Wealthfront’s Smart Beta
Smart Beta is Wealthfront’s next evolution of Stock-level Tax-Loss Harvesting, and is available to all clients with taxable account balances over $500k. Like Stock-level Tax-Loss Harvesting, it replaces the ETF normally used to represent US stocks in a Wealthfront portfolio with a combination of large- and mid-capitalization U.S. stocks and one or two additional ETFs. However, rather than hold the individual securities in proportion to their market capitalization, the weights are shifted to increase the expected after-tax return of the portfolio.
Smart Beta takes advantage of decades of time-tested academic research on the drivers of asset returns. The underlying research — recognized with two Nobel Prizes (1990, 2013) — demonstrated that a security’s return is determined by the set of risk factors that it is exposed to, rather than its standalone risk (e.g. as measured by its volatility). Thus, by optimizing the composition of the factors to which a portfolio of stocks is exposed, its expected return can be increased while leaving volatility unchanged. Using a multi-factor methodology, Smart Beta identifies securities which are likely to have the highest expected returns, and overweights them relative to their allocation in the cap-weighted benchmark.
By combining this portfolio construction methodology with our Stock-level Tax-Loss Harvesting stock-level tax loss harvesting algorithm, we expect the incremental gains to be delivered in a tax efficient manner. Specifically, employing a non-cap-weighted allocation to individual securities requires periodic trading, such that some of the incremental gains could be subject to short-term tax rates. However, the availability of tax losses harvested by our Stock-level Tax-Loss Harvesting algorithm enables the taxation of these gains to be offset, thus deferring their taxation to a future date (i.e. when the portfolio is liquidated) when a potentially lower, long-term capital gains rate will apply.
We deliver these features to eligible clients at no incremental cost to our basic service.
Relation to Stock-level Tax-Loss Harvesting
In 2013 Wealthfront launched Stock-level Tax-Loss Harvesting, a stock level tax-loss harvesting service. Stock-level Tax-Loss Harvesting offered two significant advantages over our basic, ETF-level tax loss harvesting service, which is available to all clients. First, by replacing the ETF used to gain exposure to U.S. equity markets with individual securities, it lowered the portfolio’s expense ratio. Second, trading in individual securities enabled Stock-level Tax-Loss Harvesting to harvest more losses than the ETF-level harvesting service which only relies on a pair of ETFs (Vanguard’s Total Stock Market Index, VTI, and Schwab’s U.S. Broad Market ETF, SCHB).
Unlike Smart Beta, the individual stock component of the Stock-level Tax-Loss Harvesting portfolio tracks a market capitalization-weighted benchmark. In other words, it seeks to deliver a return matching that of a broad U.S. equity index, while maximizing the quantity of harvested losses. Smart Beta balances two objectives: (a) delivering a superior return relative to the cap-weighted benchmark by relying on a multi-factor portfolio construction methodology; and, (b) maximizing the quantity of harvested tax losses. Both Smart Beta and Stock-level Tax-Loss Harvesting, continue to hold the other ETFs common to Wealthfront investments (such as Foreign Stocks, Emerging Markets, Dividend Stocks, etc.). All Smart Beta accounts also benefit from our Daily Tax-Loss Harvesting service – which harvests tax-savings at the ETF-level — outside the U.S. stock component of the portfolio.
Costs and Minimums
We offer two levels of Smart Beta:
- Smart Beta 500: For taxable accounts between $500,000 and $1 million, we replace the ETF normally used to represent a broad market of US stocks (Vanguard’s Total Stock Market ETF, VTI) with up to 500 of the largest capitalization US stocks that comprise the S&P 500 index and the Vanguard Extended Market ETF (VXF) used to represent the remaining smaller capitalization companies.
- Smart Beta 1000: For taxable accounts above $1 million, we replace our Vanguard Total Stock Market ETF with up to 1,000 of the largest capitalization US stocks (typically representing US large cap and mid cap companies) that comprise the S&P 1500 index and the Vanguard Small-Cap ETF (VB) used to represent the remaining small-capitalization US companies.
By replacing VTI with a combination of individual securities, you no longer pay the VTI expense ratio. This means the combined portfolio of individual securities and a completion ETF has a lower combined expense ratio.
More than 80% of the Smart Beta position is comprised of individual stocks and thus has no expense ratio. The remaining 20% or so is comprised of either the Vanguard Extended Market ETF (VXF) for $500,000 accounts or the Vanguard Small-Cap ETF (VB) for accounts over $1 million. The combined portfolio remains subject to Wealthfront’s advisory fee
Our minimums for Smart Beta are based on dollar amounts required to hold a reasonable collection of individual US stocks while continuing to track the performance of the broad US market. We found that holding hundreds of stocks requires a US equity allocation of roughly $150,000 – thus necessitating an account minimum of about $500,000 (because approximately 30% of our portfolios are allocated to US Equities). This minimum increases again as you exceed 500 stocks and start holding some of the smaller mid-cap stocks in the position – thus necessitating a $1 million minimum for the higher level of Smart Beta.
Investment Returns and Factors
The expected return on an asset can be decomposed into compensation for the passage of time, and compensation for exposure to common, non-diversifiable sources of risk (i.e. “factors”). The compensation for passage of time is captured by the risk-free rate, which measures the opportunity cost of parting with your money, and can be proxied by rates of return on U.S. Treasuries. The compensation for bearing risk is captured by the product of a security’s exposure to a factor (measured by its beta) and the premium for bearing that risk. “Active” managers seek to identify deviations from this relationship, but decades of research now show that they fail to outperform their properly risk-matched benchmarks.
The earliest model of security returns was the Capital Asset Pricing Model (CAPM), proposed independently by Treynor (1961), Sharpe (1964), Mossin (1965), and Lintner (1966), and recognized with the Nobel Prize in 1990. The CAPM is known as a single-factor model, where the only source of compensated risk (the “factor”) is a security’s exposure to the broad equity market. In other words, the single predictor of a security’s future expected return is how much or little it comoves with the aggregate market.
Over time researchers identified several deficiencies with the CAPM. For example, Banz (1980) showed that small firms — as measured by their market capitalization — have a tendency to outperform large firms, and this return differential cannot be explained by differences in their market betas. A decade later, a sequence of influential papers by Professors Eugene Fama and Kenneth French, documented an even more significant shortcoming of the CAPM, the “value effect.” Specifically, cheap stocks (firms with a low ratio of market equity to book equity) on average outperformed expensive stocks (firms with a high ratio of market equity to book equity), even though their market betas were lower. These findings led them to extend the single-factor CAPM model to include a value factor (HML, high-minus-low) and a size factor (SMB, small-minus-big). The Fama-French three-factor model (1992, 1993) was an early multi-factor model, and has become an important benchmark in academic work. Around the same time, Jegadeesh and Titman (1993) identified a “momentum effect,” or the tendency for securities that have been the best (worst) performers over the last six to twelve months to continue to be best (worst) performers. This effect could not be rationalized by Fama and French’s three-factor model, leading researchers to include momentum as an additional, fourth factor. Professor Kenneth French tabulates the historical performance of these factors on his website.
Over the ensuing few decades, academia and industry alike embarked on a search for additional characteristics that can be used to sort stocks in a manner that produces a return spread that cannot be accounted for by exposure to the market, size, value, and momentum factors. Out of the hundreds of characteristics that have been considered, some findings disappeared following publication and others were shown to be concentrated in areas of capital markets where they cannot be reliably incorporated in portfolios at scale (e.g. among stocks with small capitalizations or where it is difficult to short). However, a small subset has proven to be robust across time, across geographies (i.e. work reliably outside of the United States), and — in some instances — across asset classes. These include the tendency for high dividend yield, low market beta, and low volatility securities to deliver superior returns to low dividend yield, high market beta, and high volatility securities, respectively.
Smart Beta blends five single-factor strategies (value, momentum, high dividend yield, low market beta, and low volatility) with the cap-weighted market index to generate a modified index. This index then serves as the benchmark for our stock-level tax loss harvesting algorithm (Stock-level Tax-Loss Harvesting), which seeks to maximize the quantity of harvested losses, while minimizing the tracking error from the supplied benchmark.
Our portfolio construction procedure begins by constructing single-factor strategies. Each of these strategies ranks securities based on a single characteristic, and then invests in a subset of securities based on this characteristic (e.g. value, momentum, etc.). The resulting portfolio can be interpreted as a simple single-factor strategy and its composition is periodically refreshed. Tables 1a and 1b display the historical returns for each of the five single-factor strategies over the period from 1964 to 2016, matching the span of data available from CRSP (Center for Research in Security Prices). Each strategy is separately implemented in the 500 or 1000 largest securities in the U.S. equity universe (labeled CRSP 500 and CRSP 1000, respectively), which will be relevant to the two different tiers of Smart Beta. Strategy returns are presented gross of fees.
For each strategy, Tables 1a and 1b report the annualized (arithmetic) mean return, volatility, and Sharpe Ratio (computed relative to the one-month T-Bill return). To provide additional insight into the relative performance of each strategy and the cap-weighted index in adverse circumstances, we also report a mean shortfall metric. This metric reports the mean annualized performance differential in three year periods in which the single-factor strategy underperformed the cap-weighted index. The data demonstrate that over the long run each of the factors individually has been able to deliver incremental returns over and above the passive, cap-weighted benchmark. However, there are three-year windows in which the single-factor strategies have significantly underperformed the cap-weighted benchmark.
The challenge with relying on single-factor strategies is that they can underperform the cap-weighted index by a significant amount. For example, investing in low volatility stocks selected from among the CRSP 500 delivered 3.74% less per year relative to the cap-weighted index in the three-year periods with underperformance. Despite this shortcoming, 95% of existing Smart Beta ETFs are designed to track a single factor (Morningstar, 2016).
Importantly, there are substantial benefits to be gained from factor diversification. To illustrate this, Tables 2a and 2b report the correlation matrix of the return differentials of each factor relative to the cap-weighted index. The factors exhibit low (and sometimes negative) correlations, such that different factors are likely to contribute incremental returns at different points in time. A second benefit of the low factor correlations is that ranking securities on different characteristics will generally point to different securities, resulting in a well-diversified portfolio once the single-factor strategies are combined. Smart Beta exploits this feature by combining the five single-factor strategies.
Constructing the Modified Index
After constructing the single-factor strategies we combine them to produce a multi-factor “overlay” portfolio, which includes securities with good value, strong momentum, high dividend yield, low market beta, and low volatility. This overlay portfolio is blended with the cap-weighted index to produce a modified index, which serves as the benchmark relative to which the stock-level tax loss harvesting algorithm will seek to minimize tracking error. The goal of this construction is to overweight securities with high expected returns, while ensuring the modified index remains close to the cap-weighted benchmark, thus keeping overall portfolio risk unchanged.
Table 3 follows the format of Tables 1a and 1b, but compares the performance of the cap-weighted benchmark and the modified index implemented within the CRSP 500 and CRSP 1000 U.S. equity universes. As before, the underlying data span the period from 1964 to 2016. Over this period the modified index has outperformed the cap-weighted index by 0.82% per year before taxes, when implemented using the 500 largest securities by market capitalization, and 0.98% per year, when implemented using the 1000 largest securities. These results were accomplished while maintaining portfolio volatilities that were no greater than the volatilities of the cap-weighted indexes. As a result, the Sharpe Ratios of the modified indexes are 20% greater, indicating an improvement in the risk-adjusted return from deploying the multi-factor overlay.
An additional benefit of combining the single-factor strategies is that the mean shortfall of the modified index is considerably smaller than for any individual, single-factor strategy. Recall, from Tables 1a and 1b, the mean shortfall captures the mean annual amount by which the modified index underperformed the cap-weighted index in the three-year periods with underperformance. These values are now -0.62% per year, or 5-10x smaller than observed among the single-factor strategies individually.
To provide more detail on their relative performance, Figure 1 below displays the annualized return differential between the modified and capitalization-weighted indexes implemented in the CRSP 500 universe over three-year rolling windows. Positive (negative) values indicate that the modified index has outperformed (underperformed) the cap-weighted benchmark. Although the modified index outperformed in over three quarters of the three-year periods, it is important to note that there have been three-year periods in which the annual underperformance equaled almost 2% per year. The return differential of the two indices implemented in the CRSP 1000 universe is similar, and is omitted for brevity.
Combining the Modified Index with Stock-Level Tax Loss Harvesting
Smart Beta combines the multi-factor overlay described in the previous section with Wealthfront’s stock-level tax loss harvesting service (Stock-level Tax-Loss Harvesting) to deliver the portfolio gains tax efficiently. Since the overlay requires periodic rebalancing, a portion of the gains that it generates will be from securities held less than one year, and therefore “short-term” in nature for tax purposes. These gains would typically be subject to the relatively higher tax rates applicable to short-term capital gains. However, by offsetting these gains with harvested tax losses it is possible to defer their taxation into the distant future (even indefinitely in some circumstances). We implement this by using the modified index as the benchmark for the Stock-level Tax-Loss Harvesting algorithm, which then seeks to maximize the quantity of harvested losses, while minimizing the tracking error from the modified index. We refer the reader to the Stock-level Tax-Loss Harvesting whitepaper for additional details of the algorithm, the underlying optimization objective, as well as, assumptions used to derive Tax Alpha estimates.
To evaluate the economic benefit of Smart Beta we carry out a backtest using data from 2000 to 2016. The date range is determined by the availability of S&P index constituent data, used as the baseline cap-weighted indices. The backtest reflects a retroactive application of the proposed portfolio construction methodology. It does not reflect the realized performance of any client, and does not represent a guarantee that the future benefit of the strategy is likely to be similar.
Figure 2 compares the cumulative performance of Smart Beta 500 (AI500) and Smart Beta 1000 (AI1000) to the performance of the US Stock segment of the Wealthfront diversified portfolio, represented by Vanguard’s Total Stock Market Index (VTI). All strategy returns are displayed net of Wealthfront’s advisory fee.
Benefits of Smart Beta
We measure the effectiveness of Smart Beta on three dimensions: how much incremental return it generates from the multi-factor overlay (Tracking Difference), how much potential tax benefit it generates from stock-level tax-loss harvesting (Tax Alpha), and how significantly it deviates from the original ETF it replaced to achieve the first two objectives (Tracking Error). We also compute the annual trading turnover of the strategy (Turnover) to evaluate the impact of transaction costs. Portfolio turnover is computed as the gross dollar value of the trades (sales and purchases) divided by the value of the portfolio. The assumptions used to compute Tax Alpha are reported in Wealthfront’s Stock-level Tax-Loss Harvesting whitepaper.
We use the term Tracking Difference to describe the difference between the portfolio’s return and the original ETF’s return in a given time period and Tracking Error for the standard deviation of the tracking differences. Tax Alpha is computed assuming the client has sufficient external gains to take advantage of all harvested losses, unless stated otherwise. The client is assumed to be subject to a combined federal and state short-term and long-term capital gain rates of 44.5% and 24.5% respectively. The derivation of these assumptions is contained in the Stock-level Tax-Loss Harvesting whitepaper. Unlike what is described in the Stock-level Tax-Loss Harvesting whitepaper, in this case we conservatively assume clients make an initial deposit necessary to meet the minimum account size of the desired level of Smart Beta and make no additional deposits.
Table 4 summarizes these metrics for Smart Beta 500 (AI500) and 1000 (AI1000) over the period 2000-2016. As reflected by the Tracking Difference, the overlay contributed an incremental 1.61% per year in return relative to the cap-weighted index (proxied by Vanguard’s Total Stock Market Index, VTI) for AI500; and 1.55% per year for the AI1000 strategy. Even after utilizing some of the harvested losses to offset gains generated by the overlay, the stock-level tax loss harvesting delivers 1.50% per year in Tax Alpha. The Tracking Error reflects the combined effects of deviations induced by the multi-factor overlay and tax loss harvesting, and ranges between 3.44% and 3.56%. The annualized portfolio turnover is approximately 75%.
Table 5 presents an estimate of the aggregate after-tax benefit from Smart Beta. The After-tax Benefit is computed as the sum of the Tracking Difference and the Tax Alpha less an estimate of the trading costs incurred implementing the strategy. The trading cost penalty assumes that the roundtrip cost for large-cap U.S. equities used in the construction of the overlay during this time period was 0.20% (Fisher, et al. (2016)). We report the After-tax Benefit under two assumptions on the client’s ability to utilize the harvested losses to offset external capital gains. The first column assumes that the client is able to fully utilize the harvested losses, and therefore realizes the full Tax Alpha; the second column conservatively assumes that the client is able to utilize only 25% of the Tax Alpha. The After-Tax Benefit of Smart Beta ranges from over 1.85%, under the assumption that the client is able to utilize only one quarter of the harvested losses, to over 2.97% under full utilization of harvested losses.
In order to provide further insight into the backtested performance of Smart Beta, we display the Tracking Difference and Tax Alpha for each year in the sample. The Tracking Difference captures the benefit generated by tracking the non-cap-weighted index and is computed relative to the performance of VTI (the ETF used to track a broad market of US stocks in a Wealthfront portfolio). The Tax Alpha is the potential benefit an investor could realize, assuming they were able to utilize all of the harvested losses to offset capital gains realized outside the Wealthfront portfolio.
Figure 3 indicates that the multi-factor overlay contributed positively in 12 out of the 17 years for both variants of Smart Beta. The overlay generated particularly strong performance in the first year of the sample, and was driven by the momentum of Internet stocks. Following the bursting of the Internet Bubble, the value, low beta and low volatility factors contribute positively, while momentum contributed negatively, illustrating the diversification benefits of a multi-factor overlay construction.
Finally, Figure 4 displays the Tax Alpha delivered by the two levels of Smart Beta. Crucially, the Tax Alpha is computed net of any losses applied to offset taxable gains generated by the multi-factor overlay, and only reflects the Tax Alpha captured from the US Stock component of the portfolio. All Smart Beta accounts also benefit from our Daily Tax-Loss Harvesting service – which harvests tax-savings at the ETF-level — outside the U.S. stock component of the portfolio.
Smart Beta tends to generate positive Tax Alpha in periods with negative market performance, and negative Tax Alpha in periods when the market is advancing. In years with market gains, harvested losses tend to be consumed by the gains realized in rebalancing the overlay portfolio, and also due to turnover in the composition of the cap-weighted index. Importantly, harvested losses can be carried forward for use in subsequent years. Due to this, the average Tax Alpha generated by Smart Beta over the seventeen year period was strongly positive, as reported in Table 4.
Realized Results – 2019
Wealthfront’s Smart Beta launched in June 2017, giving us over two years of realized results to analyze. This section compares the actual performance of Wealthfront Smart Beta with similar mutual funds and ETFs from July 1, 2017 (launch of the service) to September 30, 2019, and explores the underlying factors that drive differences in performance. All returns shown are annualized over this period.
For simplicity we focus on Wealthfront Smart Beta 500, which trades the individual constituents of the S&P 500, and holds the VXF ETF to provide exposure to small-cap US stocks. To isolate the effect of the smart beta “tilts” – the overweights and underweights of individual stocks relative to their benchmark weight – we only compute performance of the portion of client portfolios holding individual US stocks. For each trading day, we calculate the return for each client enrolled in Wealthfront Smart Beta 500 and compute a value-weighted average. The number we display is a compounded and annualized total return over the comparison window.
As a benchmark, we display the total return of the S&P 500. We also display the return of a theoretical portfolio that trades exactly to the target Smart Beta portfolio each time the weights are updated. A client’s portfolio may differ from this theoretical portfolio due to tax-loss harvesting, tax considerations deterring sales of appreciated assets, or other trading-related constraints.
Table 6 shows these three returns. Comparing “Theoretical Smart Beta” and “S&P 500”, we see that the smart beta tilts failed to add value over this period – the return of the theoretical portfolio is slightly less than the S&P 500. The average realized return of Smart Beta clients is less still, by .6%, reflecting the extra constraints and competing tax-alpha objective in live portfolios. Note that the Smart Beta returns do not include Wealthfront’s 0.25% annual advisory fee, which includes management of a diversified global portfolio and our tax-loss harvesting service. Had this fee been deducted, the Smart Beta performance number would be 10.9%.
Next, we compare the returns of Wealthfront Smart Beta with several Smart Beta mutual funds and ETFs over the same period. All the funds we include in the comparison invest in diversified portfolios across all sectors of US stocks and, like Wealthfront Smart Beta, employ a “multi-factor” approach. The funds differ in the selection of factors used, as well as the exact universe of stocks included in the portfolios. Some salient characteristics of the funds used in the comparison are in Table 7. Besides the visible differences, there are a number of decisions in portfolio construction which can drive differences in returns – these include (potentially time-varying) factor weightings, choices of underlying descriptors to construct factor scores, trading frequency, and portfolio constraints . Note that we include two funds managed by Dimensional Fund Advisors – both funds invest in the same universe and use the same set of factors, but the US Core Equity 2 fund takes more tracking error than US Core Equity 1, with the intent of achieving higher returns above the benchmark.
Figure 5 shows the total return of each fund, in addition to the total returns of Wealthfront Smart Beta and two index ETFs: VTI, which tracks the CRSP US Total Market Index, and VOO, which tracks the S&P 500. We include both VTI and VOO as comparisons because the funds we selected to compare trade over different universes, and we don’t want to misrepresent a choice of universe as underperformance – for example, the S&P 500 outperformed the Russell 3000 over this time period, but we would not say that a fund designed to track the Russell 3000 underperformed because it returned less than the S&P 500. As mentioned above, the Wealthfront Smart Beta returns do not not deduct Wealthfront’s 0.25% advisory fee, which includes the management of assets in other asset classes, as well as our tax-loss harvesting service. If this fee was deducted, the performance of Wealthfront Smart Beta would be 10.9%. The returns of the other funds are net of the funds’ respective management fees, but also do not deduct any additional advisory fee. According to AdvisoryHQ , the average AUM-based advisory fee from 2017 to 2019 was 1.02% for a $1 million account.
Of the six smart-beta funds in our study, only GSLC has outperformed the VOO and VTI index funds over the comparison period. The others underperformed to varying degrees, with QCELX and LRGF lagging VOO by four percentage points annually.
We stress that two years is a very short period to compare realized performance – our intent is to show that the performance of Wealthfront’s Smart Beta portfolios relative to index funds during this period was not abnormal. We do, however, point out that Wealthfront’s Smart Beta is available at no extra cost beyond the Wealthfront advisory fee, unlike other funds in the study, which all have higher expense ratios than both VOO and VTI and don’t offer tax-loss harvesting which can add significant value.
All of the returns in Figure 5 are pre-tax. In our research, we find that Wealthfront’s Stock-level Tax-Loss Harvesting, which is automatically included at no extra fee in smart-beta portfolios, in some instances may add over 2% in average annual after-tax performance. On an after-tax basis, Wealthfront Smart Beta potentially may have relatively better performance than an index or a fund that does not include tax-loss harvesting.
Finally, we examine the performance of the individual factors used in Wealthfront Smart Beta, to indicate which factors contributed to or detracted from the performance of Smart Beta. Consistent with other analysis in this section, we calculate daily returns from July 1, 2017 to September 30, 2019.
As described in previous sections, each factor portfolio is an equally-weighted basket of stocks in the S&P 500. The baskets are defined by ranking the stocks in the S&P 500 by a certain characteristic, and choosing subsets of the highest-ranked stocks. The baskets are refreshed approximately monthly as we obtain new market data.
Figure 6 shows the total return of each factor portfolio over the comparison period, along with an equally-weighted average and the total return of the S&P 500. We observe that the momentum, low-volatility, and low-beta factors all outperformed the S&P 500, while dividend yield slightly underperformed and value massively underperformed, trailing the S&P 500 by more than ten percentage points annually. On average, the performance of the factor portfolios is slightly lower than that of the S&P, which we saw reflected in the performance of client portfolios. We note that these performance numbers are hypothetical – they do not represent the performance of any actual portfolio, and do not incorporate the costs of trading or other constraints placed on Wealthfront Smart Beta accounts. Their purpose is only to help explain the performance of Wealthfront Smart Beta at a more granular level.
The individual factor returns can help explain the performance of the smart-beta mutual funds and ETFs we considered as well. As an example, to try to explain the outperformance of GSLC seen in Figure 5, we regressed the fund’s daily “active” returns (in excess of the S&P 500) onto the active returns of each of the factor portfolios. This showed that the active returns of GSLC load negatively on the returns of the value factor’s active positions, explaining over 60% of GSLC’s outperformance over the S&P 500 in this time period. Again, this negative regression coefficient could be due to any number of choices in portfolio construction.
We see results consistent with our factor performance in the returns of “single-factor” funds from other managers. iShares, Fidelity, and Invesco each offer a suite of ETFs comprised of funds designed to capture one single factor. Not all managers offer the same set of factors, but all three offer a value-tilted fund. Table 8 shows the annualized returns of each manager’s offerings, along with the returns of Wealthfront’s factor portfolios over the same time period. The returns of other managers’ funds are net of management fees, while the Wealthfront returns do not include a fee – as noted above, they are purely hypothetical.
Within each family of funds, we see the same pattern of value significantly lagging the other factors. And, as with the factor portfolios, Fidelity and Invesco’s Low-Volatility funds are the best performers in their respective groups. This highlights again the value of diversification from a multi-factor approach.
Although it is not difficult to identify the value factor as the one dragging down the performance of Smart Beta, a deeper explanation of why value is underperforming is much more difficult. The value factor portfolio can be thought of as the sum of two portfolios: one holding members of the S&P 500 at their index weights, and a portfolio of “tilts”: small positive values on the stocks with more attractive value characteristics, and small negative values on the stocks with less attractive characteristics. Because the positions in the “tilt” portfolio are so small and numerous, it is nearly impossible to find a simple explanation for the performance of the overall portfolio.
Wealthfront Smart Beta provides clients with taxable accounts in excess of $500k access to a tax-efficient multi-factor investing strategy at no incremental cost. Unlike single-factor Smart Beta ETFs available from other managers Wealthfront Smart Beta relies on multiple factors when selecting which securities to overweight. This ensures the portfolio remains highly diversified, and minimizes the chances of underperforming the cap-weighted index. Moreover, by combining this approach with our stock-level tax loss harvesting service (Stock-level Tax-Loss Harvesting), clients can defer the taxation of the incremental gains relative to the cap-weighted index until portfolio liquidation. All this, at no incremental cost to our standard advisory fee.
Over the period where we have actual results, the value factor has negatively impacted Smart Beta, and as a result Smart Beta accounts have, on average, failed to outperform an index fund on a pre-tax basis. This is not unique to Wealthfront – five of the six multi-factor funds we considered underperformed index funds over the same period. Despite the disappointing recent performance of value, we continue to trust the decades of empirical evidence and believe that a diversified multi-factor approach, delivered at a reasonable price, can add value over an index in the long term.
 Intuitively, even though an asset class as a whole may be appreciating, thus offering few opportunities for harvesting at the asset class level, there will almost always be some securities within the asset classes that have declined in value, and can be used to harvest losses. As a result, Stock-level Tax-Loss Harvesting is able to magnify the quantity of harvested losses, delivering more potential Tax Alpha to clients, while lowering the expense ratio of the portfolio.
Wealthfront Smart Beta 500 is a strategy available to Wealthfront clients whose account balances meets a qualifying minimum value. Wealthfront Smart Beta 500 is not available as a mutual fund or ETF.
As an example, the annualized returns of the S&P 500, Russell 1000, and Russell 3000 – three cap-weighted indices comprised of US Stocks – over our analysis period were 12.2%, 12.0%, and 11.5%, respectively.
https://funds.aqr.com/our-funds/equity-funds/multi-style-funds/large-cap-multi-style-fund, https://www.ishares.com/us/products/272824/ishares-msci-usa-multi-factor-etf, https://www.gsam.com/content/gsam/us/en/advisors/fund-center/etf-fund-finder/goldman-sachs-activebeta-u-s–large-cap-equity-etf.html, https://am.jpmorgan.com/blob-gim/1383303740431/83456/FS-JPUS.PDF?segment=AMERICAS_US_ADV&locale=en_US
Nothing in this whitepaper should be construed as tax advice, a solicitation or offer, or recommendation, to buy or sell any security. Financial advisory services are only provided to investors who become Wealthfront Inc. clients pursuant to a written agreement available at www.wealthfront.com, which investors are urged to read carefully. All securities involve risk and may result in losses. While the data Wealthfront uses from third parties is believed to be reliable, Wealthfront does not guarantee the accuracy of the information. For more information please visit www.wealthfront.com or see our Full Disclosure.
The client was assumed to live in California and to be subject to a combined federal and state short-term and long-term capital gain rates of 44.5% and 24.5% respectively. The charts displaying performance and Tax Alpha from Smart Beta are based on backtesting, and assume initial investments of $500,000 (AI500) and $1,000,000 (AI1000), and no additional deposits. Backtested results are hypothetical and not an indicator of any investor’s actual current or future experience and are provided for illustrative purposes only. Hypothetical performance is developed by the retroactive application of a model designed with the benefit of hindsight and has inherent limitations. Specifically, hypothetical results do not reflect actual trading or the effect of material economic and market factors on the decision-making process. Wealthfront assumed we would have been able to purchase the securities recommended by the model and the markets were sufficiently liquid to permit all trading. Material economic and market factors might have occurred during the time period and could have had an impact on decision-making. Investors evaluating this information should carefully consider the processes, data, and assumptions used by Wealthfront in creating its simulations which are described above in the white paper. Backtested results were adjusted to reflect the reinvestment of dividends and other earnings and are presented net of advisory fees.
Wealthfront’s investment strategies, including portfolio rebalancing and tax loss harvesting, can lead to high levels of trading. High levels of trading could result in (a) bid-ask spread expense; (b) trade executions that may occur at prices beyond the bid ask spread (if quantity demanded exceeds quantity available at the bid or ask); (c) trading that may adversely move prices, such that subsequent transactions occur at worse prices; (d) trading that may disqualify some dividends from qualified dividend treatment; (e) unfulfilled orders or portfolio drift, in the event that markets are disorderly or trading halts altogether; and (f) unforeseen trading errors.
Wealthfront does not represent in any manner that the outcomes described herein will result in any particular tax consequence. Prospective investors should confer with their personal tax advisors regarding the tax consequences based on their particular circumstances. The effectiveness of the tax-loss harvesting strategy to reduce the tax liability of the client will depend on the client’s entire tax and investment profile, including purchases and dispositions in a client’s (or client’s spouse’s) accounts outside of Wealthfront and type of investments (e.g., taxable or nontaxable) or holding period (e.g., short- term or long-term). The performance of the new securities purchased through the tax-loss harvesting service may be better or worse than the performance of the securities that are sold for tax-loss harvesting purposes.
There is a chance that Wealthfront trading attributed to tax loss harvesting may create capital gains and wash sales and could be subject to higher transaction costs and market impacts. In addition, tax loss harvesting strategies may produce losses, which may not be offset by sufficient gains in the account and may be limited to a $3,000 deduction against income. The utilization of losses harvested through the strategy will depend upon the recognition of capital gains in the same or a future tax period, and in addition may be subject to limitations under applicable tax laws, e.g., if there are insufficient realized gains in the tax period, the use of harvested losses may be limited to a $3,000 deduction against ordinary income and distributions.
Wealthfront only monitors for tax-loss harvesting for accounts within Wealthfront. The client is responsible for monitoring their and their spouse’s accounts outside of Wealthfront for “wash sales.” A client may request spousal monitoring online or by calling Wealthfront at (844) 995-8437. If Wealthfront is monitoring multiple accounts to avoid the wash sale disallowance rule, the first taxable account to trade a security will block the other account(s) from trading in that same security for 30 days. A wash sale is the sale at a loss and purchase of the same security or substantially similar security within 30 days of each other. If a wash sale transaction occurs, the IRS may disallow or defer the loss for current tax reporting purposes.