Vanguard’s economic and investment outlook

Vanguard research

• This paper presents Vanguard’s global perspectives on the future of growth, inflation, interest rates, and stock and bond returns over the next ten years. As in past outlooks, we anticipate that the modest global recovery will likely endure at a below-average pace through a period of low interest rates, continuing high unemployment and debt levels, and elevated policy uncertainty. • We detail how, after years of slightly disappointing 2% real growth, the U.S. in 2014–2015 faces cyclical risks tilted toward better-than-trend growth for the first time since the onset of the global financial crisis. Our economic outlook, in short, is one of resiliency. • We also explain why last year’s unease about the “reach for yield” is now joined by concern about “froth” in certain equity markets. Market volatility is likely as the Federal Reserve undertakes the multistep, multiyear process of unwinding its extraordinarily easy monetary policy. Rather than frame this process as a negative, we view it as an indication of increasing economic strength. Note: The authors would like to thank the members of Vanguard’s Investment Strategy Group for their valuable feedback and contributions, in particular Harshdeep Ahluwalia, Vytautas Maciulis, Christos Tasopoulos, and Ravi Tolani. Connect with Vanguard >  vanguard.com

January 2014

Authors Joseph Davis, Ph.D. Roger Aliaga-Díaz, Ph.D. Charles J. Thomas, CFA Andrew J. Patterson, CFA

Vanguard’s distinct approach to forecasting To treat the future with the deference it deserves, Vanguard believes that market forecasts are best viewed in a probabilistic framework. This publication’s primary objectives are to describe the projected long-term return distributions that contribute to strategic asset allocation decisions and to present the rationale for the ranges and probabilities of potential outcomes. This analysis discusses our global outlook from the perspective of a U.S. investor with a dollar-denominated portfolio.

Global market outlook summary Global economy. For the first time since the financial crisis, our leading indicators point to a slight pickup in near-term growth for the United States, parts of Europe, and other select developed markets. Continued progress in U.S. consumer deleveraging, strong corporate balance sheets, firmer global trade, and less fiscal drag indicate U.S. growth approaching 3%. That said, this cyclical assessment should be placed against a backdrop of high unemployment and government debt; ongoing structural reforms in Europe, China, and Japan; and extremely aggressive monetary policy with exit strategies that have yet to be tested. Inflation. In the near term, reflationary monetary policies will continue to counteract the deflationary bias of a high-debt world still recovering from a deep financial crisis. As was suggested in previous outlooks, consumer price inflation remains near generational lows and, in several major economies, below the targeted rate. Key U.S. drivers generally point to higher but modest core inflation trends in the 1%–3% range for the next several years. For now, the risk of returning to the high inflationary regime of the 1970s is low despite the size of central bank balance sheets; in parts of Europe and in Japan, the specter of deflation remains a greater risk. Monetary policy. Tapering of the Federal Reserve’s quantitative easing (QE) program has begun, although an actual tightening is likely some time off. The Fed’s forward guidance implies that the federal funds rate will remain near 0% through mid-2015; the risk that this “lift-off” date will be further delayed is notably lower than it was in prior periods. However, real (inflation-adjusted) short-term interest rates will probably remain negative through perhaps 2017. Globally, the burdens on monetary policymakers are high as they contemplate exiting from QE policies to prevent asset bubbles on one hand and remain mindful of raising short-term rates too aggressively on the other. The exit may induce market volatility at times, but long-term investors should prefer that to no exit at all. Interest rates. The bond market continues to expect Treasury yields to rise, with a bias toward a steeper yield curve until the Federal Reserve raises short-term rates. Compared with last year’s outlook, our estimates of the “fair value” range for the 10-year Treasury bond have risen; the macroeconomic environment justifies a ten-year yield in the range of 2.75%–3.75% at present. However, we continue to hold the view that a more normalized environment in which rates move toward 5% based on stronger growth, inflation, and monetary tightening may be several years away. We maintain that the odds of a U.S. fiscal crisis and a sharp spike in yields are less than 10% at the moment, although they rise later in the decade based on the expected trajectory of U.S. federal debt. Global bond market. As in past editions, the return outlook for fixed income is muted, although it has improved somewhat with the recent rise in real rates. The expected ten-year median nominal return of a broad, globally diversified fixed income investment is centered in the 1.5%–3.0% range, versus last year’s expected range of 0.5%–2.0%. It is important to note that we expect the diversification benefits of fixed income in a balanced portfolio to persist under most scenarios. We believe that the prospects of losses in bond portfolios should be weighed against the magnitude of potential losses in equity portfolios, because the latter have tended to exhibit much larger swings in returns.

2

Global equity market. After several years of suggesting that strong equity returns were possible despite a prolonged period of subpar economic growth, our outlook for global equities has become more guarded. The expected ten-year median nominal return is below historical averages and has shifted toward the bottom of the 6%–9% range compared with this time last year, a reflection of less constructive market valuations (i.e., price/earnings ratios) in the United States and some other developed markets. A notably wide range of outcomes is possible, even over long horizons, making us hard-pressed to identify market “bubbles.” However, we are uneasy about signs of froth in certain segments of the global equity market. Because the premium compensating increased equity risk appears to have come down recently, we would encourage investors to exercise caution in making strategic or tactical portfolio changes that increase this risk. Asset allocation strategies. Broadly speaking, the outlook for risk premiums is lower across a range of investments than was the case just two or three years ago. Our simulations indicate that balanced portfolio returns over the next decade are likely to be below long-run historical averages, with those for a 60% stock/40% bond portfolio tending to center in the 3%–5% range, adjusted for inflation. Even so, Vanguard still firmly believes that the expected risk-return trade-off among stocks and bonds leaves the principles of portfolio construction unchanged. Specifically, our simulated mean-variance frontier of expected returns is upward sloping—it anticipates higher strategic returns for more aggressive portfolios, accompanied by greater downside risk. We believe that a long-term, strategic approach with a balanced, diversified, low-cost portfolio can remain a high-value proposition in the decade ahead.

The asset-return distributions shown here represent Vanguard’s view on the potential range of risk premiums that may occur over the next ten years; such long-term projections are not intended to be extrapolated into a short-term view. These potential outcomes for long-term investment returns are generated by the Vanguard Capital Markets Model® (VCMM—see the description in the Appendix) and reflect the collective perspective of our Investment Strategy Group. The expected risk premiums—and the uncertainty surrounding those expectations—are among a number of qualitative and quantitative inputs used in Vanguard’s investment methodology and portfolio construction process. IMPORTANT: The projections or other information generated by the VCMM regarding the likelihood of various investment outcomes are hypothetical in nature, do not reflect actual investment results, and are not guarantees of future results. Distribution of return outcomes from VCMM, derived from 10,000 simulations for U.S. equity returns and fixed income returns. Simulations as of November 30, 2013. Results from the model may vary with each use and over time. For more information, please see the appendix on page 30.

All investing is subject to risk, including the possible loss of the money you invest. Past performance is no guarantee of future returns. Investments in bond funds are subject to interest rate, credit, and inflation risk. Foreign investing involves additional risks, including currency fluctuations and political uncertainty. Diversification does not ensure a profit or protect against a loss in a declining market. There is no guarantee that any particular asset allocation or mix of funds will meet your investment objectives or provide you with a given level of income. The performance of an index is not an exact representation of any particular investment, as you cannot invest directly in an index. Stocks of companies in emerging markets are generally more risky than stocks of companies in developed countries. U.S. government backing of Treasury or agency securities applies only to the underlying securities and does not prevent price fluctuations. Investments that concentrate on a relatively narrow market sector face the risk of higher price volatility.

3

Global growth outlook

• Firming global trade and manufacturing.

Similar to our stance last year, we view the global recovery as likely to proceed at a modest pace (Figure 1a). Over the next decade, expansion should occur at varying speeds, with trend growth likely lower than during the past 20 years based on economic, demographic, and other factors that we will discuss later. Figure 1b presents estimates of potential growth rates for the major world economies.

• An aging capital stock and pent-up investment demand.

Nevertheless, for the first time since the financial crisis, our leading indicators point to the possibility of a slight cyclical pickup in near-term growth for the United States and other selected economies. Positive factors contributing to this outlook include: • Continued progress in U.S. consumer deleveraging. • Rising consumer wealth levels in many markets.

• Less fiscal drag in 2014–2015 (see Figure 2b on page 6).1 This cyclical growth assessment should, however, be placed within the context of a stubbornly lower-growth world marked by high levels of both government and private-sector debt (see Figure 2a on page 6), less favorable demographic trends, and extremely aggressive monetary policy with untested exit strategies. Downside economic tail risks appear somewhat lower than they were two years ago, although lingering doubts remain regarding the prospects for a hard landing in China and for euro stability. Various other geopolitical risks are inherently difficult to forecast.

Indexes used in our calculations The long-term returns for our hypothetical portfolios are based on data for the appropriate market indexes through November 2013. We chose these benchmarks to provide the best history possible and split the global allocations to align with Vanguard’s guidance in constructing diversified portfolios. U.S. bonds: Standard & Poor’s High Grade Corporate Index from 1926 through 1968, Citigroup High Grade Index from 1969 through 1972, Lehman Brothers U.S. Long Credit AA Index from 1973 through 1975, and Barclays U.S. Aggregate Bond Index thereafter. Ex-U.S. bonds: Citigroup World Government Bond Ex-U.S. Index from 1985 to January 1989 and Barclays Global Aggregate ex-USD Index thereafter. Global bonds: Prior to 1985, 100% U.S. bonds, as defined above. After 1985, 80% U.S. bonds and 20% ex-U.S. bonds, rebalanced monthly. U.S. equities: S&P 90 Index from January 1926 through March 3, 1957; S&P 500 Index from March 4, 1957, through 1974; Dow Jones Wilshire 5000 Index from 1975 through April 22, 2005; and MSCI US Broad Market Index thereafter. Ex-U.S. equities: MSCI World ex USA Index from January 1970 through 1987 and MSCI All Country World Index ex USA thereafter. Global equities: Prior to 1970, 100% U.S. equities, as defined above. After 1970, 70% U.S. equities and 30% ex-U.S. equities, rebalanced monthly.

1 Fiscal drag occurs when government spending cuts, tax increases, or both reduce the pace of overall spending and GDP growth.

4

Figure 1.

Global trend growth has been heading down

a. A cyclical upturn in a world of lower trend growth

b. Trend growth is lower across most large markets

Annualized growth of the global economy

Historical and future trend growth in the world’s ten largest economies Percentage Pre-recession Long-run of world trend growth potential economy (1995–2007) growth

6%

5.5%

Average annualized growth

5 3.9%

4

3.4%

3.1%

3

2.7%

3.0% 2.5%

2 1 0 1960s

1970s

1980s

1990s

2000s

2010– 2013

2014– 2016 (projected)

Notes: World GDP is shown at market exchange rates, in constant U.S. dollars. Data are from World Bank for 1960 through 1969 and International Monetary Fund’s World Economic Outlook (WEO), October 2013, for subsequent years. Projected growth is from the IMF. Sources: Vanguard calculations, based on IMF and World Bank data.

United States

22.8%

3.2%

2.5%

Eurozone

17.3%

2.3%

1.6%

China

12.2%

10.0%

7.0%

Japan

6.8%

1.2%

1.1%

United Kingdom

3.4%

3.3%

2.3%

Brazil

3.0%

3.0%

3.5%

Russia

2.9%

3.9%

3.5%

Canada

2.5%

3.1%

2.2%

India

2.4%

6.9%

6.7%

Australia

2.0%

3.7%

3.0%

Notes: Percentage of world economy is based on IMF estimates of nominal GDP. Pre-recession trend is based on average annualized real GDP growth from the IMF. Long-run potential for the United States is defined as the average growth in 2018 of the Congressional Budget Office’s real potential GDP estimate; for other regions, it is defined as the 2018 growth rate from the IMF’s October 2013 WEO. Sources: Vanguard calculations, based on IMF and CBO data.

Europe In Europe, the broad economy is likely to grow at a positive albeit anemic pace after several years of deep, deleveraging-related recession (see Figure 3a on page 7). We believe that the euro is likely to survive intact, although a more vibrant and balanced European economy seems several years away.

Structurally, labor costs within Europe are continuing their painful internal adjustment as the peripheral economies aim to become more competitive relative to the core nations (see Figure 3b on page 7). We believe it is unlikely that the European economy as a whole will grow sustainably above 1% in the near future because significant deflationary competitive

adjustment is still occurring on the periphery. Growth is likely to be held back by fiscal restructuring and banking-sector deleveraging, although this should ease in the next few years. Unemployment rates in some peripheral economies of more than 20%, particularly for younger workers, present a moderate risk of social unrest leading to political instability. Although we expect Europe to continue meandering through its quandaries with a modest acceleration in growth over the next few years, investors should still prepare for periodic market volatility driven by flare-ups in political risk, concerns regarding the capitalization of the European banking system, and the potential for further debt write-downs, particularly in Greece.

5

Figure 2.

Debt and fiscal drag are significant factors affecting global growth

a. Debt levels remain high but vary by country and type Debt dashboard for selected economies as a percentage of GDP Central government

Households

Nonfinancial corporations

Financial institutions

United States Eurozone Germany France Italy Spain Netherlands Greece Ireland Portugal China Japan United Kingdom Canada Australia Less than 50%

Between 50% and 100%

More than 100%

Notes: We caution readers against making concrete assessments based on this general analysis, as the level of debt that a given country or sector may be able to sustain involves many factors. China’s government debt is shown on a gross basis; all others are net. Data show the latest available quarterly value in 2013 as a percentage of the trailing four-quarter average nominal GDP. Sources: Vanguard calculations, based on data from IMF, European Central Bank, Thomson Reuters Datastream, Moody’s Analytics, Australian Bureau of Statistics, Reserve Bank of Australia, Federal Reserve, Bank for International Settlements, and Bank of Japan.

b. Fiscal drag is expected to ease over the next two years Change in the pace of fiscal consolidation, 2012–2013 versus expected 2014–2015

Change in fiscal drag as annualized percentage of GDP

3%

Policy becoming less restrictive

Policy becoming more restrictive

2 1 0 –1 –2

Japan

China

Australia

United Kingdom

Canada

Ireland

Eurozone

Germany

France

Portugal

United States

Spain

Italy

Greece

Netherlands

–3

Notes: Figure displays the change in the pace of government deficit adjustment, measured as the difference in the change in the primary (excluding interest) structural government balance over the two years through 2013 relative to the expected change in the two years through 2015. Estimates for the next two years are from the IMF’s October 2013 WEO. Sources: Vanguard calculations, based on IMF data.

6

Figure 3.

In Europe, growth is expected to remain lackluster and divergent as imbalances slowly correct

a. Real GDP in the Eurozone

b. Unit labor costs in the Eurozone Cumulative change relative to Eurozone average since Q4 2000

102

Real GDP, Q1 2008=100

100 98 96 94 92 90 2005

2006

2007

2008

2009

2010

2011

2012

2013

15%

10

5

0

–5

–10 2001

2003

2005

2007

2009

2011

2013

CEPR recession Eurozone Eurozone core Eurozone periphery

Notes: “Core” is defined as the GDP-weighted average of changes in unit labor costs in Germany, France, and the Netherlands. “Periphery” is defined as the GDP-weighted average of changes in unit labor costs in Italy, Spain, Greece, Portugal, and Ireland. CEPR refers to the Centre for Economic Policy Research. Sources: Vanguard calculations, based on data from Moody’s DataBuffet.com, Thomson Reuters Datastream, France’s National Institute of Statistics and Economic Studies (INSEE), Deutsche Bundesbank, Statistics Netherlands, Ireland’s Central Statistics Office (CSO), Italian National Institute of Statistics (ISTAT), Hellenic Statistical Authority (ELSTAT), Instituto Nacional de Estatística–Portugal (Statistics Portugal), and European Commission: Eurostat.

Asia China is likely to grow at a 7% pace over the next two to three years, in line with market expectations but notably slower than its previous trend. Policymakers are attempting through structural supply-side reforms to strategically alter the country’s growth model, which heretofore has relied on investment and exports as its sole drivers. As illustrated in Figure 4a on page 8, investment currently represents a notably high share of Chinese GDP. However, capital-to-labor ratios are fairly low, meaning that although investment is running at a high rate, the country is starting from a low base in its capital stock. With capital per person at less than one-fifth that of the United States, much more investment is still needed.

The challenge for China is to ensure that such fastpaced investment spending flows toward the most productive uses of capital, avoiding misallocation and overinvestment in certain sectors. Some of the recently announced pro-market reforms are promising, because credit and investment will respond more to market signals (as would emerge with interest-rate liberalization) than to short-term policy targets or strict controls. However, the transition is not free of risks. Normal swings in market-driven investment and credit flows coupled with the current high weight of investment spending in GDP growth could easily cause a sharp economic slowdown. Gradual and flexible implementation of the reforms will be critical. Under a multiyear schedule, a growth guideline of about 7% to 7.5% for 2014 should be within reach.

7

Figure 4.

In Asia, China needs rebalancing as Japan attempts reflation

a. China’s investment spending: An unsustainable rate, but unlikely to soon collapse

$300,000

Investment as percentage of GDP

50%

250,000

40

200,000 30 150,000 20 100,000 10

50,000

0

0 China today

China average past decade

Asia Tigers at China’s current income levels

15 largest economies average today

Capital-to-labor ratio, constant 2005 dollars

Investment as percentage of GDP and capital-to-labor ratios for selected countries and time periods

United States today

Investment as percentage of GDP Capital-to-labor ratio

Notes: Investment as a percentage of GDP is from the IMF’s October 2013 WEO and April 2002 WEO for the stated time periods. “Today” is defined as the average for 2013. Asia Tigers comprise South Korea, Hong Kong, Taiwan, and Singapore when each was at China’s 2013 real per capita GDP level ($6,500 in 2013 U.S. dollars). The 15 largest economies are defined using 2013 GDP. Capital-to-labor ratio is from the Penn World Tables, in 2005 U.S. dollars, with “today” defined as the average for 2011. Sources: Vanguard calculations, using data from the IMF and Penn World Tables.

b. “Abenomics” in Japan: More reflation than real growth Japan’s historical and Abenomics real GDP growth and inflation 5%

Average annual rate

4

4.8%

3.6%

3

1.9%

2

1

0.8%

1.0%

0.0%

0 1975–1991

1992–2012

Successful Abenomics 2013–2018

Inflation Real GDP growth

Notes: Figure assumes that Abenomics achieves its goals of 2% inflation and 3% nominal GDP per capita growth by 2015. Transition assumes IMF’s WEO October 2013 baseline is realized in 2013–2014. Sources: Vanguard calculations, based on data from Thomson Reuters Datastream and IMF.

8

Japan’s aggressive Abenomics initiative aimed at jump-starting its economy is promising, although real growth will be challenged to exceed 1% for some time. The main goal of Abenomics seems to be reflation of prices rather than real economic growth. Even assuming full success of its policies,2 by 2015 Japan would be growing at a pace not much higher than its average for the last 20 years. However, if successful, Abenomics would bring inflation to a significantly higher level than it has been in the recent past (see Figure 4b). This would be an encouraging development, although as displayed in Figure 4b, it will be a difficult task after two decades of deflation. United States As in past outlooks, we maintain that U.S. trend growth (in terms of real GDP) is near 2%, versus its historical average of 3.0% to 3.5% since 1947. This projection is based on several structural headwinds, including slower labor force and population growth, the potential for continued consumer deleveraging, and higher levels of structural unemployment and debt than was the case over the past three decades.3 Indeed, actual real GDP growth has averaged 2.3% since the recovery began in 2009, well below the experience in previous recoveries (see Figure 5a on page 10). Nevertheless, we believe that our U.S. economic outlook can best be described as one of resiliency rather than of secular stagnation, for the reasons outlined below.4

Significant progress has been made to date in reducing consumer debt (see Figure 5b on page 10). Although this debt may not reach more sustainable levels of 60%–70% of GDP until 2016 or so, lower interest rates to service it combined with rising stock and home values have substantially aided the transition to a “passive deleveraging” phase of the cycle (see Figure 5b). For growth, it is the pace of consumer deleveraging that matters most, not the absolute level of debt outstanding, and that pace is slowing. As a result, the consumer need not “lever up” and save less in order for the country to achieve stronger growth in 2014–2015. Rather, this can emerge from fewer drags and shocks hitting the economy. In fact, the U.S. private sector (real GDP excluding the government sector) grew at a 2.9% pace in the four quarters through September 2013. Over the next two to three years, further acceleration of business investment (potentially driven by an unleashing of so-called animal spirits) is critical if U.S. economic growth is to exceed its recent threshold. The health of corporate balance sheets and profit margins (see Figure 5c on page 11) indicate that this is feasible, albeit not assured. The biggest risk is that policy uncertainty spikes and again exerts an “uncertainty tax” on the economy (similar to that surrounding, say, the debt-ceiling debate in mid-2011). Figure 5d, on page 11, shows that lower levels of uncertainty may be associated with a further pickup in capital spending over the next 12 to 18 months.

2 Japanese Prime Minister Shinzo Abe has announced targets of 3% growth of nominal income per capita and 2% inflation by 2015. 3 For more on the prospects for U.S. growth over the next decade, see Davis (2012) on a potential ”third industrial revolution.” In that presentation, we discuss how the factors that could contribute to better-than-expected growth may be driven by marked increases in capital investment. This could be sparked by the widespread adoption of cost-saving technologies, increased housing and infrastructure spending after a prolonged period of depressed activity, substantial U.S. energy independence from rising domestic production, and a lower trade deficit. 4 See Summers (2013) and Gordon (2012) for in-depth discussions supporting the view of secular stagnation.

9

Figure 5.

In the United States, a modest recovery may be set to accelerate

a. Deleveraging and fiscal belt-tightening have led to a modest recovery

Contribution to annualized real GDP growth

Contribution to growth in real U.S. GDP 4.5% 4.0%

4.0 3.5 3.0 2.5

2.5%

2.3%

2.0

1.5%

1.5 1.0

0.7%

0.5%

0.5

0.6%

0.4%

0.2%

0.0

Real GDP

Consumer spending

–0.1%

–0.2%

–0.3%

–0.5

Government spending

Business investment

Residential investment

Net trade

Average of recoveries since 1950s Current recovery, Q2 2009 to Q3 2013

Sources: Vanguard calculations, based on data from U.S. Bureau of Economic Analysis.

b. Consumer deleveraging is past the worst, but not over Household debt as a percentage of GDP and of assets

Household debt as percentage of GDP

18 16

75 Potential equilibrium

14 12

50

10 8

25 1960

1965

1970

1975

1980

1985

1990

1995

2000

2005

Percentage of GDP Percentage of household assets

Sources: Vanguard calculations, based on data from Moody’s DataBuffet.com, Federal Reserve, and U.S. Bureau of Economic Analysis.

10 

2010

Household debt as percentage of household assets

20%

100%

Figure 5.

In the United States, a modest recovery may be set to accelerate continued

c. Businesses are positioned to expand and hire . . .

11%

5.5%

10

5

9

4.5

8

4

7

3.5

6

3

5

2.5

4

Corporate cash holdings as percentage of total corporate assets

Corporate after-tax profits as percentage of GDP

Corporate profits and cash balances

2 1970

1975

1980

1985

1990

1995

2000

2005

2010

Corporate after-tax profits as percentage of GDP Corporate cash holdings as percentage of total assets

Sources: Vanguard calculations, based on data from Moody’s DataBuffet.com, Federal Reserve, U.S. Bureau of Economic Analysis, and National Bureau of Economic Research.

d. . . . as long as uncertainty doesn’t crimp confidence

200

15%

175

14

150 13 125 12 100 11

75

50 1985

10 1988

1991

1994

1997

2000

2003

2006

2009

Fixed business investment as percentage of GDP

Policy Uncertainty Index, 12-month moving average

Policy Uncertainty Index and business investment as a percentage of GDP

2012

Policy Uncertainty Index Fixed business investment

Note: For additional information on the Policy Uncertainty Index, including the methodology behind its construction, see Baker, Bloom, and Davis (2012). Sources: Vanguard calculations, based on data from Moody's DataBuffet.com, U.S. Bureau of Economic Analysis, National Bureau of Economic Research, and Baker, Bloom, and Davis (2012).

11

2014 U.S. economic outlook: improving odds of above 2% growth After years of disappointing 2% real growth, the U.S. in 2014–2015 faces cyclical risks tilted toward betterthan-trend growth for the first time since the onset of the global financial crisis. As evident in Figure 6a, our proprietary U.S. leading indicators dashboard points toward a slight acceleration. The most positive leading indicators are those associated with the housing market, manufacturing activity, and financial conditions (especially those tied to the stock market). The “red signals,” associated with credit growth, confidence, and excess capacity, exemplify the lingering effects of the global financial crisis.

Figure 6.

Using simple regression analysis, we can map our proprietary set of indicators to a distribution of potential scenarios for U.S. economic growth as shown in Figure 6b. We estimate a 25% probability that U.S. real GDP growth preserves the status quo and averages 1.5%–2.5% in 2014. The odds of growth exceeding 2.5% in 2014 (48%) are roughly double that of the potential for it to stagnate and fall below 1.5% (27%). Our base case is a cyclical rebound in 2014, with growth in real GDP of close to 3% on average over the course of the year.

Odds favor an acceleration in U.S. growth in 2014

a. Vanguard’s dashboard of leading economic indicators

b. Vanguard’s 2014 U.S. economic growth outlook Estimated distribution of growth outcomes 35% 30

75

Odds of a slowdown

27%

50

Trailing three-year growth

Odds of an acceleration

2.2% 25%

25 Probability

Share of indicators above/below trend

100%

48%

25%

20 16%

16%

15 11%

25

10 7% 5

0 2000

2003

2006

2009

2012

Above-trend growth Housing Financial conditions Stock market Real rates Manufacturing Below trend but positive momentum Global trade Business loans Below trend and negative momentum Confidence Lending demand Excess capacity Government

Sources: Vanguard calculations, based on data from Moody’s DataBuffet.com.

0 Recession: Less than 0.5% Stagnation: 0.5% to 1.5% Status quo: 1.5% to 2.5% Cyclical rebound: 2.5% to 3.5% Robust above-trend growth: 3.5% to 4.5% Overheating: Above 4.5%

Historical real GDP growth 1926–2013

3.3%

2000–2013

1.8%

Past three years

2.2%

2014 consensus

2.8%

Notes: Distribution of growth outcomes generated by bootstrapping the residuals from a regression based on a proprietary set of leading economic indicators and historical data, estimated from 1960 to 2013 and adjusting for the time-varying trend growth rate. The 2014 consensus is from the Federal Reserve Bank of Philadelphia’s Q4 2013 Survey of Professional Forecasters. Sources: Vanguard calculations, based on data from U.S. Bureau of Economic Analysis and Federal Reserve.

12 

Figure 7.

Monetary policy of unprecedented size and scope to avert the prospect of global deflation

a. Global central bank assets as a percentage of a region’s 2008 GDP

b. Core inflation across key central bank markets, 2001–2013 10%

Onset of global financial crisis

Year-on-year percentage change

Total assets (percentage of 2008 GDP)

50%

40

30

20

10

0 2007

2008

2009

2010

2011

2012

2013

Federal Reserve European Central Bank Bank of Japan Bank of England

Onset of global financial crisis 8 6 4 2 0 –2

2001

2003

2005

2007

United States Japan United Kingdom Eurozone

2009

2011

2013

Typical range of developed market central bank inflation targets (1%–3%)

Note: Total assets for each central bank are shown as a percentage of that country’s or region’s 2008 GDP.

Note: Figure displays the year-on-year percentage change in each country’s or region’s core (excluding food and energy) consumer price index.

Sources: Vanguard calculations, based on data from Federal Reserve, Bank of England, European Central Bank, Bank of Japan, and IMF.

Sources: Vanguard calculations, based on data from U.S. Bureau of Labor Statistics, Statistics Bureau of Japan, U.K. Office for National Statistics, and Eurostat.

Outlook for inflation

regime is low, despite the size of central bank balance sheets; in parts of Europe and in Japan, the specter of deflation remains a greater risk.

In the near term, reflationary monetary policies will continue to counteract the deflationary bias of a highdebt world still recovering from a deep recession. Although central bank balance sheets have risen to a combined total of more than $8 trillion since the onset of the financial crisis (see Figure 7a), core inflation trends are low (see Figure 7b). Indeed, recent consumer price inflation remains near generational lows and, in several major economies, below the targeted rate. In the United States, the year-over-year CPI inflation rate lingered at about 1% as of late 2013. As stated in previous Vanguard outlooks, trend inflationary pressures in the United States and most other developed markets are modest. The recent patterns in key drivers such as labor costs, economic slack, commodity and import prices, and the velocity of money suggest that core U.S. inflation is likely to stay within its recent range of 1%–3% for the next one to two years (see Figure 8 on page 14). For now, the risk of returning to a high inflationary

For the next ten years, our VCMM simulations project a median inflation rate averaging close to 2.0%–2.5% per year for the U.S. Consumer Price Index (see Figure 9 on page 15). This is roughly consistent with the Federal Reserve’s long-term goal of inflation stability and is also near longer-term break-even rates in the Treasury Inflation-Protected Securities (TIPS) market. Of note, Vanguard’s median secular inflation expectation is approximately 1% lower than the average U.S. CPI inflation rate observed since 1950. All else being equal, this implies that nominal assetclass returns may be 1% lower than historical longrun averages, even if their expected average real (inflation-adjusted) returns are identical. We discuss this point further in the section on stocks, bonds, and asset allocation strategies.

13

Figure 8.

U.S. drivers point to modest inflation

Inflation indicators, long-term average versus recent value 10 8 6 4 2 0 –2 –4 –6 Money multiplier

Credit growth

Disposable income growth

Nominal GDP growth

U.S. import prices

Commodity prices

Output gap/slack (%)

Long-run average Latest value

Notes: Money multiplier is the ratio of M2 money supply to M0, or base money; latest value is October 2013, and average is from January 1959. Credit growth is year-on-year percentage change of the 36-month trailing average level; latest value is October 2013, and average is from January 1960. Disposable income growth is year-on-year percentage change in the three-month trailing average level; latest value is October 2013, and average is from January 1960. Nominal GDP growth is year-on-year percentage change; latest value is Q3 2013, and average is from Q1 1950. Import prices are year-on-year percentage change from the National Accounts; latest value is Q3 2013, and average is from January 1960. Commodity prices are year-on-year percentage change in the Commodity Research Bureau (CRB) spot commodity index; latest value is November 2013, and average is from January 1960. Output gap is percentage deviation from potential GDP based on an average of CBO estimates and a statistical filter of real GDP; latest value is Q3 2013, and average is set at 0%. Sources: Vanguard calculations, based on data from U.S. Bureau of Economic Analysis, Federal Reserve, CRB, and CBO.

Looking ahead, we continue to believe that the countervailing forces of fiscal deleveraging and monetary-policy reflation in the United States and Europe will reinforce an “inflation paradox.” On the one hand, we expect that some investors will continue to have significant concerns about future inflation. We estimate a nearly one-third probability that the trend inflation rate runs above 3% over the next ten years. As a result, conversations about portfolio construction will include much discussion about inflation protection and the performance of various asset classes under expected and unexpected scenarios (Davis, AliagaDíaz, Thomas, and Zahm, 2012).

14 

On the other hand, monetary policymakers in developed markets are likely to continue to guard against the pernicious deflationary forces of debt deleveraging for an extended period. Indeed, our VCMM simulations reveal that the prospects of secular Japan-style deflation (in which the average CPI inflation rate over the next decade is –1% or less) are approximately 10%. It is worth emphasizing that despite aggressive monetary policy, some developed markets would be a recession away from realizing deflation.

Global monetary policy has been extremely aggressive over the past several years. Central banks have lowered short-term interest rates close to zero and, with their traditional policy tool thus constrained, expanded the size and composition of their balance sheets.5 At the same time, they have provided a new level of communication about the length of time markets can expect short-term rates to remain unchanged. The Bank of Japan, the Bank of England, and the European Central Bank have all, to some extent and in various forms, adopted policies similar to those pursued by the U.S. Federal Reserve. Based on our economic outlook, it seems reasonable that the Fed was the first of those four banks to begin to exit these very accommodative policies.

Figure 9.

Projected U.S. CPI inflation rate, current and 2012 ten-year outlooks

VCMM-simulated distribution of annualized expected average 40%

38%38%

30 Probability

Outlook for U.S. monetary policy and interest rates

20 15%

17% 14% 14%

15%

17% 14%

16%

10

0 Less than 0%

0% to 1% 1% to 3% 3% to 4%

Greater than 4%

Ten-year annualized CPI inflation rate

Tapering of the QE program has begun, although actual monetary tightening is likely some time off. The unwinding of the Fed’s extraordinarily easy monetary policy will be a multistep, multiyear process, with the initial taper only the first in a series of actions on the road to normalization. Stronger forward guidance will likely be used, for example by pairing the 6.5% unemployment threshold with, perhaps, a “floor rate” for core inflation or by lowering the unemployment threshold itself. New policy tools such as the interest paid on excess reserves and reverse repo operations are also likely to make an appearance. We believe such actions may be effective at the margin in anchoring long-term interest rates, but ultimately the future pattern of job growth and the unemployment rate will dictate when the Fed raises short-term rates, probably sometime in 2015.

Current ten-year outlook Ten-year outlook as of year-end 2012

Median inflation 1950–2013

3.0%

Median inflation 2000–2013

2.5%

10-year TIPS break-even inflation as of November 29, 2013

2.1%

Sources: Vanguard calculations, based on data from U.S. Bureau of Labor Statistics.

5 Balance sheet expansion in most large central banks has occurred through quantitative easing—the purchase of longer-term securities, typically government bonds, financed by creating bank reserves. This policy is intended to influence longer-term interest rates when the traditional short-term policy rate is near 0%. The European Central Bank is an exception—its balance sheet expansion occurred through policies designed to promote liquidity in the banking system, referred to as long-term refinancing operations, or LTROs.

15

Figure 10.

Questions about structural unemployment are a wild card for the Fed’s exit

a. Demographics and structural change cast doubt on the headline unemployment rate

b. Labor force participation has fallen, but will it improve as the recovery continues?

Unemployment rate under hypothetical scenarios for labor force participation

Change in labor force participation rates across demographic groups, 2007–2013

13%

2% 1.1% Change in labor force participation rate since December 2007

12

Unemployment rate

11 10 9 8 7 6 5

1 0 –1 –2 –2.2% –3

–3.0%

–4

4 2007

2008

2009

2010

2011

2012

2013

Actual unemployment Assuming labor force participation rate constant at December 2007 levels Constant labor force participation, adjusted for aging population

Notes: Figure displays actual unemployment rate along with two adjusted measures. The first assumes the labor force participation rate stays constant at the December 2007 level of 66%, with any labor force dropouts being added to the unemployment rate calculation. The second assumes labor force participation is held at December 2007 levels but controls for the impact of demographics, with workers shifting to and from age groups with different participation rates.

–5

Overall

–4.6% 16–24

25–54

55+

Ages

Note: Figure displays change in the labor force participation rate from December 2007 to November 2013 for the stated age groups. Sources: Vanguard calculations, based on data from U.S. Bureau of Labor Statistics.

Sources: Vanguard calculations, based on data from U.S. Bureau of Labor Statistics and Census Bureau.

When will the Fed raise rates? The timing depends on two critical labor-market issues: (1) how quickly the actual unemployment rate falls, and (2) why it falls.

As illustrated in Figure 10a, the unemployment rate may be understated because of the severe drop in the labor force participation rate, as illustrated in Figure 10b. We find it unlikely that this figure is driven solely by an aging population (i.e., the aging and retirement of the baby boomers). In our estimates, approximately one-half of the reduction in the working-age-adjusted labor force is due to cyclical factors (e.g., weak demand, return

16 

to schooling) rather than more permanent, structural changes (e.g., skill mismatches, demographics). Put another way, the “real” unemployment rate, corrected for the cyclical drop in participation, is likely somewhere between the blue and purple lines in Figure 10a, closer to 8.0%–8.5% than to the official 7.0% rate (as of November 2013). By extension, this implies that the federal funds rate will remain near 0% through mid-2015. However, as noted in Figure 11, the risk that this lift-off date is further delayed to, say, 2016 is lower than it has been in previous years. Correct calibration of monetary policy based on current labor market

Figure 11.

Handicapping Fed tightening and the end of financial repression

VCMM-simulated probabilities of the level of nominal and real 3-month Treasury bill yields 100% 90 75.8%

80

65.4%

Probability

70

57.5%

60

51.8%

50

50.2% 44.4%

40.1%

40

31.7% 25.3%

30

21.2%

20 10 0

One year out

Two years out

Three years out

Four years out

Five years out

Probability of T-bill rate near 0% Probability of T-bill rate below inflation

Note: Figure displays VCMM-projected probability that the 3-month Treasury bill yield will be less than 0.5%, or less than the 12-month trailing inflation rate at the end of the stated year. Source: Vanguard.

conditions hinges on an accurate portrayal of labor market slack. Wage growth and inflation expectations are critical indicators in light of the marked (and partly cyclical) drop in labor force participation.6 Whether or not the Fed raises short-term rates in 2015, real (inflation-adjusted) short-term interest rates are likely to remain negative through perhaps 2017. Globally, the burdens on monetary policymakers are high as they contemplate moving away from QE policies to prevent asset bubbles on one hand and remain mindful of raising short-term rates too aggressively on the other. The unwinding process may induce market volatility at times, but long-term investors should prefer that to no exit at all.

U.S. Treasury yield curve The bond market continues to expect Treasury yields to rise, with a bias toward a steeper curve until the Federal Reserve raises short-term rates. Compared with last year’s outlook, our estimates of the fair value range for the 10-year Treasury bond have risen, with the current macroeconomic environment justifying a 10-year yield in the range of 2.75% to 3.75% (see Figure 12a on page 18). Based on our estimates of the fundamental drivers of Treasury bond yields, we are hard pressed to find a bubble in Treasury securities. With the recent rise in longterm interest rates since summer 2013, we find that bond yields are toward the middle of the range of our fair-value estimates.

6 Broad measures of wage growth currently are near 2%; growth above 3% would imply that the amount of slack in the economy is overestimated (i.e., the nonaccelerating inflation rate of unemployment [NAIRU] is higher).

17

Think twice before adjusting duration to avoid a bubble in Treasuries

a. Fair value factors suggest that long-term Treasury yields are reasonable, with the market already pricing an increase Vanguard’s fair value model of the 10-year Treasury yield, with forward-inferred and VCMM-simulated projections 12%

10-year Treasury bill yield

10 8 6 4 2 0 1990

1995

2000

2005

2010

2015

Fair value range Actual Forward curve (November 2013) VCMM 50th percentile VCMM 25th/75th percentile

b. Short-duration tilts involve giving up significant income Difference in yield between a broad-market and a short-duration U.S. bond investment

Difference in yield, broad bond market versus short-duration bonds

Figure 12.

2.5%

2.0

1.5

1.0

0.5

0 1995

1998

2001

2004

2007

2010

2013

Note: Figure displays the difference in yield to worst between the Barclays U.S. Aggregate Bond Index and the Barclays U.S. Aggregate 1–3 Year Index. Sources: Vanguard calculations, based on data from Barclays.

Notes: Historical fair value is based on a model derived from Warnock and Warnock (2009) and includes expected inflation, expected real GDP growth, expectations regarding monetary and fiscal policy, and domestic and foreign capital flows. Range reflects standard error margin of plus or minus 0.5 percentage points. Forward curve is derived from the Federal Reserve data set provided by Gürkaynak, Sack, and Wright (2006). VCMM projections reflect data through November 2013. Sources: Vanguard calculations, based on data from the Federal Reserve.

The future path for interest rates, including our fair value estimates, will likely change over time in response to growth, inflation, and monetary conditions. In our simulations, the bias is toward a rise over the next several years, a view that is consistent with the forward market and therefore reflected in today’s bond prices.7 We continue to believe that a bond bear market in which rates move toward 5% based on stronger growth, inflation, and monetary tightening may be several years away. We estimate the odds of a U.S. fiscal crisis and sharp spike in yields at less than 10% at the moment; these odds rise later in the decade based on the expected trajectory of federal debt.

In the long run, short-term rates tend to rise more than long-term rates in substantially more than onehalf of our VCMM scenarios. This so-called “bear flattening” of the yield curve is typical in a Fed tightening cycle with stable inflation expectations. This has important implications for those inclined to strategically tilt the duration exposure of their bond portfolios away from that of the broad fixed income market. In a Fed tightening cycle, the prospects for near-term losses in short-term bond portfolios are elevated as well. A short-duration strategy entails forgone income (see Figure 12b). Focusing solely on avoiding capital losses ignores the fact that a steep yield curve produces significant income differences

7 For further detail on the forward curve and the implications of the bond market pricing forward interest rate expectations, see Davis et al. (2010).

18 

Figure 13.

Projected global fixed income ten-year return outlook

VCMM-simulated distribution of expected average annualized return of the global fixed income market, estimated as of year-end 2013 and 2012 30% Global bond returns 1926–2013 5.4% 1926–1969 3.1% 7.7% 1970–2013 2001–2013 4.6%

25

Probability

20 15 10 5 0 Less than 0.5%

0.5% to 1%

1% to 1.5%

1.5% to 2%

2% to 2.5%

2.5% to 3%

3% to 3.5%

More than 3.5%

Ten-year annualized return Current ten-year outlook Outlook as of year-end 2012

Notes: Figure displays projected range of returns for a portfolio of 80% U.S. bonds and 20% ex-U.S. bonds, rebalanced monthly. Benchmarks used for historical returns are defined on page 4. Source: Vanguard.

among duration strategies. In other words, “going short duration” may not necessarily outperform a broadly diversified fixed income portfolio in the years ahead.

Asset-class outlook: Bonds Expected range of returns for the broad taxable bond market As in past editions, the return outlook for fixed income is muted, although it has improved somewhat with the recent rise in real rates. As displayed in Figure 13, the expected ten-year median return of the broad taxable U.S. fixed income market is centered in the 1.5%–3.0% range, as opposed to last year’s expected

0.5%–2.0% range. This return is near current benchmark yields and thus most closely resembles the historical bond returns of the 1950s and 1960s. Expected diversification effects We expect the diversification benefits of fixed income in a balanced portfolio to persist under most scenarios. Although yields in most developed markets are at historically low levels, diversification through exposure to hedged non-U.S. dollar-denominated bonds should help offset some of the risks specific to the U.S. fixed income market. Less-than-perfect correlation between two of the main drivers of bond returns— interest rates and inflation—is one potential benefit.8

8 For additional details, see Philips et al. (2012) and Philips and Thomas (2013).

19

Figure 14.

Bonds can provide ballast in an equity bear market

Median return of various asset classes during the worst decile of monthly equity returns, 1988–2012 2%

0

Median return

–2

–4

–6

–8

–10 U.S. stocks

Emergingmarket stocks

REITs

Dividend Commod- High-yield Emergingstocks ities bonds market bonds

Hedge funds

Corporate bonds

Treasury bonds

InterInternational national bonds bonds (unhedged) (hedged)

Notes: U.S. stocks, U.S. bonds, and international bonds represented by indexes listed on page 4. Emerging-market stocks represented by FTSE Emerging Index and emerging-market bonds by Barclays Emerging Markets Tradable USD Sovereign Bond Index. REITs represented by FTSE NAREIT Equity REIT Index, dividend stocks by Dow Jones U.S. Select Dividend Total Return Index, commodities by S&P GSCI Commodity Index, high-yield bonds by Barclays U.S. Corporate High Yield Index, hedge funds by median hedge fund-of-funds return as identified by Morningstar, Inc., corporate bonds by Barclays U.S. Corporate Investment Grade Index, and Treasury bonds by Barclays U.S. Treasury Index. Sources: Vanguard calculations, based on data from S&P, Citigroup, Barclays, Dow Jones, MSCI, CRSP, and FTSE.

We encourage investors to evaluate the role of fixed income from a perspective of balance and diversification rather than outright return. High-grade or investment-grade bonds act as ballast in a portfolio, buffering losses in riskier assets.9 Figure 14 shows how these bonds have performed during the most significant equity downturns of the past 25 years. They may very well not provide the same magnitude of benefit during periods of flight-to-quality when interest rates are low.10 However, as shown in Figure 15, the

prospects of significant near-term annual losses of –5% or more are higher for equities than they are for investment-grade bonds. Therefore, investors should approach with caution any decision to replace bonds with riskier assets.11 Although history is not necessarily indicative of future results, 200 instances of rolling 12-month losses of 5% or more occurred in a portfolio of global equities since 1926, compared with 38 in bonds (see Figure 15).

9 We define high-grade or investment-grade bonds as those fixed income securities rated Baa3 and above by Moody’s. 10 For additional details, see Kinniry and Scott (2013). 11 For additional details, see Davis (2013b).

20 

Figure 15.

Focus on the potential magnitude of losses

Relative risk of loss in equities and bonds over a rolling 12-month period 200%

Rolling 12-month return

150

100

50

0

–50

–100

1926

1932

1938

1944

1950

1956

1962

1968

1974

1980

1986

1992

1998

2004

2010

Equities Bonds

Note: Benchmarks used for historical returns are defined on page 4. Sources: Vanguard calculations, based on data from S&P, Citigroup, Barclays, Dow Jones, MSCI, CRSP, and FTSE.

Corporate bonds and TIPS The median expected total return of an investmentgrade corporate bond index in our VCMM scenarios modestly exceeds that of a similar-duration U.S. government bond portfolio. This expected positive risk premium, a function of the current level of corporate bond spreads, is not realized in all scenarios because of corporate bonds’ sensitivity to credit risk.

The probability of realizing a positive “spread return” in investment-grade or high-yield corporate bonds has decreased in the past few years as yield spreads over less risky U.S. Treasury bonds have narrowed (see Figure 16 on page 22). At the same time, real Treasury yields have increased, particularly during the past summer. This indicates that the payoff for tilting bond portfolios to riskier segments of the market is lower than was the case, say, two or three years ago.

21

Figure 16.

Treasury yields are up, and corporate spreads have compressed

Treasury and corporate bond yields for selected time periods 16% High risk premiums

14 12

Yield

10 8 6

6.6% Falling risk premiums

Low risk premiums 2.5% 1.0%

4.2%

4 2.4%

2

1.3%

1.7% 1.7%

2.0%

1.7%

Prerecession 2004–2007 average

Recession 2008–2009 average

0

2.8%

3.4%

2.2%

2.2% 0.6%

0.4% Recovery 2010–2012 average

Current

Real yield Break-even inflation Investment-grade corporate spread High-yield corporate spread

Notes: Real yield is defined as the yield of the active 10-year TIPS bond. Break-even inflation is the difference between the active 10-year nominal Treasury and the 10-year TIPS bond. Investment-grade corporate spread is the option-adjusted spread (OAS) of the Barclays U.S. Investment-Grade Corporate Index. High-yield corporate spread is the difference between the OAS of the Barclays U.S. High Yield Corporate Index and the Barclays U.S. Investment-Grade Corporate Index. “Current” is as of November 2013. Sources: Vanguard calculations, based on data from Federal Reserve and Barclays.

In the inflation-linked segment of the bond market, the distribution in our VCMM scenarios of TIPS returns is wider than that of nominal Treasury bonds. The expected median long-term return on a U.S. TIPS portfolio is lower than that of a similarduration nominal Treasury portfolio by a modest amount that represents the estimated inflation risk premium. As would be expected, TIPS generally outperform nominal Treasuries in scenarios

22 

featuring higher-than-average inflation rates over a ten-year outlook. On a more cautionary note, TIPS have displayed a higher probability of negative returns over shorter investment horizons because of their sensitivity to a rise in real rates. Balancing these considerations, investors should continue to evaluate the role of TIPS in providing protection against inflation risk—that is, the possibility of higher-than-expected inflation.

Figure 17.

Projected global equity ten-year return outlook

VCMM-simulated distribution of expected average annualized return of the global equity market, estimated as of year-end 2013 and 2012 25% Shift in distribution of outcomes relative to last year

Probability

20

15

10

5

0 Less than 0%

0% to 3%

3% to 6%

6% to 9%

9% to 12%

12% to 15%

15% to 18%

More than 18%

Ten-year annualized return Current ten-year outlook Outlook as of year-end 2012

Historical global equity returns 1926–2013

10.2%

1926–1969

9.7%

1970–2013

10.6%

2001–2013

5.6%

Notes: Figure displays the projected range of returns for a 70% U.S., 30% ex-U.S. equity portfolio, rebalanced monthly. Benchmarks used for historical returns are defined on page 4. Source: Vanguard.

Asset class outlook: Global equities Centered in the 6%–9% range, the long-term median expected return for global equity markets is moderately below the historical average and revised downward from this time last year, mainly because of current market valuations and their implications for the equity risk premium (see Figure 18 on page 24). Similar to the situation in riskier segments of the bond market, the premium compensating risk in the equity market appears to have fallen recently. When returns are adjusted for future inflation, we estimate a roughly 40% likelihood that a global equity portfolio will fail to produce a 5% average real return over the decade 2013–2023.

Our return outlook is informed by valuation metrics (such as price/earnings ratios) that relate accounting measures of value to the market’s aggregate price. Valuations today are elevated in relation to both their lows in 2009 and their historical averages (see Figure 18a on page 24). Although Figure 18b (on page 24) shows some divergence across regions, we caution investors against implementing tactical tilts based on this. Historically, emerging-market stocks have tended to possess lower relative market valuations in recognition of their higher perceived investment risk, and divergences today are less pronounced than those in the mid-2000s that led to large return differentials. To account for this, we aggregate our global return outlook in Figure 17. The expected central ranges of long-run returns on various regional equity investments are statistically similar to one another, especially after accounting for differences in expected volatility.

23

Figure 18.

Some differences exist across metrics and regions, but valuations are generally elevated

a. Most developed-market valuation metrics are above long-term averages

b. Regional valuations are more similar today than in the middle of the previous decade

Valuation metrics for the U.S. equity market relative to historical average value

Price over 36-month trailing earnings for selected global equity indexes 60

8 50

7 6 Price/earnings ratio

Standard deviations from long-term mean

9

5 4 3 2 1

40 30 20

0 10

–1 –2 –3 1927

1942

1956

1970

Broad market price/earnings Broad market price/book Broad market price/sales

1985

1999

2013

Shiller CAPE (3-year) Shiller CAPE (10-year)

Notes: Figure displays valuation metrics standardized to have a long-term average of 0.0 and a standard deviation of 1.0. Broad market price/earnings displays the market value of domestic corporations from the Federal Reserve Flow of Funds database relative to the trailing four-quarter average of after-tax corporate profits from the BEA’s national accounts. Broad market price/sales displays the market value of domestic corporations from the Flow of Funds database relative to the Gross Value Added of Corporate Business from the BEA’s national accounts. Broad market price/book displays the market value of domestic corporations relative to the net worth at historical cost of Nonfinancial Corporate Business, both from the Flow of Funds database. Shiller CAPE (10-year) is the ten-year cyclically adjusted price/earnings ratio as defined in Shiller (2000). Shiller CAPE (3-year) is Shiller’s measure, adjusted to smooth earnings over a trailing 36-month period.

0 2003

2005

2007

2009

2011

2013

Emerging markets United States Developed markets ex-U.S.

Notes: Figure displays the price/earnings ratio with 36-month trailing average earnings. United States is defined as the FTSE United States Index, developed markets ex-U.S. are defined as the FTSE All-World Developed ex US Index, and emerging markets are defined as the FTSE All-World Emerging Markets Index. Sources: Vanguard calculations, based on data from FTSE.

Sources: Vanguard calculations, based on data from Federal Reserve, U.S. Bureau of Economic Analysis, and Robert Shiller’s website, aida.wss.yale.edu/~shiller/data.htm.

Of note, the projected distribution of annualized ten-year global equity returns shown in Figure 17 on page 23 displays wide and fat tails. As discussed in Davis, Aliaga-Díaz, and Thomas (2012), valuations are the most useful metric in estimating forward-looking expected returns of equity markets. However, they still leave more than half the volatility of long-run returns unexplained. Although we emphasize a focus on the wide distribution, we note that the central tendency of our projected returns has come down

24 

since last year, reflecting the valuation levels shown in Figures 18a and 18b. Figure 19 displays the historical relationship between U.S. valuations and subsequent real ten-year returns. The results underscore the fact that today’s valuation levels have been associated with lower average returns but with a significant range around this average. Indeed, every valuation bucket has been associated with subsequent real returns near the historical average in at least some time periods.

Figure 19.

Today’s valuation levels have been associated with modest average returns but a wide range of possible outcomes

Distribution of future ten-year real equity returns

Initial cyclically adjusted price/earnings ratio and subsequent range of ten-year real U.S. equity returns, 1926–2013 20%

19% 16%

15

10

14%

Current valuation levels

5

12%

10%

10%

5% 3% 1%

0%

9%

8%

8%

8%

4%

0

Percentiles:

14% 12%

8%

5%

95th

75th

Median

3%

0%

–2%

Key

16%

25th

–2% –5

5th

–4% Top

2nd

Middle

4th

Bottom

Quintile ranking of initial Shiller CAPE (10-year) U.S. average real return = 7%

Notes: Figure displays the initial valuation quintile of the S&P 500 Index, defined as the ten-year cyclically adjusted price/earnings ratio (CAPE) as defined by Shiller (2000), with the 5th/25th/50th/75th/95th percentile range of subsequent realized ten-year real returns on the U.S. equity market. Data represent January 1926–September 2013. U.S. equities are defined as on page 4, deflated by U.S. consumer price inflation. Sources: Vanguard calculations, based on data from Robert Shiller’s website, aida.wss.yale.edu/~shiller/data.htm; S&P; Dow Jones; MSCI; and U.S. Bureau of Labor Statistics.

The results in Figure 19 are driven by the tendency of valuations to revert to a long-term average level, at least over this sample. In a simple return decomposition, this reversion has been the largest driver of the movement of long-term equity returns over time (see Figure 20 on page 26). This supports our long-held view that valuations, not growth, are the most significant drivers of returns (Davis et al., 2013). But to what level will valuations revert? As shown in Figure 18a, most valuation metrics have been above their long-term averages for more than two decades, raising questions about the potential for structural shifts. Without certainty as to where exactly valuations will move in the future, it is very

difficult to pin down a precise estimate of the equity risk premium. This is a key reason for our distributional approach to forecasting. In short, although we are hard-pressed to identify market bubbles, there is some evidence of froth in global equity markets. The uncertainty associated with forward-looking return estimates underscores the fact that today’s valuation levels present a range of potential outcomes. However, because the premium compensating increased equity risk appears to have fallen recently, we would encourage investors to exercise caution in making strategic or tactical portfolio changes that increase this risk.

25

Figure 20.

For reference, the figure also shows how the hypothetical portfolios would have performed over two past periods: 1926–2013 and 2000–2013. The results have several important implications for strategic asset allocation, as discussed next.12

Valuation movement is the largest component of returns, but predicting this is extremely difficult

Components of ten-year real U.S. equity returns

Components of ten-year real U.S. equity returns

15%

Modest outlook for long-run real returns Amid widespread concern over the current low level of dividend and long-term U.S. Treasury yields, Figure 21’s real long-run return profile for balanced portfolios may seem better than expected. However, Vanguard believes it’s important for investors to consider real-return expectations when constructing portfolios because today’s low dividend and Treasury yields are, in part, associated with lower expected inflation than those of 20 or 30 years ago.

10 5 0 –5 –10 –15 1935

1948

1961

1974

1987

2000

2013

Average dividend yield Earnings growth Valuation adjustment return

Notes: Figure displays the backward-looking return components of the S&P 500 Index. Dividend yield is the average trailing dividend yield of the S&P 500 Index, taking an average of each monthly observation of 12-month trailing yield in each ten-year time period. Earnings growth is the average annualized growth of the ten-year smoothed real earnings for the S&P 500 Index constituents. Valuation adjustment return is the annualized percentage change of the cyclically adjusted price/earnings ratio, as defined in Shiller (2000). Sources: Vanguard calculations, based on data from Robert Shiller’s website, aida.wss.yale.edu/~shiller/data.htm.

Implications for asset allocation strategies To examine the potential portfolio construction implications of Vanguard’s range of expected longrun returns, Figure 21 presents simulated real (inflation-adjusted) return distributions for 2013−2023 for three hypothetical portfolios ranging from more conservative to more aggressive:

The figure does show that the inflation-adjusted returns of a balanced portfolio for the decade ending 2023 are likely to be moderately below long-run historical averages (indicated by the small boxes for 1926−November 2013 and 2000–November 2013). But the likelihood of achieving real returns in excess of those since 2000 for all but the most conservative portfolios is considerably higher. Specifically, our VCMM simulations indicate that the average annualized returns of a 60% equity/40% bond portfolio for the decade ending 2023 are expected to center in the 3.1%–5.2% real-return range, below the actual average real return of 5.5% for the same portfolio since 1926. Viewed from another angle, the likelihood that our portfolio would achieve the 1926–2013 average real return is estimated at approximately 40%, and the odds of attaining a higher real return than that achieved since 2000 (2.1%) are near 70%.

• 20% equities/80% bonds. • 60% equities/40% bonds. • 80% equities/20% bonds.

12 See the Appendix for the range of returns in nominal terms before adjusting for inflation.

26 

Figure 21.

Projected ten-year real return outlook for balanced portfolios

VCMM-simulated distribution of expected average annualized inflation-adjusted return of balanced global equity and global fixed income portfolios, estimated as of year-end 2013 16%

Key

14

History, 1926−2013 History, 2000−2013

Average annualized return

12 10

95th percentile

8 6 4

25th−75th percentile

2 0 –2 –4 –6

5th percentile

20%/80%

60%/40%

80%/20%

Portfolio stock/bond allocation

Underlying data for this figure

25th percentile

0.0%

1.3%

1.7%

Notes: Figure displays the 5th/25th/75th/95th percentile range of VCMMprojected returns for balanced portfolios. Historical returns are computed using the indexes defined on page 4.

75th percentile

3.4%

7.1%

9.0%

Source: Vanguard.

Top 95th percentile

5.8%

11.4%

14.3%

Portfolio stock/bond allocation Bottom 5th percentile

Annualized portfolio volatility

20%/80%

60%/40%

80%/20%

−2.5%

−2.9%

−3.6%

6.2%

11.2%

14.4%

History, 1926−2013

3.5%

5.5%

6.3%

History, 2000−2013

2.4%

2.1%

1.8%

Principles of portfolio construction are intact Contrary to suggestions that this decade warrants some radically new investment strategy, Figure 21 reveals that the simulated ranges of expected returns are upward sloping. Simply put, higher risk accompanies higher (expected) return. More aggressive allocations have a higher—and wider— range of expected returns, with greater downside risk if the equity risk premium is not realized over the next decade.13 To put this in context, the risk premium over bonds may be lower than it has been in the past few years, but it is still positive. Indeed, the expected risk-return trade-offs among stocks and bonds show why the principles of portfolio construction remain unchanged, in our view, even if expected returns are lower.

The upward-sloping and wider-tailed pattern in Figure 21 reaffirms the beneficial role that bonds should be expected to play in a broadly diversified portfolio, despite their currently low yields and regardless of the future direction of interest rates. Although our scenarios generate slim, below-average nominal returns for a broad taxable bond index for the next ten years—a central tendency of 1.5%– 3.0% annually, on average—bonds should be expected to moderate the volatility in equity portfolios in the years ahead. Still, we are concerned that the low nominal rate environment may encourage savers and bond investors to very aggressively pursue higher nominal total returns by making investment decisions that

13 Although the downside tails may appear somewhat similar across the portfolios, we note that these are ten-year distributions. The downside risk for a more equity-oriented portfolio increases substantially over a shorter horizon, as demonstrated in Figure 15 on page 21.

27

increase risk, often based solely on asset-class yields rather than on a more holistic total return approach. Popular considerations at the moment include moving away from conservative bond portfolios and into either higher-yielding junk bonds or incomeoriented equity funds such as dividend-focused equity funds or REIT funds. Recent cash flows suggest that, in addition to focusing on income, investors have begun to move strongly into equities, indicating that risk-taking behavior is increasing. As discussed throughout, investors reaching for yield and moving out of bonds into equities should realize that risk premia—the compensation for taking on this extra risk—are likely lower now than at any point in the past five years. As the recent performance of stocks and bonds over the 14 years through 2013 reminds us (see Figure 21 on page 27), investors who increase their allocation to riskier segments of the capital markets should realize that portfolio volatility will likely increase as a result. We encourage investors to evaluate the trade-offs involved in a move toward risky asset classes, whether that means tilting a bond portfolio toward corporate and high-yield investments or a wholesale move from bonds into equities. Having a realistic expectation of the extra return to be gained from such a strategy and an understanding of the implications for holistic portfolio risk is crucial to maintaining the discipline needed for long-term success.

Key terms Beta. A measure of the volatility of a security or portfolio relative to a benchmark. Price/earnings ratio. The ratio of a stock’s current price to its per-share earnings over a designated period. Risk premium. The amount by which an asset’s expected return exceeds the risk-free interest rate.

28 

References Baker, Scott R., Nicholas Bloom, and Steven J. Davis, 2012. Measuring Economic Policy Uncertainty; available at policyuncertainty.com. Davis, Joseph H., Roger Aliaga-Díaz, Charles J. Thomas, and Ravi G. Tolani, 2013. The Outlook for Emerging Market Stocks in a Lower-Growth World. Valley Forge, Pa.: The Vanguard Group. Davis, Joseph H., 2013a. Look Back Before Looking Ahead. Vanguard blog posting; available at vanguardblog.com. Davis, Joseph H., 2013b. Bond Risk—A Theory of Relativity. Vanguard blog posting; available at vanguardblog.com. Davis, Joseph H., 2012. Our Economic Future, Vanguard video; available at institutional.vanguard.com. Davis, Joseph H., and Roger Aliaga-Díaz, 2012. Vanguard’s Economic and Investment Outlook. Valley Forge, Pa.: The Vanguard Group. Davis, Joseph H., Roger Aliaga-Díaz, and Charles J. Thomas, 2012. Forecasting Stock Returns: What Signals Matter, and What Do They Say Now? Valley Forge, Pa.: The Vanguard Group. Davis, Joseph H., Roger Aliaga-Díaz, Charles J. Thomas, and Nathan Zahm, 2012. The Long and Short of TIPS. Valley Forge, Pa.: The Vanguard Group. Davis, Joseph H., Roger Aliaga-Díaz, and Andrew J. Patterson, 2011. Asset Allocation in a Low-Yield and Volatile Environment. Valley Forge, Pa.: The Vanguard Group. Davis, Joseph H., Roger Aliaga-Díaz, Donald G. Bennyhoff, Andrew J. Patterson, and Yan Zilbering, 2010. Deficits, the Fed, and Rising Interest Rates: Implications and Considerations for Bond Investors. Valley Forge, Pa.: The Vanguard Group.

Davis, Joseph H., Roger Aliaga-Díaz, Julieann Shanahan, and Charles J. Thomas, 2009. Which Path Will the U.S. Economy Follow? Lessons From the 1990s Financial Crises of Japan and Sweden. Valley Forge, Pa.: The Vanguard Group.

Philips, Christopher B., Francis M. Kinniry Jr., Brian J. Scott, Michael A. DiJoseph, and David J. Walker, 2013. Risk of Loss: Should the Prospect of Rising Rates Push Investors From High-Quality Bonds? Valley Forge, Pa.: The Vanguard Group.

Gordon, Robert J., 2012. Is U.S. Economic Growth Over? Faltering Innovation Confronts the Six Headwinds. Working Paper. Cambridge, Mass.: National Bureau of Economic Research; available at nber.org/papers/w18315.

Philips, Christopher B., Joseph Davis, Andrew J. Patterson, and Charles J. Thomas, 2012. Global Fixed Income: Considerations for U.S. Investors, Valley Forge, Pa.: The Vanguard Group.

Gürkaynak, Refet S., Brian Sack, and Jonathan H. Wright, 2006. The U.S. Treasury Yield Curve: 1961 to the Present. Washington, D.C.: Federal Reserve Board, Divisions of Research & Statistics and Monetary Affairs, Finance and Economics Discussion Series; available at federalreserve.gov. Kinniry Jr., Francis M., 2013. Same As It Ever Was. Vanguard blog posting; available at vanguardblog.com.

Shiller, Robert J., 2000. Irrational Exuberance, second edition. New York: Broadway Books. Summers, Lawrence H., 2013. Economic Forum: Policy Responses to Crises, International Monetary Fund video; available at imf.org. Warnock, Francis E. and Veronica Cacdac Warnock, 2009. International Capital Flows and U.S. Interest Rates. Journal of International Money and Finance 28: 903–919.

Kinniry Jr., Francis M., and Brian J. Scott, 2013. Reducing Bonds? Proceed With Caution, Valley Forge, Pa.: The Vanguard Group. Philips, Christopher B., and Charles J. Thomas, 2013. Fearful of Rising Interest Rates? Consider a More Global Bond Portfolio. Valley Forge, Pa.: The Vanguard Group.

29

Appendix: Vanguard Capital Markets Model IMPORTANT: The projections or other information generated by the Vanguard Capital Markets Model regarding the likelihood of various investment outcomes are hypothetical in nature, do not reflect actual investment results, and are not guarantees of future results. VCMM results will vary with each use and over time.

The VCMM projections are based on a statistical analysis of historical data. Future returns may behave differently from the historical patterns captured in the VCMM. More important, the VCMM may be underestimating extreme negative scenarios unobserved in the historical period on which the model estimation is based.

The Vanguard Capital Markets Model® is a proprietary financial simulation tool developed and maintained by Vanguard’s primary investment research and advice teams. The model forecasts distributions of future returns for a wide array of broad asset classes. Those asset classes include U.S. and international equity markets, several maturities of the U.S. Treasury and corporate fixed income markets, international fixed income markets, U.S. money markets, commodities, and certain alternative investment strategies. The theoretical and empirical foundation for the Vanguard Capital Markets Model is that the returns of various asset classes reflect the compensation investors require for bearing different types of systematic risk (beta). At the core of the model are estimates of the

Figure A-1. Projected ten-year nominal return outlook for balanced portfolios VCMM-simulated distribution of expected average annualized return on balanced global equity and global fixed income portfolios, estimated as of year-end 2013 18%

Key

Average annualized return

16 14

History, 1926−2013 History, 2000−2013

12

95th percentile

10 8 25th−75th percentile

6 4 2 0 –2

5th percentile

20%/80%

60%/40%

80%/20%

Portfolio stock/bond allocation

Underlying data for this figure Portfolio stock/bond allocation

20%/80%

60%/40%

80%/20%

Bottom 5th percentile

1.5%

0.4%

−0.4%

25th percentile

2.8%

3.9%

4.2%

75th percentile

4.6%

8.6%

10.5%

Top 95th percentile

6.0%

12.2%

15.3% 14.4%

Annualized portfolio volatility

30 

6.2%

11.2%

History, 1926−2013

6.6%

8.6%

9.5%

History, 2000−2013

4.9%

4.6%

4.3%

Notes: Figure displays the 5th/25th/75th/95th percentile range of VCMMprojected returns for balanced portfolios. Historical returns are computed using the indexes defined on page 4. Source: Vanguard.

Figure A-2. VCMM simulation output for broad U.S. stock market (10,000 simulations)

Annualized ten-year returns

40% 1950s

30% 20%

1980s 1990s

10%

1940s

1960s

0%

1930s

1970s 2000s

−10% −20% 0

5

10

15

20

25

30

35

40

45%

Annual volatility 10,000 simulations

Median simulation

History

Source: Vanguard.

dynamic statistical relationship between risk factors and asset returns, obtained from statistical analysis based on available monthly financial and economic data from as early as 1960. Using a system of estimated equations, the model then applies a Monte Carlo simulation method to project the estimated interrelationships among risk factors and asset classes as well as uncertainty and randomness over time. The model generates a large set of simulated outcomes for each asset class over several time horizons. Forecasts are obtained by computing measures of central tendency in these simulations. Results produced by the tool will vary with each use and over time. The primary value of the VCMM is in its application to analyzing potential client portfolios. VCMM assetclass forecasts—comprising distributions of expected returns, volatilities, and correlations—are key to the evaluation of potential downside risks, various risk– return trade-offs, and diversification benefits of various asset classes. Although central tendencies are generated in any return distribution, Vanguard stresses that focusing on the full range of potential outcomes for the assets considered, such as the data presented in this paper, is the most effective way to use VCMM output.

The VCMM seeks to represent the uncertainty in the forecast by generating a wide range of potential outcomes. It is important to recognize that the VCMM does not impose “normality” on the return distributions, but rather is influenced by the so-called fat tails and skewness in the empirical distribution of modeled asset-class returns. Within the range of outcomes, individual experiences can be quite different, underscoring the varied nature of potential future paths. Indeed, this is a key reason why we approach asset return outlooks in a distributional framework, as shown in Figure A-1, which highlights balanced portfolio returns before adjusting for inflation. Figure A-2 further illustrates this point by showing the full range of scenarios created by the model. The scatter plot displays 10,000 geometric average ten-year returns and standard deviations for U.S. equities. The dispersion in returns and volatilities is wide enough to encompass historical market performance for various decades.

31

P.O. Box 2600 Valley Forge, PA 19482-2600

Connect with Vanguard® > vanguard.com

Vanguard research > Vanguard Center for Retirement Research Vanguard Investment Strategy Group E-mail > [email protected] For more information about Vanguard funds, visit vanguard.com or call 800-662-2739 to obtain a prospectus. Investment objectives, risks, charges, expenses, and other important information about a fund are contained in the prospectus; read and consider it carefully before investing. CFA ® is a trademark owned by CFA Institute.

© 2014 The Vanguard Group, Inc. All rights reserved. Vanguard Marketing Corporation, Distributor.



ICREIO 022014