![]()
Betterment’s portfolio construction methodology
Betterment’s portfolio construction methodology Learn more about the process that underpins all the portfolios we build on behalf of customers. Table of contents Introduction Global diversification and asset allocation Portfolio optimization Tax management using municipal bonds The Value Tilt portfolio strategy The Innovative Technology portfolio strategy The Socially Responsible Investing portfolio strategies Conclusion Citations I. Introduction Betterment builds investment portfolios designed to help you make the most of your money and live the life you want. This guide lays out our portfolio construction process, one informed by real-world evidence and systematic decision-making. The Betterment Core portfolio serves as the foundation for all of the globally-diversified portfolios we construct. From there, specific adjustments are applied to other portfolios based on the investment objective of their particular strategies. These adjustments include additional allocations to value-focused or innovative stocks, or adherence to Socially Responsible Investing (SRI) criteria. For more information on the third-party portfolios we offer, such as the Goldman Sachs Smart Beta portfolio, see their respective pages and disclosures. When building a portfolio, any investment manager faces two main tasks: asset class selection and portfolio optimization. We detail our approach to these in the sections that follow. Our fund selection process, while equally as important, is covered in a separate methodology. II. Global diversification and asset allocation An optimal asset allocation is one that lies on the efficient frontier, which is a set of portfolios that seek to achieve the maximum objective for any given feasible level of risk. The objective of most long-term portfolio strategies is to maximize return for a given level of risk, which is measured in terms of volatility—the dispersion of those returns. In line with our approach of making systematic decisions backed by research, Betterment’s asset allocation is based on a theory by economist Harry Markowitz called Modern Portfolio Theory.1 A major tenet of Modern Portfolio Theory is that any asset included in a portfolio should not be assessed by itself, but rather, its potential risk and return should be analyzed as a contribution to the whole portfolio. Modern Portfolio Theory seeks to maximize expected return given an expected risk level or, equivalently, minimize expected risk given an expected return. Other forms of portfolio construction may legitimately pursue other objectives, such as optimizing for income, or minimizing loss of principal. Asset class selection Our approach to asset allocation starts with a universe of investable assets, which could be thought of as the “global market” portfolio.2 To capture the exposures of the asset classes for the global market portfolio, we evaluate available exchange-traded funds (ETFs) that represent each class in the theoretical market portfolio. We base our asset class selection on ETFs because this aligns portfolio construction with our fund selection methodology. All of our portfolios are constructed of the following asset classes: Stocks U.S. stocks International developed market stocks Emerging market stocks Bonds U.S. short-term treasury bonds U.S. inflation-protected bonds U.S. investment-grade bonds U.S. municipal bonds International developed market bonds Emerging market bonds We select U.S. and international developed market stocks as a core part of the portfolio. Historically, stocks exhibit a high degree of volatility, but provide some degree of inflation protection. Even though significant historical drawdowns, such as the global financial crisis in 2008 and pandemic outbreak in 2020, demonstrate the possible risk of investing in stocks, longer-term historical data and our forward expected returns calculations suggest that developed market stocks remain a core part of any asset allocation aimed at achieving positive returns. This is because, over the long term, developed market stocks have tended to outperform bonds on a risk-adjusted basis. To achieve a global market portfolio, we also include stocks from less developed economies, called emerging markets. Generally, emerging market stocks tend to be more volatile than U.S. and international developed stocks. And while our research shows high correlation between this asset class and developed market stocks, their inclusion on a risk-adjusted basis is important for global diversification. Note that we exclude frontier markets, which are even smaller than emerging markets, due to their widely-varying definition, extreme volatility, small contribution to global market capitalization, and cost to access. We incorporate bond exposure because, historically, bonds have a low correlation with stocks, and they remain an important way to dial down the overall risk of a portfolio. To promote diversification and leverage various risk and reward tradeoffs, we include exposure to several asset classes of bonds. Asset classes excluded from Betterment portfolios While Modern Portfolio Theory would have us craft a portfolio to represent the total market, including all available asset classes, we exclude some asset classes whose cost and/or lack of data outweighs the potential benefit gained from their inclusion. Our portfolio construction process excludes commodities and natural resources asset classes. Specifically, while commodities represent an investable asset class in the global financial market, we have excluded commodities ETFs because of their low contribution to a global stock/bond portfolio's risk-adjusted return. In addition, real estate investment trusts (REITs), which tend to be well marketed as a separate asset class, are not explicitly included in our portfolios. We do provide exposure to real estate, but as a sector within stocks. Adding additional real estate exposure by including a REIT asset class would overweight the exposure to real estate relative to the overall market. Incorporating awareness of a benchmark Before 2024, we managed our portfolios in a “benchmark agnostic” manner, meaning we did not incorporate consideration of global stock and bond indices in our portfolio optimization, though we have always sought to optimize the expected risk-adjusted return of the portfolios we construct for clients. The “risk” element of this statement represents volatility and the related drawdown potential of the portfolio, but it could also represent the risk in the deviation of the portfolio’s performance relative to a benchmark. In an evolution of our investment process, in 2024 we updated our portfolio methodology to become “benchmark aware,” as we now calibrate our exposures based on a custom benchmark that expresses our preference for diversifying across global stocks and bonds. A benchmark, which comes in the form of a broad-based market index or a combination of indices, serves as a reference point when approaching asset allocation, understanding investment performance, and aligning the expectations of portfolio managers and clients. In our case, we created a custom benchmark that most closely aligns with our future expectations for global markets. The custom benchmark we have selected is composed of: The MSCI All Country World stock IMI index (MSCI ACWI IMI) The Bloomberg U.S. Universal Bond index The S&P US Treasury Bond 0-1 Year Index (for <40% stock allocations) Our custom benchmark is composed of 101 risk levels of varying percentage weightings of the stock and bond indexes, which correspond to the 101 risk level allocations in our Core portfolio. At low risk levels (allocations that are less than 40% stocks), we layer an allocation to the S&P US Treasury Bond 0-1 Year Index, which represents short-term bonds, into the blended benchmark. We believe that incorporating this custom benchmark into our process reinforces the discipline of carefully evaluating the ways in which our portfolios’ performance could veer from global market indices and deviate from our clients’ expectations. We have customized the benchmark with 101 risk levels so that it serves clients’ varying investment goals and risk tolerances. As we will explore in the following section, establishing a benchmark allows us to apply constraints to our portfolio optimization that ensures the portfolio’s asset allocation does not vary significantly from the geographic and market-capitalization size exposures of a sound benchmark. Our benchmark selection also makes explicit that the portfolio delivers global diversification rather than the more narrowly-concentrated and home-biased exposures of other possible benchmarks such as the S&P 500. III. Portfolio optimization As an asset manager, we fine-tune the investments our clients hold with us, seeking to maximize return potential for the appropriate amount of risk each client can tolerate. We base this effort on a foundation of established techniques in the industry and our own rigorous research and analysis. While most asset managers offer a limited set of model portfolios at a defined risk scale, our portfolios are designed to give customers more granularity and control over how much risk they want to take on. Instead of offering a conventional set of three portfolio choices—aggressive, moderate, and conservative—our portfolio optimization methods enable our Core portfolio strategy to be customized to 101 different stock-bond risk levels. Optimizing portfolios Modern Portfolio Theory requires estimating variables such as expected-returns, covariances, and volatilities to optimize for portfolios that sit along an efficient frontier. We refer to these variables as capital market assumptions (CMAs), and they provide quantitative inputs for our process to derive favorable asset class weights for the portfolio strategy. While we could use historical averages to estimate future returns, this is inherently unreliable because historical returns do not necessarily represent future expectations. A better way is to utilize the Capital Asset Pricing Model (CAPM) along with a utility function which allows us to optimize for the portfolio with a higher return for the risk that the investor is willing to accept. Computing forward-looking return inputs Under CAPM assumptions, the global market portfolio is the optimal portfolio. Since we know the weights of the global market portfolio and can reasonably estimate the covariance of those assets, we can recover the returns implied by the market.3 This relationship gives rise to the equation for reverse optimization: μ = λ Σ ωmarket Where μ is the return vector, λ is the risk aversion parameter, Σ is the covariance matrix, and ωmarket is the weights of the assets in the global market portfolio.5 By using CAPM, the expected return is essentially determined to be proportional to the asset’s contribution to the overall portfolio risk. It’s called a reverse optimization because the weights are taken as a given and this implies the returns that investors are expecting. While CAPM is an elegant theory, it does rely on a number of limiting assumptions: e.g., a one period model, a frictionless and efficient market, and the assumption that all investors are rational mean-variance optimizers.4 In order to complete the equation above and compute the expected returns using reverse optimization, we need the covariance matrix as an input. This matrix mathematically describes the relationships of every asset with each other as well as the volatility risk of the assets themselves. In another more recent evolution of our investment process, we also attempt to increase the robustness of our CMAs by averaging in the estimates of expected returns and volatilities published by large asset managers such as BlackRock, Vanguard, and State Street Global Advisors. We weight the contribution of their figures to our final estimates based on our judgment of the external provider’s methodology. Constrained optimization for stock-heavy portfolios After formulating our CMAs for each of the asset classes we favor for inclusion in our portfolio methodology, we then solve for target portfolio allocation weights (the specific set of asset classes and the relative distribution among those asset classes in which a portfolio will be invested) with the range of possible solutions constrained by limiting the deviation from the composition of the custom benchmark. To robustly estimate the weights that best balance risk and return, we first generate several thousand random samples of 15 years of expected returns for the selected asset classes based on our latest CMAs, assuming a multivariate normal distribution. For each sample of 15 years of simulated expected return data, we find a set of allocation weights subject to constraints that provide the best risk-return trade-off, expressed as the portfolio’s Sharpe ratio, i.e., the ratio of its return to its volatility. Averaging the allocation weights across the thousands of return samples gives a single set of allocation weights optimized to perform in the face of a wide range of market scenarios (a “target allocation”). The constraints are imposed to make the portfolio weights more benchmark-aware by setting maximum and minimum limits to some asset class weights. These constraints reflect our judgment of how far the composition of geographic regions within the portfolio’s stock and bond allocations should differ from the breakdown of the indices used in the benchmark before the risk of significantly varied performance between the portfolio strategy and the benchmark becomes untenable. For example, the share of the portfolio’s stock allocation assigned to international developed stocks should not be profoundly different from the share of international developed stocks within the MSCI ACWI IMI. We implement caps on the weights of emerging market stocks and bonds, which are often projected to have high returns in our CMAs, and set minimum thresholds for U.S. stocks and bonds. This approach not only ensures our portfolio aligns more closely with the benchmark, but it also mitigates the risk of disproportionately allocating to certain high expected return asset classes. Constrained optimization for bond-heavy portfolios For Betterment portfolios that have more than or equal to a 60% allocation of bonds, the optimization approach differs in that expected returns are maximized for target volatilities assigned to each risk level. These volatility targets are determined by considering the volatility of the equivalent benchmark. Manually established constraints are designed to manage risk relative to the benchmark, instituting a declining trend in emerging market stock and bond exposures as stock allocations (i.e., the risk level) decreases. Meaning that investors with more conservative risk tolerances have reduced exposures to emerging market stocks and bonds because emerging markets tend to have more volatility and downside-risk relative to more established markets. Additionally, as the stock allocation percentage decreases, we taper the share of international and U.S. aggregate bonds within the overall bond allocation, and increase the share of short-term Treasury, short-term investment grade, and inflation-protected bonds. This reflects our view that investors with more conservative risk tolerances should have increased exposure to short-term Treasury, short-term investment grade, and inflation-protected bonds relative to riskier areas of fixed income. The lower available risk levels of our portfolios demonstrate capital preservation objectives, as the shorter-term fixed income exposures likely possess less credit and duration risk. Clients invested in the Core portfolio at conservative allocation levels will likely therefore not experience as significant drawdowns in the event of waves of defaults or upward swings in interest rates. Inflation-protected securities also help buffer the lower risk levels from upward drafts in inflation. IV. Tax management using municipal bonds For investors with taxable accounts, portfolio returns may be further improved on an after-tax basis by utilizing municipal bonds. This is because the interest from municipal bonds is exempt from federal income tax. To take advantage of this, we incorporate municipal bonds within the bond allocations of taxable accounts. Other types of bonds remain for diversification reasons, but the overall bond tax profile is improved by incorporating municipal bonds. For investors in states with some of the highest tax rates—New York and California—Betterment can optionally replace the municipal bond allocation with a more narrow set of bonds for that specific state, further saving the investor on state taxes. Betterment customers who live in NY or CA can contact customer support to take advantage of state-specific municipal bonds. V. The Value Tilt portfolio strategy Existing Betterment customers may recall that historically the Core portfolio held a tilt to value companies, or businesses that appear to be potentially undervalued based on metrics such as price-to-earnings (P/E) ratios. Recent updates, however, have deprecated this explicit tilt that was expressed via large-, mid-, and small-capitalization U.S. value stock ETFs, while maintaining some exposure to value companies through broad market U.S. stock funds. We no longer favor allocating to value stock ETFs within our portfolio methodology in large part as a result of our adoption of a broad market benchmark, which highlights the idiosyncratic nature of such tilts, sometimes referred to as “off benchmark bets.” We believe our chosen benchmark that represents stocks through the MSCI ACWI IMI, which holds a more neutral weighting to value stocks, more closely aligns with the risk and return expectations of Betterment’s diverse range of client types across individuals, financial advisors, and 401(k) plan sponsors. Additionally, as markets have grown more efficient and value factor investing more popularized, potentially compressing the value premium, we have a marginally less favorable view of the forward-looking, risk-adjusted return profile of the exposure. That being said, we have not entirely lost conviction in the research supporting the prudence of value investing. The value factor’s deep academic roots drove decisions to incorporate the value tilt into Betterment’s portfolios from our company’s earliest days. For investors who wish to remain invested in a value strategy, we have added the Value Tilt portfolio, a separate option from the Core portfolio, to our investing offering. The Value Tilt portfolio maintains the Core portfolio’s global diversification across stocks and bonds while including a sleeve within the stock allocation of large-, mid-, and small-capitalization U.S. value funds. We calibrated the size of the value fund exposure based on a certain target historical tracking error to the backtested performance of the latest version of the Core portfolio. Based on this approach, investors should expect the Value Tilt portfolio to generally perform similarly to Core, with the potential to under- or outperform based on the return of U.S. value stocks. With the option to select between the Value Tilt portfolio or a Core portfolio now without an explicit allocation to value, the investment flexibility of the Betterment platform has improved. VI. The Innovative Technology portfolio strategy In 2021, Betterment launched the Innovative Technology portfolio to provide access to the thematic trend of technological innovation. The portfolio’s investment premise is based upon the thesis that, over the long term, the companies innovating and disrupting their respective industries are shaping our global economy and may be the winners of the next industrial revolution. Some of these themes the portfolio seeks to provide increased exposure to are: Artificial intelligence Alternative finance Clean energy Manufacturing Biotechnology Similar to the Value Tilt portfolio, the Core portfolio is used as the foundation of construction for the Innovative Technology portfolio. With this portfolio strategy, we calibrated the size of the innovative technology funds’ exposure based on a certain target historical tracking error to the backtested performance of the latest version of the Core portfolio. Through this process, the Innovative Technology portfolio maintains the same globally-diversified, low-cost approach that is found in Betterment’s investment philosophy. The portfolio, however, has increased exposure to risk given that innovation requires a long-term view, and may face uncertainties along the way. It may outperform or underperform depending on the return experience of the innovative technology funds’ exposure and the thematic landscape. To learn more, read the Innovative Technology portfolio disclosure. VII. The Socially Responsible Investing portfolio strategies Betterment introduced its first Socially Responsible Investing (SRI) portfolio in 2017 and has since expanded the options to include three distinct portfolios: Broad Impact, Social Impact, and Climate Impact. These SRI portfolios are built on the same foundational principles as the Core portfolio, utilizing various asset classes to create globally-diversified portfolios. However, they incorporate socially-responsible ETFs that align with specific Environmental, Social, and Governance (ESG) and shareholder engagement mandates, tailored to each SRI focus. Betterment’s SRI approach emphasizes three core dimensions: Reducing exposure to companies engaged in unsustainable activities Increasing investments in those addressing environmental and social challenges Allocating to funds that utilize shareholder engagement to promote responsible corporate behavior. This methodology ensures diversified, cost-efficient portfolios that resonate with investors' values. For more information, read our full Socially Responsible Investing portfolios methodology. VIII. Conclusion After setting the strategic weight of assets in our various Betterment portfolios, the next step in implementing the portfolio construction process is our fund selection methodology, which selects the appropriate ETFs for the respective asset exposure in a generally low-cost, tax-efficient way. In keeping with our philosophy, that process, like our portfolio construction process, is executed in a systematic, rules-based way, taking into account the cost of the fund and the liquidity of the fund. Beyond ticker selection is our established process for allocation management—how we advise downgrading risk over time. The level of granularity in allocation management provides the flexibility to align to multiple goals with different timelines and circumstances. Most of our portfolios contain 101 individualized risk levels (each with a different percentage of the portfolio invested in stocks vs. bonds, informed by your financial goals, time horizon and risk tolerance). Finally, our overlay features of automated rebalancing, tax-loss harvesting, and our methodology for automatic asset location, which we call Tax Coordination, are designed to be used to help further maximize individualized, after-tax returns. Together these processes put our principles into action, to help each and every Betterment customer maximize value while invested at Betterment and when they take their money home. IX. Citations 1 Markowitz, H., "Portfolio Selection".The Journal of Finance, Vol. 7, No. 1. (Mar., 1952), pp. 77-91. 2 Black F. and Litterman R., Asset Allocation Combining Investor Views with Market Equilibrium, Journal of Fixed Income, Vol. 1, No. 2. (Sep., 1991), pp. 7-18. Black F. and Litterman R., Global Portfolio Optimization, Financial Analysts Journal, Vol. 48, No. 5 (Sep. - Oct., 1992), pp. 28-43. 3 Litterman, B. (2004) Modern Investment Management: An Equilibrium Approach. 4 Note that the risk aversion parameter is essentially a free parameter. 5 Ilmnen, A., Expected Returns.
