TIAA-CREF Lifecycle Funds: Methodology and Design

Transcription

1 Introduction The TIAA-CREF Lifecycle Funds are a target retirement date fund family that includes a total of 12 funds: 11 target retirement date funds at five-year intervals for retirement dates 2010 through 2060, and a retirement income fund for those in retirement. The Lifecycle Funds are designed to provide an effective and convenient means for investors who prefer to have their investments professionally managed to help successfully prepare for and fund their retirement years. The Lifecycle Funds invest in a carefully constructed selection of equity, fixed income and direct real estate funds, providing a diversified, professionally managed portfolio. The overall asset allocation of each Lifecycle Fund, which includes the selection, portfolio weight, and mix among the equity, fixed-income and direct real estate funds, evolves as each Lifecycle Fund approaches its target retirement date. TIAA developed the current Lifecycle Funds design with a focus on accumulating wealth during the pre-retirement years while maximizing the opportunity for income in the retirement years, all within a risk-controlled framework. The funds are consistently managed by a dedicated team focused on (1) the daily management of the funds, and (2) the development and ongoing refinement of the funds design and glidepath methodology to reflect both our latest investment insights and industry innovations. Why TIAA-CREF Lifecycle Funds? TIAA Asset Management expertise Dedicated Multi-Asset Investments Team Underlying funds leverage TIAA s vast equity, fixed-income and direct real estate research/expertise Glidepath design Focuses on accumulating wealth during the pre-retirement years while maximizing the opportunity for income during retirement years, all within a risk-controlled framework Incorporates the growing life expectancy of the U.S. population in an effort to strategically balance market risk with longevity risk Risk management through diversification across and within asset classes Blending fundamental and quantitative funds should help lower volatility of the alpha over time Inclusion of lower correlated asset classes to extend diversification Competitive fees Among the lowest fees for actively managed target-date funds 1

2 Glidepath design An important feature of any target-date investment is the structure of the glidepath, or planned progression of asset allocation changes over time. The TIAA-CREF Lifecycle Funds have what is known as a through retirement glidepath. The asset allocation mix evolves both during the working years, when an individual is accumulating assets (the accumulation phase), and during the retirement years, when an individual is withdrawing assets (the drawdown income phase). The glidepath design takes into account the potential for returns as well as sources of risk under a wide range of possible future market conditions. TIAA has been managing retirement savings for its clients since Given this long history, our Multi-Asset Investments Team benefits from having access to a rich set of customer data on how much and how consistently individuals contribute to their retirement plans, and, through our actuarial mortality tables, individual life expectancies at different ages. The Multi-Asset Investments Team incorporates the growing life expectancy of the U.S. population into the glidepath design in an effort to strategically balance market risk with longevity risk. Exhibit 1: TIAA-CREF Lifecycle Funds Glidepath Inflation-Protected Assets Short-Term Fixed Income Allocations % Fixed Income Direct Real Estate International Equity U.S. Equity Years to Retirement Years to Retirement U.S. Equity Int l Equity Direct Real Estate Fixed Income Short-Term Fixed Income Inflation- Protected Assets % 28.5% 2.5% 2.5% 0.0% 0.0% % 28.1% 2.5% 3.8% 0.0% 0.0% % 27.8% 2.5% 5.0% 0.0% 0.0% % 27.4% 2.5% 6.3% 0.0% 0.0% % 27.0% 2.5% 7.5% 0.0% 0.0% % 25.0% 2.5% 16.0% 0.0% 0.0% % 21.9% 2.5% 24.5% 0.0% 0.0% % 19.4% 2.5% 29.0% 2.0% 2.0% % 16.8% 2.5% 33.5% 4.0% 4.0% % 14.3% 2.5% 38.0% 6.0% 6.0% % 12.8% 2.5% 39.0% 8.0% 8.0% -10 to % 11.3% 2.5% 40.0% 10.0% 10.0% Some totals may not add up to 100% due to rounding. 2

3 Glidepath methodology The Lifecycle Funds glidepath has been structured with the goal of replacing income in retirement for investors, based on their retirement date, after having consistently saved during their working years. The investment efforts of the TIAA Multi-Asset Investments Team are overseen by the TIAA Asset Allocation Committee. The Asset Allocation Committee oversees the strategic and tactical asset allocations of TIAA-CREF Lifecycle Funds including the glidepath allocations. The Committee includes: Chief Investment Officer, Multi-Asset Investments Multi-Asset Investments Team Chief Investment Officer, TIAA Global Asset Management Chief Investment Officer, Global Equity Investments Chief Investment Officer, Global Public Markets (Fixed Income) Chief Risk Officer, TIAA Global Asset Management Chief Economist, TIAA Global Asset Management Global Investment Strategist, TIAA Global Asset Management The Multi-Asset Investments Team views the rationale for the glidepath s overall asset allocation mix primarily through the lens of a Human Capital/Financial Capital (HC/FC) Investment Framework. This framework is based on the ability of younger people to take on more investment risk because they have many years of earnings/savings ahead of them (i.e., human capital). As they age and enter retirement, they need to have a more conservative portfolio to offset their declining human capital. Human Capital/Financial Capital Investment Framework Economists define human capital as a financial translation of a person s economic productivity, which is a function of a variety of factors, including formal education, training, and experience. One estimate of human capital is a present value at any age of the individual s remaining ability to generate income through employment. 2 Human capital is generally highest for young people that have many years remaining in their careers in which to generate earnings. As an individual s career progresses, the potential for future income declines, resulting in a gradual reduction of human capital over the course of one s career. However, human capital can be improved gradually over time through investments in additional schooling or training. Financial capital is represented by an individual s accumulated wealth in stocks, bonds, and other investment assets. It has an inverse relationship to human capital, generally starting low for the youngest individuals and growing over time. Compared to human capital, financial capital markets respond rapidly to news and new information. Given human capital s lower sensitivity to market-related events, economists view the volatility of human capital as being less than that of financial markets. 3 In the investment field, a number of studies have analyzed how to consider allocating human capital and financial capital. In this context, the term human capital is not the present value of total earnings the definition that economists use but rather the present value of total retirement savings (earnings saved after consumption). In a 1992 paper on human capital/financial capital, Bodie, Merton and Samuelson conclude that under normal circumstances, an individual will tend to exhibit more conservative investment behavior as he nears retirement. 4 Hence, given that human capital declines as one ages, more conservative financial assets are needed to replace this declining, less volatile asset. 3

4 In 2007, the HC/FC framework was further popularized by a CFA Institute monograph written by Ibbotson, Milevsky, Chen and Zhu. 5 The exhibit below is adapted from that monograph. It shows the relationships between human capital, financial capital, and the asset allocation decision. Exhibit 2: Human Capital/Financial Capital Framework Age, Income Expected returns HUMAN CAPITAL Savings rate FINANCIAL CAPITAL Risk aversion, ρ between HC/FC ASSET ALLOCATION DECISION Source: Adapted from Ibbotson, et al. (2007): p. 17, Figure 2.1 An example of a human capital/financial capital chart is shown in Exhibit 3, which demonstrates how human capital and financial capital contribute in different proportional amounts to an individual s total capital over the course of one s lifetime. Human capital in this chart includes both the present value of savings along with Social Security benefits.. Exhibit 3: Sample Human Capital/Financial Capital Chart 100 Total Capital Human Capital Financial Capital 80 Capital Age Retirement age shown as gray vertical line at age 65 4

5 The Human Capital/Financial Capital framework informs the glidepath decision. Within this HC/FC framework, the Asset Allocation Committee considers a wide range of information from a variety of sources when selecting the strategic glidepath allocation. One input is a mathematical formulation of the HC/FC problem, shown below, which helps inform the overall downward sloping shape of the glidepath: Max (i=1,n) w i r i k * ½ (i=1,n) (j=1,n) w i w j σ i σ j ρ ij (1) s.t. where: w 1 = Human capital at given age. (i=1,n) w i = 1 Max (i=1,n) w i r i k σ i σ j ρ ij s.t. w 1 (i=1,n) w i = Maximize this objective function = Sum of assets 1 through n = Weight in asset i = Expected return of asset i = Relative risk aversion coefficient = Standard deviation of asset i = Standard deviation of asset j = Correlation between asset i and asset j = Such that (the two equations below are the constraints) = Weight in asset 1, which is Human Capital = Sum of weights of Human Capital and all financial assets As part of TIAA s internal research effort, the Lifecycle Portfolio Management Team: evaluates optimization tests based on the HC/FC investment framework; conducts Monte Carlo simulations of individuals saving during their working years in the Lifecycle Funds and generating income through systematic withdrawals in retirement; analyzes an alpha risk budget; and monitors the underlying portfolios. Before reaching a final decision regarding glidepath changes, other sources of information and insights beyond the Team s human capital/financial capital framework, simulations, alpha risk budget, and monitoring are considered by members of the Asset Allocation Committee. These sources include other internal research, external research, economic and capital market insights, sector-specific insights in collaboration with the underlying portfolio managers, investment experience, and product considerations, combined with the Asset Allocation Committee members expert judgment. The selected glidepath is believed to be appropriate for a large majority of our shareholders. Glidepath assumptions The ideal glidepath incorporates the changing human capital of an individual over their working years and in retirement, taking into account reasonable assumptions for starting salary, raises, contributions, and retirement age (among other inputs), as shown in Exhibit 4. 5

6 Exhibit 4: Assumptions Variable Assumptions Start working 22 yrs Starting salary $29,414 Salary increase 2.15% Contribution rate 7% (inclusive of employee contributions and employer match) Retirement age 65 yrs Start working: The Multi-Asset Investments Team considers a person who begins saving for retirement at age 22 at a 7% contribution rate. Starting salary: The starting salary is $29,414 at age 22 and grows with the salary increase rate. The salary amount is near the average reported by the Employee Benefit Research Institute (EBRI). Assumptions regarding starting salary are significant in that projected Social Security benefits may differ substantially in proportion to varying salary levels. Resulting variations in Social Security payments as a percentage of total capital can have a strong bearing on an appropriate glidepath. Salary increase: We assume an annual salary increase during one s working years that is equal to the current consensus forecast for the average inflation rate expected over the next 10 years by the Society of Professional Forecasters. Contribution rate: The EBRI has shown that the average American contributes approximately 7% of his or her salary each year to a retirement account, inclusive of employer and employee contributions. We use the EBRI average. Proprietary asset class expected returns The Multi-Asset Investments Team, working in concert with the Asset Allocation Committee, gauges the implied returns in market prices given available information using proprietary asset class expected returns. These capital market returns consider current market information such as forecasted economic growth, stock market dividend yields, stock buybacks, valuations, profit margins, forecasted earnings growth, bond yields and capitalization rates on real estate investment properties. The forecasted expected returns are integral to the design of the strategic asset allocations for our Lifecycle Funds glidepath and also serve as important inputs into our tactical asset allocation process discussed on page 15. We have two equity return models: (1) a Components of Return Model, and (2) a Dividend Discount/Share Buyback Model. The Components of Return Model builds an expected return assuming a long-run sales real growth estimate, a long-run inflation estimate, and current dividend yields. Current Price/Earnings (P/E) valuations and profit margins are assumed to revert gradually over the long run to the median historical valuation level and profit margins. The Dividend Discount/Share Buyback Model establishes a net present value (NPV) based on future dividends and/or share repurchases. Within fixed income, we also have two long-run total return models: (1) a Current Yield to Maturity Model, which historically is a reliable predictor of future returns over the next decade, and (2) a Yield to Maturity Spread Model, which assumes that current yields revert to historical median values over the long run. For both the equity and fixed-income asset classes, the ultimate expected return is derived as a straight arithmetic mean of the two models used in each asset class. 6

7 With regard to direct real estate, to forecast expected returns we look at current capitalization rates and use a dividend discount model. A capitalization rate is the ratio of a property s net operating income (NOI) to its market value, akin to an inverse P/E ratio. Current capitalization rates provide insight on the expected NOI, and thus the expected cash flows, from the typical types of properties we target for investment. Our dividend discount model for real estate establishes a NPV based on these expected cash flows. Our expected return models are evaluated, updated, and discussed within the Multi-Asset Investments Team on a monthly basis. Monte Carlo analysis of glidepath The Multi-Asset Investments Team performs Monte Carlo simulation analyses of the glidepath for the Lifecycle Funds. The purpose of the Monte Carlo analysis is to consider a range of possible outcomes for the selected glidepath under a variety of market conditions. The analysis accomplishes this by simulating an individual investor in his or her age-appropriate target-date glidepath during the accumulation years who then makes systematic withdrawals in retirement. We borrow the framework of a stochastic return model from the American Actuarial Association (AAA), which is used to test the capital adequacy of insurance products (Ahlgrim et al., 2009). We modify the input parameters of this model to reflect our view of long-run expected returns. Specifically, we alter the input parameters to achieve our proprietary view of the long-run expected returns for stocks, bonds and direct real estate (please see proprietary asset class expected returns section on page 6 for a complete discussion of these models). In the AAA stochastic return model, interest-rate paths are generated using a stochastic log volatility model with mean reversion. A series of stochastic equations link evolving interestrate yield curves with stock, bond and direct real estate returns. In total, the model generates 10, year return series. The 73-year period reflects returns over the complete savings and retirement time horizons (from age 22 through age 95). There are many benefits to using this stochastic return model compared with simply resampling the past 100 years or so of recorded capital market returns: The stochastic return series exhibit mean reversion properties, are calibrated to our expected return models, and represent a broader range of potential outcomes. Each return series is a realistically modeled, potential sequence of future stock, bond and direct real estate returns. The Monte Carlo simulations assume a mortality table/life expectancy shifted about three years forward from the 2000 Census mortality table, in recognition of the growing life expectancy of the U.S. population and TIAA participant base. Our long-run inflation assumption is taken from the latest forecasts provided by the Society of Professional Forecasters administered by the Philadelphia Federal Reserve Bank. Glidepath design metrics The Multi-Asset Investments Team assumes individuals invest in their appropriate target-date retirement fund during their working years and make systematic withdrawals in retirement. We measure the following three metrics: 1. Range of potential savings outcomes on the date of retirement 2. Potential volatility of the portfolio under short-term market movements on the date of retirement 3. Probability of having income for one s lifetime in retirement if one pursues a systematic withdrawal program 7

8 The rationale for each of these three metrics is listed below: 1. Although our glidepath is designed as a through retirement asset allocation experience, we are sensitive to the savings outcomes on the date of retirement. We chose the first key metric, which focuses on the outcomes on the date of retirement (size and shape of retirement nest egg on the date of retirement), because individuals may make retirement lifestyle decisions for the duration of their retirement given their amount of savings on the date of retirement. Although we have a through glidepath, we encourage individuals to seek financial advice as they approach their retirement date. Throughout the analysis, return outcomes are measured in relation to the level of an investor s projected accumulated savings relative to projected ending salary. This measure of retirement wealth is used to estimate the likelihood that a Lifecycle Fund investor s projected nest egg is of sufficient size to support an extended period of regular withdrawals over the course of retirement without depleting savings. For each simulation, we look not only at the median outcome, but also at the range of nest egg outcomes for 2 and 3 standard deviation events. Tail risk, the likelihood of unusual market events (more than 3 standard deviations), is also a consideration. 2. After the dramatic volatility during the Credit Crisis in 2008, the Multi-Asset Investments Team decided to further refine its investment process to better assess the potential riskiness of the portfolio under short-term market movements on the date of retirement. We recognize the need to balance the long-term goal of remaining properly invested in a diversified portfolio that can generate sufficient income during the drawdown phase in retirement with the short-term goal of having a portfolio that can withstand the volatility of a market crisis without causing investors to exit their Lifecycle Funds portfolio in a panic if such an event were to occur. We also recognize that some of our investors may sell their Lifecycle Fund on the date of retirement and invest in other vehicles. These exiting investors will be exposed to the particular value of the market on their retirement date. For these reasons, we chose to add the additional metric of estimating the worst possible 12-month performance of the Lifecycle Funds on the date of retirement. We measure this by considering the historically worst 12-month performance for each of our asset classes together with our forward-looking estimates of asset class volatility. 3. The third key metric we measure is the probability of the Lifecycle Funds glidepath to be able to generate income for an individual s lifetime in retirement via a systematic withdrawal program. We acknowledge that our participants may have other sources of income in retirement, the most likely of which would be Social Security benefits. Additional simulations are run looking at income replacement ratios for each hypothetical shareholder s lifetime, measuring the probability of being able to meet certain income thresholds across various stages of retirement. In these cases, we not only simulate market returns, but also factor in the growing life expectancy of the U.S. population. Planned progression of asset allocation changes At 45 years to the target retirement date, a Lifecycle Fund begins with a 95% allocation to equity, a 2.5% allocation to fixed-income investments and a 2.5% allocation to direct real estate. From a Human Capital/Financial Capital framework perspective, this beginning asset allocation reflects the greater human capital and longer time horizon of investors in their 20 s and 30 s. While the direct real estate allocation is held steady at 2.5% throughout the glidepath, the equity allocation is reduced by 0.25% per year and the fixed-income allocation is increased by an offsetting amount until the target retirement date is less than 25 years 8

9 away, at which point the equity exposure is reduced, and fixed-income exposure is increased, by 1.7% per year. In this way, the portfolio makes a gradual transition to a more conservative asset allocation mix over time. To reduce trading costs and to limit potential exposure to large single-day market movements, the transition from one target asset allocation to the next is carried out over a long time period, making use of available daily fund cash flows to the extent possible to realize the desired change in allocation. At the point of the target retirement date, the allocation to equity, fixed income and direct real estate is 47.5%, 50% and 2.5%, respectively. During the initial years of retirement, the direct real estate exposure is maintained while equity exposure is further reduced by 1% per year until a final allocation of 37.5% equities, 60% fixed income and 2.5% direct real estate is reached 10 years after the target retirement date. This portion of the glidepath, which maintains a substantial allocation to equities in the initial stages of retirement, was determined based on consideration of the need for capital appreciation as well as income in the early retirement years. The allocation of the Lifecycle Funds during this stage of the glidepath is particularly significant given the growing life expectancy of the U.S. population, and the need to balance market and longevity risk. Approximately seven to 10 years after each Lifecycle Fund attains its respective retirement date, the Lifecycle Funds Board may authorize its merger into the Lifecycle Retirement Income Fund, which maintains a constant allocation of 37.5% equity, 60% fixed income and 2.5% direct real estate. Composition of benchmark The performance of the Lifecycle Funds is evaluated in relation to a series of composite benchmarks that represent the five primary equity and fixed income groupings ( market sectors ) in which each of the funds invests. Each composite benchmark is comprised of five underlying indices, representing each of these five equity and fixed income market sectors. Due to the lack of availability of a real estate benchmark that is updated on a daily basis, the real estate market sector is not represented within the Lifecycle Funds composite benchmark. Market Sectors U.S. Equity International Equity Fixed-Income Short-Term Fixed-Income Inflation-Protected Assets Benchmark Russell 3000 Index MSCI ACWI ex-usa Investable Market Index (IMI) Bloomberg Barclays U.S. Aggregate Bond Index Bloomberg Barclays U.S. 1 3 Year Government/Credit Index Bloomberg Barclays U.S. Treasury Inflation-Protected Securities (TIPS) 1 10 Year Index Composition of portfolio of underlying investments The asset allocation strategy and selection of underlying investments for the Lifecycle Funds has been determined on the basis of their contribution to return outcomes during periods of savings and over the course of an individual s retirement. When considering the composition of funds to ultimately include within the equity, fixed-income and direct real estate asset classes, we categorize funds into one of six market sectors: 1) domestic equity; 2) international equity (including both developed and emerging markets); 3) fixed income (including core, core plus, international developed, high yield, and emerging-markets debt); 4) short-term fixed income; 5) inflation-protected assets and 6) direct real estate. Additional detail regarding the composition of holdings within each of the six primary asset class groupings is provided below. 9

10 Equity portfolio composition The allocation to equities within each Lifecycle Fund is divided among domestic and international equities, with a split of 70% domestic and 30% international as a percentage of total equity assets. The international equity component is inclusive of both developed- and emerging-markets equities. Based on studies covering a range of time periods, the reduction in risk associated with international diversification historically has been maximized in a range of 25% to 30% international exposure. We believe that 30% is the appropriate international exposure, as the investable universe is growing faster outside the U.S. The diversification benefit of international equity is depicted in Exhibit 5 below, based on monthly returns from 1979 through September For any level of international equity below 40%, risk is less than or similar to that of an all-u.s. equity portfolio, as the diversification benefits of international equity outweigh the higher risk of the international equity asset class. The overall risk level is particularly low in the 25% to 35% range. Exhibit 5: Portfolio risk as function of % international equity ( ) Russell 3000 for U.S., MSCI ACWI ex-usa IMI for international 18% 17% Risk 16% 15% 14% 0% 20% 40% 60% 80% 100% % International Domestic equity portfolio composition The domestic equity portfolio provides diversified, style-consistent exposure across large-, mid-, and small-cap stocks, as well as to growth- and value-oriented companies. The selection and relative weightings of the mutual funds in this market sector are designed to represent an overall exposure similar in market capitalization, sector exposure, style, and risk characteristics of the broader domestic equity market, as represented by the Russell 3000 Index. Funds within the U.S. Large-Cap Growth and Value sectors provide exposure to both a fundamental and a quantitative approach to equity investing, offering further diversification. The mix of domestic funds in which the Lifecycle Funds invest represent distinct, targeted strategies that employ a range of stock selection techniques, using fundamental (e.g., Large- Cap Growth Fund and Large-Cap Value Fund) and quantitative stock selection (e.g., Enhanced Large-Cap Growth Index Fund and Enhanced Large-Cap Value Index Fund). By maintaining two approaches and strategies with differing levels of active risk, we seek to improve the consistency of returns of the Lifecycle Funds relative to their benchmarks over time. 10

11 Exhibit 6: Equity only glidepath 90 Emerging Markets Equity International Developed Markets Equity Allocations % U.S. Equity Years to Retirement The Lifecycle Funds may invest in the following domestic equity funds: TIAA-CREF Funds U.S. Equity Asset class category Benchmark Growth & Income Fund Domestic, Large-Cap Blend Equity S&P 500 Index Large-Cap Growth Fund Domestic, Large-Cap Growth Equity Russell 1000 Growth Index Enhanced Large-Cap Growth Index Fund Domestic, Large-Cap Growth Equity Russell 1000 Growth Index Large-Cap Value Fund Domestic, Large-Cap Value Equity Russell 1000 Value Index Enhanced Large-Cap Value Index Fund Domestic, Large-Cap Value Equity Russell 1000 Value Index Small/Mid-Cap Equity Fund Domestic, Small-Cap Blend Equity Russell 2500 Index Small-Cap Equity Fund Domestic, Small-Cap Blend Equity Russell 2000 Index International equity portfolio composition The international equity portion provides exposure to developed- and emerging-markets foreign equities, and is designed to provide overall regional exposure similar to stocks contained in the MSCI ACWI ex-usa Investable Market Index (IMI). Within developed markets, we have two funds that employ a fundamental approach to investing (International Equity Fund and International Opportunities Fund), and two funds that have an independent quantitative approach (Enhanced International Equity Index Fund and International Small-Cap Equity Fund). These two approaches to investing provide independent, uncorrelated sources of excess return. This benefit provides Lifecycle Funds investors with smoother relative performance over time. The inclusion of emerging markets provides an additional source of diversification while allowing the funds to potentially benefit from the rapid economic growth associated with emerging markets. 11

12 The Lifecycle Funds may invest in the following international equity funds: TIAA-CREF Funds International Equity Asset class category Benchmark International Equity Fund Foreign, Large-Cap Blend Equity MSCI EAFE Index Enhanced International Equity Index Fund Foreign, Large-Cap Blend Equity MSCI EAFE Index International Opportunities Fund Foreign, Large-Cap Blend Equity MSCI ACWI ex-usa Index Emerging Markets Equity Fund Diversified Emerging Markets Equity MSCI Emerging Markets Index International Small-Cap Equity Fund Foreign, Small-Cap Blend Equity MSCI ACWI ex-usa Small-Cap Index Fixed-income portfolio composition Our Fixed-Income market sector includes investments in our Core (Bond Fund), Core Plus (Bond Plus Fund), International Developed (International Bond Fund), High-Yield (High-Yield Fund), and Emerging-Markets Debt (Emerging-Markets Debt Fund) strategies. When the investor reaches age 50, the portfolio begins to pare back exposure to credit risk, duration risk, and inflation risk by purchasing investments in the Short-Term Bond Fund (categorized in a market sector of a similar name) and our Inflation-Linked Bond Fund (categorized in a market sector called Inflation-Protected Assets). The changing composition of fixed-income investments from early periods of saving through retirement reflects investors greater need for interest-rate and inflation protection and seeks to provide stability of returns for investors as they approach retirement and during their retirement years. The Lifecycle Funds may invest in the following fixed-income funds: TIAA-CREF Funds Fixed-Income Asset class category Benchmark Bond Fund Intermediate-Term Bond Bloomberg Barclays U.S. Aggregate Bond Index Bond Plus Fund Intermediate-Term Bond Bloomberg Barclays U.S. Aggregate Bond Index International Bond Fund World Bond Bloomberg Barclays Global Aggregate ex-usd (Hedged) Index High-Yield Fund High-Yield Bond BofA Merrill Lynch BB-B U.S. Cash Pay High Yield Constrained Index Emerging Markets Debt Fund Emerging-Markets Debt JPMorgan Emerging Markets Bond Global Diversified (EMBI GD) Index TIAA-CREF Funds Short-Term Fixed-Income Asset class category Benchmark Short-Term Bond Fund Short-Term Bond Bloomberg Barclays U.S. 1-3 Year Gov t/credit Bond Index TIAA-CREF Funds Inflation-Protected Assets Asset class category Benchmark Inflation-Linked Bond Fund Inflation-Protected Bond Bloomberg Barclays U.S. Treasury Inflation Protected Securities (TIPS) 1-10 Year Index* * Effective January 1, Prior to this date, the benchmark for this market sector was the Barclays U.S. Treasury Inflation Protected Securities (TIPS) Index (Series-L). 12

13 Exhibit 7: Fixed-income only glidepath Allocations 60% 50% 40% Inflation-Linked Bond Fund Short-Term Bond Fund International Bond Fund Bond Plus Fund Bond Fund 0 High Yield Fund Emerging Markets Debt Fund Years to Retirement Direct real estate portfolio composition The Lifecycle Funds will maintain a strategic target allocation of 2.5% to direct real estate throughout the glidepath. The allocation to direct real estate consists of exposure to directly held U.S. core commercial real estate properties that are well-located in select targeted cities/markets. Core real estate assets are typically well-occupied with high-quality tenants and are typically located in high-barrier-to-entry markets (e.g., New York, Washington, D.C., San Francisco). This direct real estate exposure is obtained via the Real Property Fund LP which seeks to provide broad diversification by property type (i.e., office, industrial, retail and multi-family/apartment properties) and geographic region. The Real Property Fund s portfolio composition is almost entirely comprised of properties. The portfolio has only 2% - 5% in liquid assets in order to maximize the diversification benefits directly held real estate can provide. Direct real estate has exhibited noticeably low performance correlations to equities and fixed income, and has provided relatively stable and high cash yields (stemming from the cash flows generated by long-term leases) and attractive risk-adjusted returns (i.e., attractive absolute returns, with much lower volatility relative to equities 6 ). By including direct real estate we seek to benefit from its potential to improve risk-adjusted returns, as well as its potential to be a particularly effective diversifier. The anticipated diversification benefits further promote consistency of returns and increased downside protection (this is of particular importance in the portion of the glidepath approaching retirement and in retirement). Direct Real Estate Asset class category Benchmark TIAA-CREF Real Property Fund LP* Direct Real Estate NCREIF Property Index Open End Funds *This is a private fund not available for sale to individual investors. 13

14 Benefits of active management We believe that investing in actively managed equity, fixed-income and direct real estate funds provides improved expected risk-adjusted returns compared to passive approaches that seek only to replicate the performance of an index. Active risk Historically, the TIAA-CREF Lifecycle Funds have had an active risk level of between approximately 1.5% and 2.5% annual tracking error. Annual tracking error is a measure of the targeted, expected, or realized variation (+/- 1 standard deviation) of a fund s performance relative to its benchmark. The lowest tracking errors (approximately 1.5%) are found in the Lifecycle Retirement Income Fund, while the highest tracking errors are found in the funds with the highest exposure to equities (e.g., the Lifecycle 2060 Fund). Tracking errors of actively managed funds can vary significantly depending on current financial market conditions. During periods of economic and market stress, realized tracking errors can double or even triple in reaction to volatile markets. The overall tracking errors of the Lifecycle Funds relative to their respective benchmarks are a function of the tracking errors of each underlying fund relative to their respective benchmarks, as well as the degree to which Lifecycle Fund asset allocations differ from target allocations. At the aggregate Lifecycle Fund level, the combined tracking error of the underlying mutual funds is less than a linear combination of the tracking errors because of a degree of independence of each underlying manager. In other words, the relative performance of the Lifecycle Fund can benefit from a portfolio of underlying managers who exhibit a degree of independence in their relative performance. In addition, active risk might be extended somewhat as tracking error potentially increases through the implementation of tactical asset allocation positions (described later in this paper). Alpha risk budget The Multi-Asset Investments Team monitors how much each individual portfolio manager can influence the relative performance of each Lifecycle Fund. We term this the Lifecycle Funds alpha risk budget. Our alpha risk budget provides a framework for determining allocations among funds within each asset class, taking into consideration the relative impact that a +/ 1 standard deviation variation in the relative performance of any individual underlying fund may have on a Lifecycle Fund s performance. The Lifecycle Funds alpha risk budget helps determine the broader weightings across each of the underlying funds in each sub-asset class. This includes determining the relative weightings of the fundamental and quantitative strategies within U.S. large-cap equities and international developed markets, taking into consideration characteristics inherent in each approach and the resulting impact on expected variability of a Lifecycle Fund s excess returns. In specifying our alpha risk budget, we seek to limit the contribution of any single strategy to one-quarter or less of the overall risk budget. 14

15 The following table provides an example of how our risk budgeting framework may be used to help determine relative weightings in underlying funds: Sample alpha risk budget for the Lifecycle 2020 Fund Market sector Weight Annualized tracking error Marginal contribution to tracking error Risk budget U.S. Equity Funds Large-Cap Growth 7.0% 3.9% 0.27% 14% Large-Cap Value 7.3% 2.6% 0.19% 10% Enhanced Large-Cap Growth 6.2% 1.1% 0.07% 4% Enhanced Large-Cap Value 6.4% 1.4% 0.09% 5% Growth & Income 8.0% 2.5% 0.20% 10% Small/Mid Cap Equity 1.7% 1.9% 0.03% 2% Small Cap Equity 2.3% 2.0% 0.05% 2% Subtotal: 38.9% International Equity Funds International Opportunities 4.2% 5.1% 0.21% 11% International Equity 4.4% 5.4% 0.24% 12% Enhanced International Equity 5.2% 1.4% 0.07% 3% Emerging Markets Equity 3.0% 3.4% 0.10% 5% International Small Cap Equity 0.0% 3.0% 0.00% 0% Subtotal: 16.7% Direct Real Estate Funds Real Property Fund 0.7% 5.0% 0.03% 2% Subtotal: 0.7% Fixed-Income Funds Bond 18.2% 0.9% 0.16% 8% Bond Plus 11.3% 1.4% 0.16% 8% International Bond 1.7% 1.9% 0.03% 2% High Yield 1.7% 1.1% 0.02% 1% Emerging Markets Debt 1.7% 1.9% 0.03% 2% Subtotal: 34.6% Short-Term Fixed-Income Fund Short Term 4.6% 0.5% 0.02% 1% Subtotal: 4.6% Inflation-Protected Assets Fund Inflation-Linked Bond 4.6% 0.6% 0.03% 1% Subtotal: 4.6% Grand total 100.0% Note: This sample represents the TIAA-CREF Lifecycle Retirement Income Fund s holdings prior to investment in direct real estate. 15

16 Tactical asset allocation The Asset Allocation Committee, referenced earlier in the document, oversees a modest tactical asset allocation effort in the Lifecycle Funds. Our investment process considers the direction of the economy, market valuation, investor sentiment, and the views of the portfolio managers of the underlying funds, in addition to a range of market metrics. The Committee regularly reviews the following analytical efforts in analyzing any potential tactical shifts, along with information from a variety of other sources: TIAA-CREF Lifecycle Funds proprietary long-run expected return models discussed earlier. Our long-run expected return models have been developed and are maintained by our Lifecycle fund-of-funds team and are updated monthly. These models consider current market information to forecast an average return over the next 10 years. Large moves in the long-run expected returns can signal significant short-run valuation changes. Our equity expected return model considers long-run forecasts for real growth of sales, inflation, dividend yield, and mean reversion of P/E ratios and profit margins. Our fixed-income models consider current yields and reversion of yields to normalized long-run spreads. Note that due to the illiquid/long-term nature of commercial real estate, the real estate market sector is not within the scope of the tactical asset allocation program. Consistent monitoring effort of key tactical metrics. The Asset Allocation Committee monitors key tactical metrics where time horizons are much shorter, often confined to weeks or months. These key tactical metrics support views on the economy, market valuation and investor sentiment. Examples include valuation, earnings revisions, technical measures (e.g., moving averages, VIX), sentiment (e.g., short-term bulls vs. bears), and economic growth (e.g., ISM, PMI, company surveys, consumer spending). The degree of tactical asset allocation implemented is modest relative to the impact of the strategic glidepath and the relative performance of the underlying portfolio managers. We expect the performance of the Lifecycle Funds to be primarily driven by the overall asset allocations of the strategic glidepath and the relative performance of the underlying funds, as it has been historically. At most, we anticipate tactical asset allocation to account for about 10% to 20% of the relative performance of the Lifecycle Funds versus their composite benchmark in any given year. The remaining 80% to 90% of relative performance of the Lifecycle Funds remains a function of the performance of the portfolio managers of the underlying funds relative to their respective sub-asset categories, as well as the strategic glidepath decision. The following are guidelines and expected impacts of the Lifecycle Funds tactical asset allocation practices: Holding period for tactical asset allocation trades in portfolio: We plan to hold most tactical positions for 6 to 12 months. Market sector limits: We do not expect tactical decisions to exceed +/-5% of target market-sector allocations. Tracking error estimates: We expect tactical allocation to add about 0.15% of tracking error annually to the Lifecycle Funds. As discussed earlier, without the impact of tactical asset allocation, our tracking errors in the Lifecycle Funds range from about 1.5% (the Lifecycle Retirement Income Fund) to about 2.5% (the Lifecycle 2060 Fund) under normal market conditions. Under periods of lower or higher volatility, our tracking errors will vary accordingly. This tracking error is primarily generated by the active management of the underlying funds to their benchmarks. Adding the tactical asset allocation tracking error to 16

17 our overall tracking error will increase it by about 0.15%, assuming a correlation of 0.25 (conservatively high) between the tactical asset allocation decisions and the positioning of the underlying funds. Risk budget: The tactical asset allocation program ranges from about 10% to 20% of our alpha risk budgets. The remaining 80% to 90% of the alpha risk budget is a function of how well the portfolio managers of the underlying funds held by the Lifecycle Funds perform relative to their sector-specific benchmarks. Customization of tactical decision per Lifecycle Fund: When appropriate, we adjust the tactical decisions based on the overall investment goal of each Lifecycle Fund. For example, our tactical decisions could be more aggressive in the Lifecycle Funds for people early in their working careers and thus with the longest investment time horizons, but less aggressive for people closer to retirement. Estimated turnover: We expect the tactical asset allocation decisions to add 5% to 10% to the annual turnover of the funds. This is based on the planned holding period of tactical portfolio decisions, current cash flows, and implementation speed. Historically, the turnover for the Lifecycle Funds has ranged between 10% and 25%, varying from one fund to another based on month-end rebalancing and client flows. Monitoring of underlying funds The Multi-Asset Investments Team closely monitors the portfolios, relative performance, and portfolio compositions of the underlying funds. In addition to monitoring the underlying funds individually, the Team also monitors the combined effect of collective ownership of the stocks and bonds across all of the underlying funds. Through an x-ray analysis of the underlying funds which looks through to the individual stocks and bonds held across all of the underlying funds the Multi-Asset Investments Team monitors sector biases and various risk management statistics. The individual performance of the underlying funds is discussed by the Asset Allocation Committee. The CIOs of both Equities and Fixed Income are present to address any concerns or issues. The performance is discussed in relation to individual fund objectives, risk parameters, market climate, and outlook and positioning. Tactical shifts may be initiated by the Asset Allocation Committee if determined that the current market environment is not conducive to a particular fund s investment strategy. Any long-term concerns regarding performance or suitability of the underlying fund related to the fund s investment process, objectives, or risk profiles are escalated to senior management. Monthly rebalancing Consistent management of portfolio allocations in relation to target allocations helps reduce the funds vulnerability to market bubbles and avoids having the funds risk profiles drift from targeted levels. On a monthly basis, the Multi-Asset Investments Team rebalances the relative weights of each Lifecycle Fund s holdings to within approximately 1% to 2% (depending on market volatility) of its target allocation as appropriate per the planned glidepath. For example, at the end of each month, the Lifecycle Retirement Income Fund is rebalanced to an allocation near its 37.5% equities, 60% fixed-income and 2.5% direct real estate target. Monthly rebalancing is achieved using cash flows to the extent possible, in order to minimize portfolio turnover. If sufficient rebalancing cannot be achieved using daily contributions and redemptions, the Multi-Asset Investments Team will, on approximately a monthly basis, implement trades to bring each fund back toward its target allocation. These trades leverage the use of a daily 17

18 mean-variance optimization process that assists in allocating cash inflows each day. To the largest extent possible, the optimization algorithm lowers the tracking error of the portfolio relative to its target benchmark using available cash inflows. For instance, rather than simply filling the positions that are furthest from their target allocations first, the optimizer takes into account the relative risk of various positions. Each Lifecycle Fund is considered on an individual basis in the optimizer relative to fund-specific risk characteristics. Over time, the daily optimization process helps minimize trading as well. Use of this daily optimizer allows for more consistent performance within the Lifecycle Funds through effective allocation of available cash flows. Cash management On days when the Lifecycle Funds receive large cash inflows, the Multi-Asset Investments Team invests the cash to be allocated to the U.S. and foreign-developed equity markets in ETFs that follow these markets. At the market close, we sell the ETFs and invest the proceeds in our underlying mutual funds as appropriate. This ensures that the Lifecycle Funds remain as fully invested as possible. Lifecycle Retirement Income Fund and advice offering nearing retirement Each target-date Lifecycle Fund is managed such that the ending asset allocation matches the asset allocation of the Lifecycle Retirement Income Fund 10 years after the target date. Because the asset allocation continues to glide for 10 years after retirement, our TIAA-CREF Lifecycle Funds have what is termed a through retirement glidepath. The asset allocation of the Lifecycle Retirement Income Fund remains static over time. This fund is appropriate for participants in retirement and serves as the ending portfolio. Each target-date fund may be merged with the Lifecycle Retirement Income Fund seven to 10 years after the date of retirement. We believe this post-retirement glidepath is best suited for those that may not make an active investment decision at retirement; however, we encourage all participants to review their savings goals and investment options at retirement. TIAA encourages shareholders to seek advice support from a financial advisor as they approach retirement. The discussions should include retirement goals, income needs, and risk tolerances, as well as wealth and estate planning needs. Advice offering reviews may consider a number of options, such as: 1. Annuitizing all or a portion of retirement savings for lifetime income 2. Continuing in the Lifecycle Fund for all or a portion of retirement savings to continue to have assets professionally managed across asset classes in a prudent manner 3. Other investment options Fiduciary oversight of the TIAA-CREF Multi-Asset Investments Team Several committees oversee the fiduciary decisions made within our Lifecycle Funds strategies: W W The glidepath, asset allocation, fund selection, and all other key elements of the investment decisions of the TIAA-CREF Lifecycle Funds are overseen by the Asset Allocation Committee (described earlier). The investment professionals on the Asset Allocation Committee are also involved in the management of the CREF Stock Account and the TIAA General Account, so decisions made within our Lifecycle Funds portfolios are generally consistent with these pension annuity accounts, which are among TIAA s largest and longest-standing products. This committee meets monthly and as needed based on market circumstances. 18