

In Japan by Invesco Asset Management (Japan) Limited, Roppongi Hills Mori Tower 14F, 6-10-1 Roppongi, Minato-ku, Tokyo 106-6114: Registration Number: The Director – General of Kanto Local Finance Bureau(Kin-sho) 306; Member of the Investment Trusts Association, Japan and the Japan Investment Advisers Association

© 2018 Invesco. All rights reserved. GL724.

AuthorsPaul JacksonHead of Research, EMEA ETFsTelephone +44(0)20 3370 1172Email [email protected]

Andras VigMulti-Asset StrategistTelephone +44(0)20 3370 1152Email [email protected]

Further information

Telephone +44 20 3370 1100Email [email protected]

Portman Square House, 43-45 Portman Square, London W1H 6LY

the big picture43b029b3-81cc... · the big picture 2019 outlook: war, inflation, recession? in...

Documents