Asset Class and Capital Market Assumptions Methodology. Portfolio Strategy and Research

Transcription

1 Asset Class and Capital Market Assumptions Methodology Portfolio Strategy and Research I N V E S T M E N T S T R AT E G Y & R E S E A R C H

2

3 Introduction Loring Ward s Portfolio Strategy and Research (PSR) team periodically reviews asset class definitions and capital market assumptions in support of its Asset Class investing philosophy. The analysis presented supports changes to: Definition of asset classes Selection of asset class proxies Estimates of the return, standard deviation and correlation of the asset classes The results of this analysis will be reflected in: Updated asset class proxies and benchmarks for the Investment Planning Center (IPC) and the new Envestnet proposal tool Expected return and standard deviation and correlations forecasts for the Investment Planning Center (IPC) and the ENV2 proposal tool Updated asset classes and asset class proxies for new and existing simulation tools Capital market assumptions for mean-variance optimization (MVO) of Loring Ward GPS model portfolios Research on improving Loring Ward s model portfolio and SA Funds lineup IMPORTANT: The projections or other information generated by the Investment Planning Center and ENV2 Proposal tools regarding the likelihood of various investment outcomes are hypothetical in nature, do not reflect actual investment results and are not guarantees of future results. Results may vary with each use and over time. For Investment Professional Use Only 1

4 Asset Class Definitions Asset classes are groups of securities which share common characteristics that affect risk and expected return. We consider the following when defining asset classes: We define broad asset classes that coarsely divide securities by type (stocks, bonds, etc.) and geography (domestic, international). Examples: U.S. Stocks, U.S. Investment Grade Bonds. Within each broad asset class, we decide which securities represent core holdings and select a benchmark index as a proxy for the core holdings. For example, we have chosen the Russell 3000 index to represent core holdings within the U.S. Stocks asset class. We define two types of targeted asset classes. The first type is a subset of a broad asset class that tilts the core holdings toward one or more security characteristics that are believed to affect risk and/or expected return. Examples: U.S. Large Value Stocks, U.S. Short-Term Government Bonds. The second type is an asset class that augments the core asset holdings. These targeted asset classes may include securities that fall outside of the core holdings. Examples: U.S. High-Yield Bonds, Emerging Market Stocks. Has the asset class historically exhibited a significant difference in average return or volatility relative to a broad asset class? Does the asset class exhibit sufficiently low correlation with other asset classes to be considered distinct? Is the asset class definition consistent with industry practice? For example, is the asset class consistent with a Morningstar category? Is a commercial benchmark index available for the asset class? Index benchmarks permit the use of passive index funds to obtain asset class exposure with minimal tracking error. Benchmarks also permit detailed performance attribution. Are high-quality, low-cost funds available to obtain asset class exposure? 2 For Investment Professional Use Only

5 Asset Class Choices Six core asset classes are defined: U.S. Stocks, International Stocks, Investment Grade Bonds, Real Estate/REITs, Commodities, and Cash & Cash Equivalents. U.S. Stocks U.S. Stocks is defined as the U.S. stock market (as defined by Russell). Within U.S. Stocks, six sub-classes are defined along the value and size dimensions. We include U.S. Microcap stocks as a target asset class which augments the U.S. Stocks asset class. See the table below for details. Broad Asset Class Targeted Asset Class Targeted Asset Class (Subset of Core Holdings) (Augment Core Holdings) U.S. Stocks U.S. Large Neutral Stocks U.S. Microcap U.S. Large Value Stocks U.S. Large Growth Stocks U.S. Small Neutral Stocks U.S. Small Value Stocks U.S. Small Growth Stocks International Stocks International is defined as Developed Markets (as defined by MSCI) excluding the United States. Within International Stocks, six sub-classes are defined along the value and size dimensions. We include Emerging Markets (as defined by MSCI) as a targeted asset class which augments the International Stocks asset class. See the table below for details. Broad Asset Class Targeted Asset Class Targeted Asset Class (Subset of Core Holdings) (Augment Core Holdings) International Stocks International Large Neutral Stocks Emerging Markets Stocks International Large Value Stocks International Large Growth Stocks International Small Neutral Stocks International Small Value Stocks International Small Growth Stocks For Investment Professional Use Only 3

6 Investment Grade Bonds Investment Grade Bonds are defined as the U.S. investment grade bond market (as defined by Barclays Capital). This includes U.S. treasury, U.S. agency and U.S. high-grade credit bonds of all maturities. Within Investment Grade Bonds, five sub-classes are defined along the term and credit dimensions. Targeted asset classes which augment the Investment Grade Bonds asset classes include the sub-asset classes listed below. Broad Asset Class Targeted Asset Class Targeted Asset Class (Subset of Core Holdings) (Augment Core Holdings) U.S. Investment Grade Bonds U.S. Short Government Bonds Global Short Bonds U.S. Short Investment Grade Bonds Global Bonds U.S. Government Bonds U.S. High Yield Bonds U.S. Long Government Bonds U.S. Inflation-Protected Bonds U.S. Long Credit Bonds Municipal Bonds Cash, Commodities and Real Estate The final three asset classes include Cash & Cash Equivalents, Commodities, and Real Estate. At present, there are no sub-asset classes within these broad asset classes. Allocations In addition to our broad and targeted asset classes, we define and maintain a time series of monthly returns for several generic portfolio allocations. The Global Developed allocation includes large-cap and mid-cap stocks from developed markets (defined by MSCI). The Global includes Emerging Markets and small-cap stocks to the Global Developed allocation. The Aggressive, Moderate and Conservative portfolios are composed of Global Developed Stocks and Investment Grade Bonds. Allocation Weights Notes Global Developed Stocks 100% Global Developed Stocks DM (U.S. + Intl) Global Stocks 100% Global Stocks DM + EM Aggressive 80% Global Developed Stocks, 20% U.S. Investment Grade Bonds Moderate 50% Global Developed Stocks, 50% U.S. Investment Grade Bonds Conservative 20% Global Developed Stocks, 80% U.S. Investment Grade Bonds 4 For Investment Professional Use Only

7 Asset Class Proxies Asset Class Proxies were chosen to create a time series of monthly returns that are representative of each asset class. These proxies are comprised of popular commercial benchmarks. Several important benchmark indexes have inception dates after the beginning of our sample period. Where necessary, we have included returns from other indexes with similar properties in order to create a complete asset class return series that commence in January These Asset Class Proxies and their start dates are reported in Appendix 1. Choice of Sample Period Our sample period for statistical analysis begins in January There are several compelling reasons for this choice, namely: Global bond and money markets behaved very differently prior to the 1970 s. In a successful effort to curb high inflation, the Federal Reserve changed its operating procedures in the 1970 s. This change led to wider variation in the Fed Funds rate and increased volatility in short-term interest rates. Floating exchange rates replaced the Bretton Woods system of fixed exchange rates pegged to the U.S. Dollar. Global equity markets matured and became accessible to U.S. investors (institutional and retail) with many international equity benchmarks having inception dates during the 1970 s. Deregulation and competition drove transaction costs starting in the 1970 s, which led to increased trading volumes and liquidity. Many financial innovations (including index funds and market-traded derivatives) were introduced in the 1970 s. Capital Market Assumptions Loring Ward s investment philosophy is based, in part, on the idea that there is information contained in securities prices and that investors should take a long-term approach to reaching their financial goals. Our capital market assumptions were developed with these philosophical tenets in mind. Our approach to estimating expected returns has two steps: 1. Forecast the long-horizon risk-free rate. We focus on the long-term expected return on a twenty-year U.S. Treasury bond. 2. Forecast risk premiums that we expect asset classes to earn relative to the long-horizon riskfree rate. Our estimates incorporate historical average excess returns (relative to the twentyyear Treasury bond income return), yield data, and subjective adjustments informed by theory and professional judgment. For Investment Professional Use Only 5

8 Expected Term Structure of Interest Rates The first step is to forecast the long-horizon risk-free rate. This entails estimating the average term structure of interest rates that we expect over a twenty-year horizon. Our forecast is informed by historical data on yields. U.S. Treasury bonds are widely considered to be default risk-free. This means that, barring a default by the Federal government, the future nominal cash flows of a U.S. Treasury bond are known with certainty. Yields provide valuable information regarding future average bond returns. Dai (2015) provides a nice summary of the relevant theory and empirical evidence. The yield curve is a plot of yields vs. maturities on a given date. The figure below plots the median, 5th percentile and 95th percentile yields for constant-maturity Treasury bonds for the period January 1972 to December The constant-maturity yield curve for December 2015 is plotted as a point of reference. Note that the December 2015 yield curve reflects very low nominal yields relative to recent (post-1972) history. Figure 1. Nominal Yield Curves, Source: Federal Reserve Bank of St. Louis 14 Nominal Yield Curves, Yield (% per annum) Median 5th Percentile 95th Percentile Dec 2015 Long-term Forecast Month 1 Year 2 Years 3 Years 5 Years 7 Years 10 Years 20 Years 30 Years Treasury bonds are backed by the U.S. government and are subject to interest rate and inflation risk. Treasury bond values will decline as interest rates rise. The value of the U.S. dollar depreciates over time with inflation, so the primary risk is inflation risk. 6 For Investment Professional Use Only

9 Two points on the yield curve are of particular interest: the one-month yield and the twenty-year yield. Our forecast of average future nominal yields will be critically dependent on these two end points. We know that one-month nominal yields are historically very close to the expected inflation rate. The Federal Reserve has signaled its intention to target a 2% inflation rate for the foreseeable future, and we have no compelling reason to expect otherwise. Therefore, we anchor our forecast of the yield curve at 2% on the short-end. Note that our expectation of longrun 2% inflation is significantly lower than the average annualized inflation rate 3.99% for However, it is slightly higher than the 20-year breakeven inflation rate of 1.55% observed in December At the other end of the yield curve, we observe that the twenty-year nominal yield has historically exceeded the one-month yield by about 2%. Accordingly, we forecast that twenty-year yield will be 4%, on average, in the long run. This is our estimate of the long-horizon risk-free rate, and is one of the touchstones of our capital market assumption process. The other maturities on the forecast yield curve are also plotted in the figure above. Note that the forecast yield curve lies substantially below the median yield curve, but above the December 2015 snapshot. This reflects our belief that, over the next twenty years, nominal yields will be modestly higher than current yields, but well below historical averages. The difference between forecast yields on the 20-year Treasury bond and a shorter-maturity Treasury bond represents a term premium. For Investment Professional Use Only 7

10 Asset Class Premiums Bonds and Cash For bonds, we are concerned with two types of risk: term and default. Our approach to estimating expected returns for a bond-based asset class is to start with the long-horizon risk-free rate (rf), subtract a term discount (TD) for shorter-duration bonds, and add a default premium (DP) for bonds with default risk. This approach is consistent with one developed by Lummer, Riepe, and Siegel (1994). The approach is represented in the following simple model. E[return] = rf TD + DP We employ both historical return and historical yield data to estimate term discounts and default premiums. The term discounts are related to the slope of the forecast average yield curve discussed in the previous sub-section. Estimates of default premiums for investment grade bonds are based on (1) mean returns in excess of the 20-year Treasury bond income return, and (2) historical yields. The results are reported in the table below. All rates are reported in percent per annum. Horizon Cash Short Intermediate Long Yield Curve Maturity Duration Range (years) Real Rate 0.0% 1.4% 1.7% 2.0% Inflation Rate 2.0% 2.0% 2.0% 2.0% Nominal Rate 2.0% 3.4% 3.7% 4.0% Inv-Grade Default Premium 0.0% 0.2% 0.5% 1.0% Term Discount 2.0% 0.6% 0.3% 0.0% 1 The breakeven inflation rate for a particular maturity is the difference between the yields on a constant-maturity nominal Treasury bond and its inflation-protected counterpart. Source: Federal Reserve Bank of St. Louis. 2 Note that we are subtracting a term discount from a long-horizon risk-free rate rather than adding a term premium to a short-term risk-free rate. There are several reasons for this approach. For long-horizon investors, the long-horizon rate is more appropriate. A longhorizon bond investor can reasonably expect to earn the long-horizon yield as an income return. And, short-term rates are more volatile than long-term rates. Yield curves may swing from upward-sloping to downward-sloping, but long-horizon yields exhibit more stability. Stocks and Other Asset Classes For stocks and other asset classes, we adopt a complementary approach. We assume that the expected return on an asset class is the sum of the long-horizon risk-free rate plus a risk premium. E[return] = rf + Risk Premium Our approach to estimating expected risk premiums is straightforward. Since we have carefully constructed asset class proxies based on security characteristics, we can observe historical data to compute average excess returns relative to the long-horizon risk-free asset. We employ income returns on a 20-year Treasury bond as our proxy for the long-horizon risk-free rate. Risk premia are time-varying and the return generating process in the next twenty years may not be similar to that represented historically. We adjust our premium estimates upward or downward 8 For Investment Professional Use Only

11 from the average excess return. These adjustments are necessarily subjective, but are refined by judgment. Our estimates of premiums over the long-horizon risk-free rate for targeted equity asset classes are summarized in the tables below. The top table is for U.S. Stocks and the bottom table is for International Stocks. In each case the premium for the Large Neutral asset class can be thought of as the expected equity premium. The relative premiums (or discounts) for other targeted asset classes (in basis points, or bps) are reported in each cell. U.S. Growth Neutral Value Large 2.5% 3.5% 4.5% -100bps +100 bps Small 3.0% 4.5% 5.5% -50bps +100 bps +200bps Int l Growth Neutral Value Large 2.5% 3.5% 4.5% -100bps +100 bps Small 3.0% 4.5% 5.5% -50bps +100 bps +200bps EM 5.0% +150 bps Equity Premium. The global economy and financial markets are changing, and there are compelling reasons to believe that equity premiums in the future will be lower than historical averages. Elevated valuations are likely to be permanent, on average, and will result in lower expected real returns in the future. Empirical evidence in support of this view is reported in Fama and French (2002). Accordingly, our estimate of the expected equity premium for the U.S. is approximately 50 bps lower than its historical average for the sample period. We set our expected international equity premium equal to the expected U.S. equity premium. Small Stocks. The small-cap premium was first documented for U.S. Stocks by Banz (1981). However, average returns for small-cap stocks, both in the U.S. and abroad, have been lower, on average, since the mid-1980 s. This fact may reflect increased awareness of small-cap stocks or more developed global capital markets. We estimate that small-cap stocks will earn an expected premium of 100 bps per annum relative to large-cap stocks. This is lower than the historical average, especially for international stocks. Value/Growth Stocks. In the 1980 s, several academic papers documented that signals relating stock price to fundamentals (e.g., Price/Book, Price/Earnings, etc.) had cross-sectional forecasting power for stock returns. This empirical finding is captured by the familiar Fama-French 3-factor model (Fama and French, 1992) which employs the Book/Market (B/M) ratio to sort individual stocks into portfolios. High B/M indicates a value stock and low B/M indicates a growth stock. For both U.S. and International Stocks, we estimate that Large Value will earn +100 bps relative to Large Neutral, and that Large Growth will earn -100 bps relative to Large Neutral. For U.S. and International Small-Cap stocks, we estimate that Small Value will earn +100 bps relative to Small Neutral, For Investment Professional Use Only 9

12 and that Small Growth will earn -150 bps relative to Small Neutral. These premiums/discounts are very close to historical averages. The poor relative performance of small-cap growth stocks is documented in Lam, Markowitz and McFarland (2015) Assumptions The table in Appendix 2 reports Loring Ward s 2016 forecast of long horizon expected returns. The first column is the expected return forecast. The second column is the average return (annualized) for the January 1972-December 2016 sample period. Prior to this update, Loring Ward employed average nominal returns as expected return forecasts. So, the difference between columns 1 and 2 (reported in column 3) can be thought of as the revision in expectations. All of the asset class expected returns have been substantially revised downwards. The main culprit is the long-horizon risk-free rate. From , the average yield-to-maturity for a constant-maturity 20-year Treasury bond was 7.00%. The average income return on the same bond was 6.93%. Our forecast is 4% going forward, which represent a nearly 3% reduction to all expected returns. Of this 3%, approximately 2% is due to a decrease in inflationary expectations and another 1% is due to a decrease in the expected real yield. Note that the nominal yield on the 20-year Treasury bond as of December 31, 2015 was 2.67%. Estimating Covariance Matrices Our approach for estimating the covariance matrix returns is very simple. We compute the sample covariance matrix of monthly returns and annualize by multiplying by twelve. This procedure implicitly assumes that asset class returns exhibit no serial correlation. This is a reasonable first-order approximation of the return generating process. Estimation Error We have employed statistical methods to estimate average excess returns, and those estimates are subject to error. For example, the standard error of the mean excess return for the U.S. Stocks asset class is 68 bps. Assuming that monthly returns are normally distributed, this implies that the 95% confidence interval for our estimate of the U.S. market risk premium (over the 20-year Treasury bond yield) is 4.3% ± 1.3%. Clearly, statistical estimates are imprecise. Our estimate of the long-horizon risk-free rate, 4%, is subject to error as well. Taken together, this means that we expect the average return on U.S. Stocks over the next 20 years to be somewhere between 6.3% and 10.3%. The estimation error for more volatile asset classes is even higher. Other Estimates Many firms in the financial services industry publish their capital market assumptions. Although methodologies, purposes and horizons vary greatly, there is some value in comparing our assumptions with others in the industry. Appendix 3 reports a selection of capital market assumptions from J.P. Morgan Asset Management (JPM), State Street Global Advisors (SSGA) and AQR. In all cases, expected returns are well below historical averages. 10 For Investment Professional Use Only

13 References AQR, Inc., Capital Market Assumptions for Major Asset Classes. AQR Alternative Thinking, First Quarter, Banz, R. W., 1981, The relationship between return and market value of common stocks, Journal of Financial Economics 9(1), Dai, Wei, Empirical Evidence in the Fixed Income Markets. Dimensional Fund Advisors, December Fama, Eugene F., and Kenneth R. French, 1992, The Cross-Section of Expected Returns. Journal of Finance 47, Fama, Eugene F., and Kenneth R. French, 2002, The Equity Premium. Journal of Finance 57, Federal Reserve Bank of St. Louis, 20-year Breakeven Inflation Rate [T20YIEM], retrieved from FRED, Federal Reserve Bank of St. Louis November 30, J.P. Morgan Asset Management, 2015, 2016 Long-Term Capital Market Assumptions. Lam, Wynce, Harry M. Markowitz, and Sheldon P. McFarland, 2015, The Likelihood of Small Cap Premium Distributions. Working Paper. Lummer, Scott L., Mark W. Riepe, and Laurence B. Siegel, 1994, Taming Your Optimizer: A Guide Through the Pitfalls of Mean-Variance Optimization. In Global Asset Allocation: Techniques for Optimizing Portfolio Management, Jess Lederman and Robert A. Klein, eds., New York: John Wiley & Sons. Morningstar, Inc., 2014 Ibbotson Stocks, Bonds, Bills and Inflation (SBBI) Classic Yearbook. Chicago: Morningstar. State Street Global Advisors, Long-Term Asset Class Forecasts. March 30, For Investment Professional Use Only 11

14 Appendix 1 Asset Classes and Proxies Asset Class Benchmark Start Date Cash & Cash Alternatives BofAML U.S. Treasury Bill 3 Mon TR USD 1/1/1978 IA SBBI U.S. 30 Day TBill TR USD 1/1/1972 U.S. Short Government Bonds BofAML U.S. Trsy/Agcs AAA 1-3 Yr TR USD 10/1/1982 BofAML U.S. Treasuries 1-3 Yr TR USD 1/1/1978 IA SBBI U.S. 1 Yr Trsy Const Mat TR USD 1/1/1972 U.S. Short Investment Grade Bonds BofAML U.S. Corp&Govt 1-3 Yr TR USD IA SBBI U.S. 1 Yr Trsy Const Mat TR USD 6/1/1986 1/1/1972 U.S. Government Bonds Barclays U.S. Government TR USD 1/1/1973 IA SBBI U.S. IT Govt TR USD 1/1/1972 U.S. Investment Grade Bonds Barclays U.S. Agg Bond TR USD 1/1/1976 Barclays U.S. Govt/Credit TR USD 1/1/1973 IA SBBI U.S. IT Govt TR USD 1/1/1972 U.S. Long Government Bonds Barclays U.S. Government Long TR USD 1/1/1973 IA SBBI U.S. LT Govt TR USD 1/1/1972 U.S. Long Credit Bonds Barclays U.S. Long Credit TR USD 1/1/1973 IA SBBI U.S. LT Corp TR USD 1/1/1972 Global Short Bonds Citi WBGI 1-5 Yr Hdg USD 1/1/1985 IA SBBI U.S. IT Govt TR USD 1/1/1972 Global Bonds Citi WBGI Hdg USD 1/1/1985 IA SBBI U.S. IT Govt TR USD 1/1/1972 Municipal Bonds Barclays Municipal TR USD 2/1/1980 IA SBBI U.S. LT Corp TR USD 1/1/1972 U.S. High-Yield Bonds Barclays U.S. Corporate High Yield TR USD 7/1/1983 IA Barclays U.S. HY Corporate Bonds 1/1/1972 U.S. Inflation-Protected Bonds Barclays U.S. Treasury U.S. TIPS TR USD 3/1/1997 IA SBBI U.S. LT Govt TR USD 1/1/1972 U.S. Stocks Russell 3000 TR USD 1/1/1979 CRSP Deciles 1-10 Index 1/1/1972 U.S. Large Neutral Stocks Russell 1000 TR USD 1/1/1979 S&P 500 Index 1/1/1972 U.S. Large Value Stocks Russell 1000 Value TR USD 1/1/1979 Fama/French U.S. Large Value Index (ex utilities) 1/1/1972 U.S. Large Growth Stocks Russell 1000 Growth TR USD 1/1/1979 Fama/French U.S. Large Growth Index (ex utilities) 1/1/1972 U.S. Small Neutral Stocks Russell 2000 TR USD 1/1/1979 CRSP Deciles 6-10 Index 1/1/1972 U.S. Small Value Stocks Russell 2000 Value Index 1/1/1979 Fama/French U.S. Small Value Index (ex utilities) 1/1/ For Investment Professional Use Only

15 Asset Class Benchmark Start Date U.S. Small Growth Stocks Russell 2000 Growth Index 1/1/1979 Fama/French U.S. Small Growth Index (ex utilities) 1/1/1972 U.S. Microcap Stocks Russell Micro Cap TR USD 7/1/2000 CRSP Deciles 9-10 Index 1/1/1972 Real Estate / REITs DJ U.S. Select REIT TR USD 1/1/1987 Global Stocks (includes Int l Dev, U.S., EM) Global Developed Stocks (includes Int l Dev, U.S.) International Stocks (includes Int l Developed) FTSE NAREIT All Equity REITs TR USD 1/1/1972 MSCI ACWI IMI NR USD 6/1/1994 MSCI World NR USD 1/1/1972 MSCI World IMI NR USD MSCI World NR USD 6/1/1994 1/1/1972 MSCI World ex USA IMI NR USD 6/1/1994 MSCI World ex USA NR USD 1/1/1972 International Large Neutral Stocks MSCI World ex USA NR USD 1/1/1972 International Large Value Stocks MSCI World Ex USA Value NR USD 1/1/1975 MSCI World ex USA NR USD 1/1/1972 International Large Growth Stocks MSCI World Ex USA Growth NR USD 1/1/1975 MSCI World ex USA NR USD 1/1/1972 International Small Neutral Stocks MSCI World Ex USA Small Cap NR USD 1/1/2001 Dimensional International Small Cap Index 1/1/1972 International Small Value Stocks MSCI World Ex USA Small Value NR USD 6/1/1994 Dimensional International Small Cap Value Index 7/1/1981 Dimensional International Small Cap Index 1/1/1972 International Small Growth Stocks MSCI World Ex USA Small Growth NR USD 6/1/1994 S&P Developed Ex U.S. Small Growth TR 1/1/1990 Dimensional International Small Cap Index 1/1/1972 Emerging Markets Stocks MSCI EM NR USD 1/1/1999 MSCI EM GR USD 1/1/1988 MSCI Pacific Ex Japan NR USD 1/1/1972 Commodities Bloomberg Commodity TR USD 1/1/1991 S&P GSCI TR USD 1/1/1972 Other S&P 500 Index PR 1/1/1972 Aggressive Allocation 20% Barclays U.S. Agg Bond TR USD, 80% MSCI World IMI NR USD 6/1/ % Barclays U.S. Agg Bond TR USD, 80% MSCI World NR USD 1/1/ % Barclays U.S. Govt/Credit TR USD, 80% MSCI World NR USD 1/1/ % IA SBBI U.S. IT Govt TR USD, 80% MSCI World NR USD 1/1/1972 Conservative Allocation 80% Barclays U.S. Agg Bond TR USD, 20% MSCI World IMI NR USD 6/1/ % Barclays U.S. Agg Bond TR USD, 20% MSCI World NR USD 1/1/ % Barclays U.S. Govt/Credit TR USD, 20% MSCI World NR USD 1/1/ % IA SBBI U.S. IT Govt TR USD, 20% MSCI World NR USD 1/1/1972 Moderate Allocation 50% Barclays U.S. Agg Bond TR USD, 50% MSCI World NR IMI USD 6/1/ % Barclays U.S. Agg Bond TR USD, 50% MSCI World NR USD 1/1/ % Barclays U.S. Govt/Credit TR USD, 50% MSCI World NR USD 1/1/ % IA SBBI U.S. IT Govt TR USD, 50% MSCI World NR USD 1/1/1972 For Investment Professional Use Only 13

16 Appendix 2 Capital Market Assumptions Asset Class Expected Return (Annualized) Average Arithmetic Return (Annualized) Difference Std. Dev. of Annual Returns Sharpe Ratio (Annual Returns) Cash & Cash Alternatives 2.00% 5.24% -3.24% 3.85% 0.00 U.S. Short Government Bonds 3.40% 6.28% -2.88% 4.55% 0.23 U.S. Short Investment Grade Bonds 3.60% 6.31% -2.71% 4.03% 0.27 U.S. Government Bonds 3.70% 7.31% -3.61% 6.39% 0.32 U.S. Investment Grade Bonds 4.20% 7.50% -3.30% 6.72% 0.34 U.S. Long Government Bonds 4.00% 8.93% -4.93% 12.10% 0.30 U.S. Long Credit Bonds 5.00% 8.97% -3.97% 10.55% 0.35 Global Short Bonds 3.40% 6.38% -2.98% 5.05% 0.23 Global Bonds 3.70% 7.32% -3.62% 5.58% 0.37 Municipal Bonds 2.80% 6.47% -3.67% 9.07% 0.14 U.S. High-Yield Bonds 6.50% 9.78% -3.28% 14.64% 0.31 U.S. Inflation-Protected Bonds 3.70% 7.83% -4.13% 10.38% 0.25 U.S. Stocks 7.60% 11.91% -4.31% 17.74% 0.38 U.S. Large Neutral Stocks 7.50% 11.90% -4.40% 17.62% 0.38 U.S. Large Value Stocks 8.50% 13.35% -4.85% 17.18% 0.47 U.S. Large Growth Stocks 6.50% 11.32% -4.82% 20.45% 0.30 U.S. Small Neutral Stocks 8.50% 13.32% -4.82% 22.25% 0.36 U.S. Small Value Stocks 9.50% 15.23% -5.73% 21.69% 0.46 U.S. Small Growth Stocks 7.00% 11.71% -4.71% 25.74% 0.25 U.S. Microcap Stocks 9.00% 13.90% -4.90% 25.58% 0.34 Real Estate / REITs 9.00% 12.92% -3.92% 18.51% 0.41 Global Stocks (includes Int'l Developed, U.S., EM) 7.50% 10.24% -2.74% 18.08% 0.28 Global Developed Stocks (includes Int'l Dev, U.S.) 7.50% 10.33% -2.83% 17.84% 0.29 International Stocks (includes Int'l Developed) 7.50% 10.93% -3.43% 21.54% 0.26 International Large Neutral Stocks 7.50% 10.83% -3.33% 21.13% 0.26 International Large Value Stocks 8.50% 12.31% -3.81% 21.80% 0.32 International Large Growth Stocks 6.50% 9.65% -3.15% 21.19% 0.21 International Small Neutral Stocks 8.50% 16.65% -8.15% 28.20% 0.40 International Small Value Stocks 9.50% 17.72% -8.22% 28.95% 0.43 International Small Growth Stocks 7.00% 15.40% -8.40% 29.51% 0.34 Emerging Markets Stocks 9.00% 15.47% -6.47% 33.92% 0.30 Commodities 6.00% 10.10% -4.10% 21.83% 0.22 Other 4.50% 8.54% -4.04% 16.91% 0.20 Aggressive Allocation 6.84% 9.70% -2.86% 14.65% 0.30 Conservative Allocation 4.70% 8.00% -3.30% 6.99% 0.39 Moderate Allocation 6.25% 8.81% -2.56% 10.19% For Investment Professional Use Only

17 Appendix 3 A Sample of Industry Capital Market Assumptions Asset Class Horizon (years) LW 20 JPM SSGA 10+ Inflation 2.00% 2.25% 2.00% 2.30% Cash 2.00% 2.00% 2.50% U.S. Short Investment Grade Bonds 3.60% 3.25% U.S. Government Bonds 3.70% 2.25% 2.30% 2.40% U.S. Investment Grade Bonds 4.20% 3.00% 3.00% U.S. Long Government Bonds 4.00% 2.00% U.S. Long Credit Bonds 5.00% 3.75% Global Bonds 3.70% 2.00% 1.30% 1.80% U.S. High Yield Bonds 6.50% 5.75% 4.80% 4.40% U.S. Inflation-Protected Bonds 3.70% 3.50% 2.30% U.S. Large Neutral Stocks 7.50% 6.25% 6.00% 6.50% U.S. Large Value Stocks 8.50% 6.25% U.S. Large Growth Stocks 6.50% 6.25% U.S. Small Neutral Stocks 8.50% 7.00% 6.50% Real Estate / REIT 9.00% 6.00% International Stocks (includes Int l Developed) 7.50% 6.75% 6.20% 6.90% International Large Neutral Stocks 7.50% 6.75% 6.20% International Large Value Stocks 8.50% International Small Neutral Stocks 8.50% Emerging Markets Stocks 9.00% 9.25% 8.40% 7.70% Commodities 6.00% 3.75% 5.30% 5.30% AQR 5-10 Sources: J.P. Morgan, State Street Global Advisors, AQR. For Investment Professional Use Only 15

18 For Investment Professional Use Only Not For Public Distribution LWI Financial Inc. ( Loring Ward ) is an investment adviser registered with the Securities and Exchange Commission. Securities transactions may be offered through Loring Ward Securities Inc., an affiliate, member FINRA/SIPC. B