It is hard to imagine an industry that has not been impacted by technology, and ‘investment management’ is no exception. For years, average retail investor with net worth less than $1 million was outside the scope of a traditional investment advisor’s “potential clients” list, but today the same investor is the target market for Robo-advisors. Robo Advisors tends to take the human element out of investment advisory and enables an individual to make financial decisions by providing low cost, highly diversified portfolio solutions.
Wealthfront, one such robo-advisor, is at the front of this tech revolution in investment management. It’s a start-up in Silicon Valley that is now managing more than $3B in AUM – a progress made well under 4 years. By offering, low cost, well diversified, high net-of-fee after-tax returns to retail investors who would not have considered financial advisors because of cost, it is helping shape investment management for the better.
Before reviewing Wealthfront’s portfolio construction and how it selects investments, let’s first take a look into its investment ideology. There is no dearth of investment securities to invest in. Every now and then there are new vehicles introduced to the market. However, Wealthfront holds the view that the best investment strategy is to invest in a globally diversified portfolio that invests in low cost indexed ETFs, which are index funds closely tracking a broader market, index, or basket of assets. Burton Malkiel, the firm’s Chief Investment Officer and the author of the famous book “Random Walk Down Wall Street,” can be credited with much of the investment views the firm holds. He believes in a buy-and-hold strategy for the long term. In fact, he asserts that investing in passive, indexed ETFs for the long term will provide greater returns than almost every other investment strategy. There are funds like “Smart Beta Funds” that are touted as providing higher returns via a collection of passive investment instruments that have the same risk as a low cost index fund invested in the total stock market.
Smart Beta Funds or any other portfolio strategies that use factor tilts instead of capitalization weighted funds to achieve higher returns are, in Malkiel’s view, not smart investments. Malkiel is a strong proponent of investing in cap weighted funds using low cost indexed funds. Regarding Smart Beta Funds, he says,
“The holder of a broad-based index fund will by definition achieve the average. In the presence of cost, investing has to be a negative-sum game. The broad market portfolio can be accessed through ETFs at very little cost. “Smart Beta” funds and ETFs charge annual fees that can be a very large multiple of those charged on standard capitalized-weighted index funds.” (Source: Wealthfront, “Smart Beta”)
Malkiel’s views about investing in low cost index funds forms the foundation of the investment methodology at Wealthfront and has helped the company provide efficient and cost-effective investment products to its clients.
Modern Portfolio Theory: Investment Methodology at Wealthfront
The investment methodology that Wealthfront employs is based on Modern Portfolio Theory which holds that for a given level of risk, there is a portfolio on the “efficient frontier” that provides the highest return. In order to provide its clients with such return maximizing, risk minimizing portfolios, Wealthfront uses a mix of 11 asset classes to provide maximum diversification benefits. Figure 1 shows each asset classes’ standard deviation (i.e. risk) and expected real return.
|Asset Class||Standard Deviation(annualized)||CAPM Return||Wealthfront View||Black Litterman Gross Return||Net-of-Fee Return||After-Tax Net-of-Fee Return|
|U.S. Government Bonds||5%||-0.8%||-1.5%||-0.9%||-1.0%||-1.4%|
|Emerging Markets Bonds||7%||1.0%||2.0%||1.1%||0.8%||-0.5%|
Figure 1: Asset Class Expected ‘Real’ Return (assuming long term inflation at 2%) (Source: Wealthfront)
Using the above estimates, Wealthfront constructs portfolios for investors based on their overall levels of risk tolerance. If the above estimates for risk and return change, portfolios are annually rebalanced to reflect the new estimates. The advantages derived from diversification are well known under the Modern Portfolio Theory but how does Wealthfront implement diversification?
Wealthfront’s Portfolio Construction
Modern Portfolio Theory espouses diversification as key to reducing overall portfolio risk while maximizing return. Therefore, diversification is significant to providing optimal portfolio solutions. The next section looks into how Wealthfront brings about diversification in its product solutions.
- Asset Class Diversification
Advisors have traditionally diversified client portfolios across the following three asset classes.
- US. Stocks
- Foreign developed stocks
- US. Government Bonds
However, Modern Portfolio Theory suggests that the more there is diversification across uncorrelated asset classes, the more portfolio variance can be brought down to maximize return. Wealthfront achieves this purpose by using seven or eight asset classes on average in its portfolios. As a result, the Efficient Frontier for retirement accounts is raised by around 0.6% per year and for taxable accounts by 0.6% per year (Source: Wealthfront) when more asset classes are added to the traditional three asset class retirement and taxable accounts respectively.
For its portfolios, Wealthfront has set minimum and maximum asset allocation limits. Using this and tax efficiency of each asset class as a yardstick, it designs portfolios for its clients given their levels of risk tolerance.
|Asset Class||Minimum Allocation||Maximum Allocation||Tax Efficient?|
|Foreign Developed Stocks||5%||35%||Efficient|
|Emerging Market Stocks||5%||35%||Efficient|
|Dividend Growth Stocks||5%||35%||Efficient|
|U.S. Government Bonds||5%||35%||Inefficient|
|Emerging Market Bonds||5%||35%||Inefficient|
Table 1: Asset Class Allocation Limits (Source: Wealthfront)
- How does Wealthfront Measure Risk Tolerance?
Now that asset class diversification has been shown to improve expected reward for a portfolio with a given risk level on the Efficient Frontier, the next logical question is: what defines risk for an individual? Conventional ways of doing this by advisors included the use of questionnaires to arrive at a ‘subjective’ risk tolerance level for a client.
Wealthfront uses a two step approach to assessing true risk tolerance of a client
- It combines behavioral economics with sophisticated algorithms to fine tune nuances in an investor’s risk preferences to get a ‘subjective’ risk tolerance. For instance, a client is comfortable taking on risk in a certain investment scenario but does not want any risk in another scenario. The variation in risk preferences is seen and interpreted as an inconsistency in risk preference. Therefore, the client gets assigned a lower score for risk tolerance than what the weighted average of her answers to the questionnaire would suggest.
- Secondly, there is ‘objective’ risk tolerance measure, where a client is asked a few questions to estimate whether there will be sufficient income in retirement to support the spending needs. The more the expected retirement savings to support spending needs, the more an individual will be risk tolerant.
Combining results from the above two risk assessments – giving more weighting to the more risk averse factor – an overall risk metric is derived.
After assigning an individual a score for risk tolerance, Wealthfront seeks to maximize the following utility function, popularized by Harry Markowitz.
To explain how Wealthfront puts it into practice, consider an individual with $100,000 invested and a risk tolerance of 7 (Wealthfront assigns risk tolerance on a scale of 0 to 10, with 0 being the lowest and 10 being the highest; 7 is risk tolerance for an average Wealthfront client). Figures 4 and 5 show how asset allocations will be done for such an individual.
Once a portfolio is created using Mean-Variance Optimization (MVO) technique (that solves for the portfolios on the Efficient Frontier), it is monitored and rebalanced periodically to maintain the asset allocations at optimal levels. Due to market movements, asset allocations in a portfolio tend to deviate from their intended levels. Thus it becomes necessary to bring the asset classes back in line with their original allocations to achieve the highest expected return for a given level of risk.
Thus far we analyzed how Wealthfront constructs an optimal portfolio for its clients given their appetites for risk. But that is just one part to delivering high expected returns to its clients. The other part is the investment vehicles themselves that are used to construct a portfolio.
- How does Wealthfront chooses investment vehicles for a portfolio?
Wealthfront analyzes investment vehicles across four aspects.
It is not possible to maximize returns without minimizing fees and costs. Wealthfront believes this can only be done by picking index based ETFs with the lowest cost and tracking error, maximum liquidity, and without the need for underlying securities lending. ETF issuers often involve in lending of securities underlying their respective ETFs to hedge funds. This can expose an ETF buyer to unnecessary risk. To avoid this risk, Wealthfront, therefore, selects ETFs that either minimize lending of securities or distribute lending revenue generated with ETF investors to lower management fees.
For an investor, it is important to note how much an advisor charges in advisory fee. Typically it is upwards of 1% of account the value. Wealthfront is able to bring that cost down for its clients due to the low cost product offerings.
|Up to $10,000||Free|
Figure 6: Advisory Fee Structure (Source: Wealthfront)
In addition to advisory fees, there are product fees. Product fees are costs embedded in the ETFs themselves as it is a fee charged by the issuers in order to operate their funds.
On a weighted average basis, a Wealthfront portfolio has 0.12% in product fees. Combined with advisory fees, an average Wealthfront customer ends up paying 0.35% to 0.40% of account value, which is still lower than 1% on average charged by traditional advisors for advisory fees alone.
As can be seen from the figure above, higher fees eat into the long term return of a portfolio. Wealthfront compares two portfolios: portfolio A with total expenses of 0.57% of an invested amount of $200,722 and portfolio B with 2.26% of an invested amount of $182,581. Compared over the course of 30 years to an average Wealthfront portfolio with overall expense of 0.37%, Portfolio A yields $25,912 less in value while Portfolio B yields $291,188 less.
- Tax Efficiency
Wealthfront employs daily tax loss harvesting to maximize portfolio returns. While minimizing fees is an important element for return maximization, minimizing taxes have the ability to generate far more savings.
It also offers tax-optimized direct Indexing for accounts with more than $100,000 in investment. This method of indexing allows Wealthfront to generate tax savings in excess of what daily tax loss harvesting produces. As a result, an account with tax-optimized direct indexing reaps further 0.2% to 0.5% in annual after-tax returns beyond the returns generated through daily tax los harvesting.
- Cash Drag
Wealthfront proposes to invest any excess cash over the long term to increase portfolio return. It defines excess cash as cash sitting in a portfolio above what is needed as an emergency fund. To arrive at an estimate for an appropriate emergency fund, it assumes annual expenses to be 80% of after-tax income and the emergency fund to cover 4.5 months of expenses. This results in an emergency fund worth 30% of annual after-tax income. Any cash, if invested, above what an emergency fund requires can generate additional return for the portfolio.
Portfolio B does not have any excess cash. Therefore, cash drag cost for Portfolio B is $0. However, Portfolio A does have $7,571 in excess cash, which if invested over 30 years at an average annual rate of 4.75% (4.68% for retirement accounts) net of fee could generate $22,892 in additional income.
When it comes to diversification, most portfolios tend to suffer from ‘home bias,’ which is the tendency to over allocate in one’s home market to improve expected returns. Research proves to the contrary.
The above graphs show that over the long term, investing in a well diversified ‘global’ portfolio is the best investment strategy. It is not possible to time the market and predict how a certain market will perform in a given time frame. By investing in indexed ETFs with lowest cost and tracking error and maximum liquidity will ensure that costs do not eat into the investment returns, the portfolio is well diversified and mimics the market in performance, and can be traded in and out of at any time without liquidity issues.