investing

Variable Dollar Cost Averaging (VDCA)

Investing Psychology Matters

Fear and greed drive us to make bad investment decisions. One way to protect ourselves is to have a system and stick to it. This is one reason I love dollar cost averaging (DCA), which is a system to trade a small expected gain in exchange for a big reduction in point in time risk. Read more about it here if you aren’t familiar with it.

Over time, variable DCA is likely not any better or worse than regular DCA, though there will be individual variation. I recommend the system because it is easier to stick with when markets are tumultuous. The idea behind VDCA is simple: Invest less when assets are expensive and invest more when assets are cheap.

Decide how much and how often to invest

First, we’ll need to make some decisions on how we much we want to invest, how often we want to invest, and how complicated we want our system to be.

    1) How often will you be making an investment? Ex: I will be contributing to my investment portfolio once every 2 weeks.

    2) What is the average amount you want to invest per period? Ex. I will contribute $100 on average every two weeks. The exact amount will, however, vary week to week.

    3) How many possible investments amounts do you want? Ex. I will either invest $50, $100, or $150. You can be as coarse or as granular as you like. Simple is often better.

Create your VDCA system

Now, we do some quick analysis to set up our system. In my example, I will be contributing on average $100 to my portfolio every two weeks and will do this with one of three investment amounts ($50, $100, or $150). You can set whatever parameters you like.

    1) Find and download the historical returns of your portfolio. The more data you have, the better. If you don’t have your portfolio returns, you can use the return of the S&P500 since most portfolios are essentially the same.

    2) Calculate the historical returns for your investment frequency (e.g. 2-week periods). Note: This frequency does not need to be the same as your investment frequency, but it makes things simpler. I’m also ignoring auto-correlation here — it doesn’t matter a lot.

    S&P500 2w Return Calculations
    S&P500 2w Return Calculations

    3) Split the returns into the number of investment amounts you will be using (e.g. 3). We can do this using percentiles (e.g. 67th and 33rd) to split our 2-week returns into 3 buckets. Note: Effectively, we’re using percentiles to make the buckets equal in frequency (sort of), though this isn’t necessary. Roughly a third of the time we’ll see a return above 1.2%, a third of the time we’ll see a return between -0.2% and 1.2%, and a third of the time we will see a return below -0.2%.

    S&P500 2w Return Percentiles Calculation
    S&P500 2w Return Percentiles Calculation

    S&P500 2w Return Percentiles
    S&P500 2w Return Percentiles

    4) Apply your investment amounts to the appropriate buckets. If the market is up a lot (bucket #1), we’ll only invest $50 and if the market is down a lot (bucket #3), we’ll invest a full $150. On average, however, we should be investing about $100 every 2 weeks.

    Variable Investment Amounts
    Variable Investment Amounts

    5) Remind yourself to quickly check returns and make your investment. For example, you could put a calendar reminder every 2 weeks to make a deposit into your Wealthfront account. That’s it. You’re done.

Tweaks to make to your VDCA for fun

VDCA is merely a system to help you invest in the markets consistently. You’ll invest a little more when the markets are down and invest a little less when the markets are up. The specifics of how you do this matter much less than actually sticking to your plan through thick and thin and making the system work for you. Below are some modifications that you can make, should you so choose to.

    1) If your portfolio materially differs from the S&P500, use the historical returns of your portfolio instead of those of the S&P500. And no, the 10% in bonds don’t help! Use these 2-week returns to set your percentile buckets instead.

    2) Tweak the variable and fixed splits. In our example, we are essentially making a fixed $50 investment every 2 weeks. On top of that, we make a variable investment of $0, $50, or $100. You may tweak this mix however you like. Isn’t it great that whether the market is up or down, you’re always investing (except a bit more when it’s down and a bit less when it’s up)?

    3) Add more buckets and granularity to your investment amounts, if you really feel like it. Instead of having 3 investment amounts, you can have 4 or even 5. Don’t get too complicated.

    4) The buckets don’t have to be equal sizes. You could invest $50 if the market has a top 20% return (80th percentile) and invest $150 if it has a bottom 20% return (20th percentile) and then invest $100 the other 60% of the time (20th to 80th percentile). If you do this, don’t forget to update your expected weekly investment calculation: in other words you will be investing $100 3 times more frequently than you will be investing $50 or $150.

The big deal about Risk Parity

On the heels of Wealthfront’s recent announcement of a risk parity offering there has been renewed interest in this decades old concept. In this first post of a series about Risk Parity for personal investing I’ll provide a primer on the strategy and why it is a big deal.

Risk parity says we should allocate our risk, not our capital

Risk parity is a concept that suggests investors allocate assets based on their risk, not merely the capital allocation. A simple, common asset allocation is $80 stocks and $20 bonds. Stocks are on average 2-4 times as risky as bonds. As a result, while the share of capital is 80% stocks and 20% bonds the share of risk is closer to 95% stocks and 5% bonds. Lamest diversification ever.

Risk parity portfolios are based on the premise that allocating capital doesn’t accurately reflect the risk of the assets in the portfolio. Risk parity, not capital parity.

Risk parity has higher allocation to lower risk assets like bonds

Bridgewater Associates pioneered the risk parity strategy through their famous All Weather Portfolio in the 1970’s. Since then, various implementations of the concept have been brought to market by other asset managers like AQR and Putnam. Essentially, these implementations all have larger capital allocations to bonds and lower capital allocations to equities than standard portfolios to more evenly divide the risk in the portfolio.

Risk parity trounces standard portfolios on return/risk ratios

This is a big deal because a risk parity portfolio’s risk-based diversification offers higher returns versus the standard portfolio at equal levels of risk. A standard portfolio might achieve a 0.3 – 0.4 sharpe ratio (great primer), meaning that it provides 0.35 units of return for each unit of risk. For an average portfolio with an annualized standard deviation of 10%, this is a 3.5% annual excess return over cash. A well-managed risk parity portfolio on the other hand can deliver a sharpe ratio between 0.5 and 0.7 which translates to a 6.0% annual excess return.Note: To me the assertions here are supported by substantial evidence, but there are still smart people who will disagree with some of them. I’ll dig into some of of the assumptions in future posts.

Not everyone has access or ability to do risk parity

There are two reasons why this strategy isn’t far more widely used. First, not all investors have access to leverage and this strategy usually requires it. Second, professionally managed portfolios are expensive and/or have high minimums.

A standard $80 stock and $20 bond portfolio (Portfolio A) would return ~4.2% (12% volatility x 0.35 sharpe ratio). On the other hand, a $100 portfolio with equal stock and bond risk is $25 of stocks and $75 of bonds (Portfolio B) has a puny annual volatility of about 5%. This portfolio would return ~3% (5% volatility x 0.60 sharpe ratio). Since 4.2% is higher than 3% *and* 12% is acceptable risk for most under the age of 40, investors choose Portfolio A with a higher return.

Leverage (borrowing money to purchase securities) can double the excess return and risk of Portfolio B (200% leverage). We’ll achieve this by borrowing $100 and using it to buy another $25 of stock and $75 of bonds. This is new Portfolio C is $50 stocks and $150 bonds. Despite the use of leverage, it still has lower risk (10% vs 12%) and provides better returns (6.0% vs 4.2%).

Leverage lets us take superior, diversified portfolios and increase or decrease their risk to our desired level. We get more return for each unit of risk we take by doing this.

[yop_poll id=”1″]

cognitive biases investing

The Cognitive Biases That Plague Investors

It is very difficult to consistently invest profitably. Investment decisions are complex and as such are riddled with cognitive biases. Over a long enough time horizon even the most seasoned investors will succumb to many widely studied cognitive biases. They are human after all. This article discusses the cognitive biases that thwart investors. Some of these biases also prevent investors from sticking to tried and true investing advice.

The Investment Process

Investment decisions can be made in many different ways. Most of them involve answering different flavors of the same fundamental questions.

  • What opportunities are being sought?
  • What criteria are used to evaluate an opportunity?
  • What is the proper weighting of each criterion?
  • How are opportunities rigorously graded on each criterion?
  • What is a profitable purchase price?
  • Under what conditions should the asset be liquidated?

    • Navigating to the answers for each of these questions is fraught with risk of cognitive biases. The more these biases play into an investor’s decision the more likely the investor’s conclusions may be invalid. Below, I’ve listed some of the most common cognitive biases investors should be aware of. Additionally, I’ve included a brief example of an investment tactic or system that can mitigate the risk of this bias.

    Overconfidence Bias

    Overconfidence bias is a belief that one’s ability to beat the market is better than it truly is. More fundamentally, this effect means investors are more confident in their conclusions than they should be.

    Reducing the downside of being wrong is one way to mitigate this bias. Investor’s should always ask the question, “what if I am wrong?” and be comfortable with the answer to that question. Practically, setting (and more importantly, sticking to) position size limits can mitigate the effects of overconfidence. Setting reasonable limits ensures that investors do not “bet the farm” and lose everything if they are wrong. The risk of overconfidence is further mitigated by having a nuanced framework on how to size positions based on confidence level in the position. Forcing oneself to weigh the relative confidence of positions against one another encourages reflection on what one knows and does not know. Finally, incorporating consistent and measurable factors into the investment process will make it easier to spot overconfidence than if the entire process is qualitative. Occasionally revisiting past decisions and reviewing one’s success rate is often a humbling way to increase returns in the long run.

    Survivorship Bias

    Survivorship bias is most visible when investors are evaluating investment managers or making broad claims about the effectiveness of active management. The idea is that if hundreds of thousands of investors are betting on the markets, some will build an excellent track record not due to skill, but due to luck. This is a big reason investors often see the common disclaimer: “past performance is not necessarily indicative of future results”.

    Thoroughly understanding an investment manager’s investment process can mitigate the impact of survivorship bias. Investors should pay careful attention to the reasons a manager outperformed and not merely to the fact that they did outperform. An investor can develop a basis by which to differentiate the contribution of skill vs luck to the manager’s returns by deeply understanding the rationale behind the performance. Essentially, one must analytically assess the likelihood that a manager’s success was a byproduct of skill instead of luck. Separating the alpha and beta components of a manager’s returns can be a quantitative starting point for this discussion.

    Anchoring

    cognitive biases anchoring bias investing

    Anchoring is the tendency to place substantial (unjustified) weight on the first piece of information acquired in an investment decision. This often manifests itself as confirmation bias (don’t think I’ve forgotten) in favor of the original investment thesis. The current price of an investment is also an anchor that is frequently overvalued in decision-making

    Gathering informed opinions from others who have not yet been anchored is a method to mitigate the anchoring effect. Furthermore, managing an investment process to collect evidence before forming a hypothesis can also mitigate the effects of the anchor. A well-defined framework can specify which facts must be collected and what conclusions must be drawn without the need to start with questions about an existing hypothesis. Finally, a clear rubric assigning a weight (or ranges of weights) to various conclusions will limit anchor points from taking on more weight than they should.

    The bandwagon effect

    The bandwagon effect occurs when investment decisions are unduly influenced due to the conclusions of others. Imagine an investor who receives a stock tip from a friend. The investor may believe that his friend made an informed investment decision. In reality, that friend heard from his friend, who heard from her friend. This creates a situation where an investor may believe that 10 other disciplined investors did the work when in fact only one or two did. Many hedge fund analysts share their research with one another and as a result they have short (or long) positions in the same concentrated list of names (e.g Valeant).

    Following a disciplined investment process reduces the risk of herding. It is reasonable to weigh the opinions of others (based on your trust in them and your knowledge that the work they did deserves that trust). The risk, however, is overvaluing that opinion, especially when one has little understanding of why the recommendation was made in the first place.

    Loss Aversion

    The idea behind loss aversion is that humans have more aversion to loss than they have affinity for a gain of the same magnitude. The disposition effect is a specific case of loss aversion exhibited by investors. It is the tendency for investors to sell shares that have appreciated and hold onto shares that have an unrealized loss. No one likes taking a loss. The stock will bounce back eventually, right? right?

    Setting price limits and targets (though this is not without risks of it’s own) is a structured way to mitigate the impact of this effect. These targets and limits may be modified as new information is acquired. It is important not to let the disposition effect drive those changes, however, and ensure that the original investment process for calculating entry and exit points is followed. Finally, subordinating the weight placed on unrealized gains and losses (in fact an argument can be made that this should have almost no impact) can reduce the risk of falling prey to this bias.

    Confirmation bias

    cognitive biases confirmation bias investing

    Ah yes, the mother(fucker) of them all! Confirmation bias is the focus on finding and weighing more heavily, evidence that supports one’s initial hypothesis vs evidence that disproves it. Investors can browse their Facebook news feed for hundreds of daily examples of this phenomenon.

    As described above, adhering to the investment framework goes a long way in reducing confirmation bias. Adding detail around what criteria to use and how to measure it will reduce the likelihood that salient evidence is ignored. Additionally, playing devil’s advocate has helped substantially in my experience. Investors should independently try to make the strongest possible case for the investment as well as the strongest possible case against the investment. The investor can then think carefully about a trial where their life is at stake — they must make a case to a judge to go long or short. Which side is the investor most comfortable defending? This admittedly grim thought exercise compels decision-makers to carefully confront evidence on both sides.

    Other investing biases

    I’m tired of typing, so here is a laundry list of other biases to be cognizant of when investing:

    Availability bias
    Conservatism bias
    Zero-risk bias
    Pseudocertainty bias
    Outcome bias
    Nominal money illusion

    Unconscious Bias is Everywhere

    As should be evident, many of our investment decisions are steeped in unconscious bias. It is nearly impossible to remove all these biases from our decision-making. If we are cognizant of the pitfalls, however, we are far better armed to reduce the negative impact of the biases.

    One reason I am a huge proponent of roboadvisors is because they take everyday decisions out of our hands, frequently saving us from ourselves. A number of the systematic approaches outlined above to mitigate unconscious bias are programmed into Roboadvisors. The specifics of the algorithms have been rigorously tested.

    Investing is hard. Good luck out there.

    Socially Responsible Investing Will Soon Be The Norm

    For the average investor, I am a strong proponent of investing in a diversified set of low-cost index funds which likely provides market-matching returns over the long run (or damn near close to it). Even though most investors under-perform this easy to achieve benchmark, many cannot help but try their hand at beating the markets because they do not want to settle for “table stakes.”

    Given the reality that our cognitive biases, overconfidence, and desire to have fun in the markets frequently supplant our rational judgment when it comes to investing, why not make a positive expected value (I assert) bet with positive externalities? To this end, I recommend socially responsible investing as a way to “play the markets,” do real good, and potentially earn superior returns.

    There is a growing body of evidence that suggests that socially responsible investing (SRI) has the potential to outperform conventional portfolios over time and the purpose of this post is to provide a quick overview of logic that supports this argument.

    Go to where the money will be.

    Between today and my retirement, there is an estimated $30 Trillion of wealth that will transfer to millennials with almost $3 Trillion of that coming in the next 4 years. Pair this with the fact that millennials are 2x more likely to invest in companies that target specific social or environmental outcomes, and the fact that they are 2x more likely to exit an investment position because of objectionable corporate activity, and you are left with a large number of dollars that will systematically flow into socially responsible companies and out of objectionably behaved companies.

    This story by itself, if you believe it, is enough to generate a sensible investment thesis that is bullish on socially responsible organizations. The icing on the cake (gluten free and organic, of course) is that these large flows of capital have the potential to create powerful, self-reinforcing engines: cash flowing to these businesses lowers their cost of capital, increasing their efficiency (or growth potential if you don’t buy efficiency), and generating superior returns that in turn attract more investors.

    Too clean to fail.

    With the growing importance of these issues as expressed by millennials as well as scientific facts (global warming, water scarcity, etc.), it is not unreasonable to think government subsidies for sustainable practices, both in terms of dollars and regulatory benefits, will continue to grow considerably in coming years.

    Furthermore, there is strong evidence to suggest that the world will be afflicted with increased scarcity of resources (water, oil, fossil fuels) in our lifetimes. Business that begin to slowly shift to more sustainable practices will be well ahead of rising resource costs and business disruptions from supply shocks.

    Socially responsible investing outperforms empirically, too

    There is a study that studied studies that studied the relationship between environmental, social and governance (ESG) criteria and corporate financial performance (CFP). In total, some 2000 studies were evaluated. A staggering 90% of the studies suggests non-negative correlation between ESG and CFP. Put differently, 90% of the studies reviewed indicate that socially responsible companies perform equal to or better than those who do not follow that criteria.
    While there are macroeconomic and regulatory tailwinds for ESG-positive companies in the coming years, historical evidence has already provided significant evidence of financial out-performance. While the basket of low-cost index funds may be “the market” of today, there is reason to believe that it will under-perform the “new market” of tomorrow that is more accepting of ESG-criteria. Do you want to be left behind? And implicitly support the destruction of the earth? Of course not.

    Disclosure: I recently became an adviser to OpenInvest, a Y Combinator-backed startup whose mission is to make socially responsible investing the norm.

    Roboadvisors are good investment vehicles

    I write today in sadness and frustration.

    I hesitate to link to Blake Ross’ recent article on the tyranny of Wealthfront because it strikes a blow that sets my friends’ and generation’s investment goals back immeasurably. While I fear his colorful writing and celebrity status will likely confuse people into inaction, I hope this article will inspire you to action and shed light on some the inaccuracies and misrepresentations in Ross’ article.

    Before I begin my response, I implore readers: If you’ve been evaluating Wealthfront, Betterment, Vanguard Retirement Funds, or Schwab Intelligent portfolios over the last few months (or years) and can’t decide which to use — for the love of whatever you believe in, stop reading this post, roll a 1d4 and just pick one. They’re all better (by a large margin) than you leaving it in your checking account, spending it on something stupid, or giving it to an expensive active advisor. Go. Do it. Now. Which one did you choose? As the Rock would say, “it doesn’t matter [what you chose, as long as you chose].”

    Time for business.

    There is a lot of glossy rhetoric in Ross’ post and while I appreciate the passion and his renegade approach, I am upset by the misinformation. I’ll distill the rhetoric down to the key arguments and address those.

    Argument A: You’ll make more money with Vanguard Retirement Fund than a Roboadvisor – This is a strong claim that in Ross’ words, “There is simply no evidence, nor any theoretical reason, to believe”. It’s the hardest claim to prove but is also most important of the arguments. Simply put, the Wealthfront portfolio costs slightly more (quantified below) annually but probably makes up for this difference and then some due to tax loss harvesting (enough to close the gap alone), a likely superior asset allocation, a likely superior rebalancing strategy, and a faster rate of innovation.

    Argument B: Roboadvisor fees increase over time – This isn’t a particularly unique thing, and I’m not sure what the ultimate impact is of this argument, especially since Ross’ alternative suffers the same “flaw”. A number of business models include fee structures like this, especially when the benefit of the product increases over time as well. The fee does not rise on a percentage basis, but only on an absolute basis, and only if the Roboadvisor is making you money.

    Argument C: Tax-loss harvesting is too good to be true and the benefit is overstated – Ross agrees that there is an aftertax benefit to tax loss harvesting. His argument is that the magnitude of the impact is too small to matter. Unfortunately, even if we cut Wealthfront’s estimates by 10x and then refer to Ross’ citation about Tax Loss Harvesting, both sources claim an annualized benefit of at least 0.20%, which is already enough to push Wealthfront above Vanguard Retirement Funds. Finally, the author of Ross’ only substantive claim against the benefits of Tax Loss Harvesting actually supports Roboadvisors, and even Wealthfront specifically.

    Argument D: Roboadvisor fee structures are unprecedented – No they’re not.

    Note that while Ross decides to aim his critique primarily at Wealthfront, his arguments apply to Roboadvisors at large and so I’ll genericize my response where I can. When I use specific fee figures, I’ll use Wealthfront’s. Also note that Ross’ alternative is by no means bad, it’s just not better, and certainly not better to the degree that justifies the tone of his critique. I love Vanguard. They’re great.

    Onward…

    A. You’ll make more money with Vanguard Retirement Fund than a Roboadvisor

    1. The true fee difference is at most 0.17% in Vanguard’s favor, likely less – Ross overstates the cost difference between Wealthfront and Vanguard. Wealthfront’s portfolio costs at most 0.35% annually (closer to 0.33% for accounts with Direct Indexing). This fee is composed of a 0.25% fee paid to Wealthfront and a 0.10% fee paid to the administrators of the ETF that Wealthfront invests you in (mainly Vanguard, by the way). Ross’ “dirty secret” alternative (Vanguard Retirement Fund) charges 0.18%. Below is a comparison of total expenses for Wealthfront vs a Vanguard Retirement portfolio. Wealthfront manages $10,000 for free and adds an extra $5,000 managed for free for every client you refer to Wealthfront, reducing your blended costs. The table below assumes the minimum of $10,000 managed for free.

    Wealthfront Vanguard Fee Comparison

    Now, that means we need to see if Wealthfront can overcome 0.07% – 0.17% in annual fees. Let’s get to work.

    2. Tax Loss Harvesting by itself makes up for the difference and then some – Wealthfront claims Tax Loss Harvesting can add approximately 1.29% annually (2.03% if one includes tax-optimized direct indexing). While I agree with Ross that this figure is huge and potentially overstated, there’s (a) no evidence he provides to the contrary and (b) just ignoring it is hardly sensible. So even if we decimate Wealthfront’s claim, we get to a 0.20% benefit from TLH with TODI. So this means that 1/10th of Wealthfront’s claimed Tax Loss harvesting makes up the fee difference by itself, and then some. I contend that Ross’ arguments show a misunderstanding of what TLH is and the mechanics of how it works. I’ll dive into this in depth in the TLH section.

    3. Ross likely misrepresents Buffett’s viewpoint – Furthermore, he mis-characterizes a Buffett quote as evidence to assail robo-advisors: “… Put 10% of the cash in short-term government bonds and 90% in a very low-cost S&P 500 index fund. (I suggest Vanguard’s.) I believe the trust’s long-term results from this policy will be superior to those attained by most investors — whether pension funds, institutions or individuals — who employ high-fee managers.” First, if you asked Mr. Buffett whether 0.37% is a high-fee manager, I’d wager he wouldn’t think it was too bad given the management fees being in the 1-2% for most of his lifetime and given Ross’ own argument that WF has significantly lower fees than other advisors and given that Ross’ alternative has fees that are less than 0.20% lower. Second, I’d also place a wager that Buffett is referring to active managers (most Roboadvisors are passive asset allocators). Third, Wealthfront disputes Ross’ claim directly in an article that compares the Wealthfront portfolio to Buffett’s recommendation. Finally, Buffet’s allocation is obviously an oversimplification that can and should be improved upon. Or is it? Let’s see what the experts have to say about the importance of a diversified asset allocation.

    4. Asset Allocation matters a lot – In a seminal 1985 paper it is argued that asset allocations explained almost 94% of a portfolio’s return versus other factors contributing 6%. The administrators of Mr Ross’ alternative also believes this too as evidenced by a Vanguard-published paper stating that almost 77% of portfolio performance comes from asset allocation decisions (they removed some factors from being classified as asset allocation that the Brinson paper includes). If academics don’t convince you, fine.

    5. Good investors also think asset allocation matters a lot – Some not-so-bad investors say stuff too. “Asset allocation is critically important; but cost is critically important, too—All other factors pale into insignificance.” -John Bogle, founder of Vanguard. “The most important decision you will probably ever make concerns the balancing of asset categories (stocks, bonds, real estate, money market securities, etc.) at different stages of your life.” -Professor Burton Malkiel, author of A Random Walk Down Wall Street, way before he became Wealthfront’s CIO. “Choose your asset allocation model carefully. Asset allocation is the biggest factor in determining your overall return.” -Charles Schwab.

    6. Figuring out asset allocation difference is hard, we have to evaluate it systemically – Now, the question is whether Wealthfront’s asset allocation is better or worse than the Vanguard retirement fund’s. This is a very, very tricky question to answer because there’s really no concrete way to know aside from just seeing how they perform. The issue with that approach is you don’t know systematically which (if either), will outperform. You only know what actually happened, which is significantly influenced by chance. What we can do, however, is understand the process each manager employs to see if there are systematic biases to outperform the other or not.

    7. Vanguard portfolios are more constrained than Wealthfront’s – Vanguard’s Retirement portfolios use their own funds. Wealthfront has no such requirement. That means that in cases where an asset allocation would be better served by having a non-Vanguard fund that provides a certain set of exposures, Vanguard must find a close proxy, create a new fund, or risk underperformance. Wealthfront has the opportunity to find the best alternative (which happen to be Vanguard MOST of the time). It’s tough to quantify, but the argument is simply that a constrained portfolio is likely worse than a less constrained one.

    7. Wealthfront’s rebalance strategy is likely more sophisticated than Vanguard’s -Furthermore, Wealthfront’s rebalance strategy is optimized to minimize drift while keeping costs at bay, a strategy advocated by Vanguard. While Wealthfront has developed software to assess the trade-off between these two sides in real time (it operates continuously), Vanguard does it daily (because they receive fund inflows and outflows daily) incurring small costs on the fund each time and requiring discrete decision-making which is not real-time. Additionally, there is evidence to believe that Vanguard uses time-based and threshold based triggers, which they admit in their own white paper have issues. While the impact here is likely small, it is yet another systematic benefit to an automated strategy like Wealthfront’s.

    8. Wealthfront’s rate of innovation is fast – Finally, Wealthfront has been at the Vanguard (see what I did there?) of change in the industry for a long time. I will say that Vanguard’s innovations are great, but Wealthfront has been the Roboadvisor that has consistently pushed the envelope, developed new strategies, reduced cost via cheaper ETF’s, and made more features available to smaller investors faster. Others follow suit of the star-studded investment team from Wealthfront. Purely on a trust basis, I would trust the investment analysis (I’ve verified enough myself) of Wealthfront. Even beyond that, there’s a tangible benefit being the leader of innovation because your investors realize the benefits of the features while investors in other funds must wait for the innovations. Wealthfront discusses this specific argument here.

    9. Ross’ critique minimizes the impact of human psychology (and lack of logic) on financial decisions – Ross says, “If you open a retirement account, and you invest some of your paycheck each month into a Vanguard Target Retirement Fund, and you just…leave it… until retirement… you don’t do anything when the folks on CNBC announce that the sky is falling; you don’t do anything when Cousin Eddy calls from a secure underground bunker in the badlands and says that the fed is printing money and it’s time to liquidate and ammo up; you don’t think it’s a sign that your parrot said “fuhgeddaboutit” but you thought she said “get a nugget” and surely that must mean a gold nugget? and you looked online and noticed that the price of shiny yellow metal was crashing and wait your parrot is also yellow and I’ll be damned if that isn’t a sign to buy… no, if you just leave it there to compound over decades… then you will probably make more money than … if you used Wealthfront.” That’d be nice, but it rarely happens. The story of Peter Lynch and the Magellan Fund is a cautionary tale. He ran the Fidelity Magellan fund for 13 years and achieved a ~30% annualized return. The average investor in the fund, however, actually lost money. This happened because investors would fear on the dips and sell and chase returns after a big run up. They essentially broke the cardinal rule of investing by selling low and buying high, over and over again. An automated advisor protects you from yourself marginally more than a portfolio you need to manage more (or even a Vanguard Retirement Fund which is psychologically easier to view as a single investment rather than money with an advisor).

    10. Vanguard Retirement Funds are either less tax efficient (in taxable accounts) or they are less optimized (in non-taxable accounts) than the Wealthfront portfolio — Wealthfront offers different asset allocations for taxable and non-taxable accounts in order to optimize tax treatment. The Vanguard Retirement Fund is a single entity that holds a single asset allocation is not individually optimized for one type of account or the other. The impact of this is non-negligible.

    B. Roboadvisor fees increase over time

    1. Yes, they do – But only if they make you money. There doesn’t seem to be a differential implication presented.

    2. It’s pretty common – In fact, many businesses (even those not on Wall Street) have a business model that charges increased fees based on usage. Many SaaS companies charge by seat. Many data businesses charge by volume of data processed/stored. There seems to be no distinction drawn between Roboadvisors and these other businesses.

    3. Vanguard’s got the same “issue” – In fact, the suggested alternative (Vanguard Retirement Fund) charges exactly the same way.

    4. What do you believe? There’s a philosophical / value-based question on whether you think the fee should be proportional. If the value being provided is proportional (it is for many SaaS companies, and it is for Wealthfront too), then I personally find it fair to charge a fee in this way. You’ll have to figure out for yourself what you believe on this one.

    5. The structure aligns incentives – Furthermore, Wealthfront is better than a number of the alternative subscription-based models in that the fee only increases if your outcomes improve (portfolio value increases). SaaS models that charge by seat cost you more as your # of seats grow, but don’t cost you less if the seats do dumb things with the product or if the seats use the product less. It’s incentive alignment, not predatory pricing.

    6. Ross misrepresents rarity of the model – He argues, “It’s not just that Wealthfront charges users for its software, which is rare.” This one made me angry so I had to take a break to watch Netflix and listen to Spotify before coming back to blog on my personal domain. He also argues, “It’s also that, on average, Wealthfront increases its subscription fee every day.” Yes, I agree. But the question is whether you are OK watching your investment balance grow by $100 while watching the fee grow by $0.25. If you’re not, then I ask about the alternative. His alternative is structured similarly.

    7. Roboadvisors are not merely providing advice – Finally he says, “Stop charging proportional fees for advice,” to which I say Roboadvisors provide more than merely advice. They provide services like trading for you, keeping up with changes in ETF fees and landscape, rebalancing intelligently instead of periodically, tax loss harvesting and performance tracking. As mentioned earlier, a number of the services above have benefits that in fact do scale with the size of the account. The implication of the statement above is that one should pay fixed fees but gain bigger benefit over time from those fees.

    8. Wealthfront has never been profitable – In response to the “Wealthfront helps itself to such margins” argument, it is important to note that the company is not profitable and won’t be profitable any time in the next few years. You can write that in pen. On the 2bn they manage, they’re revenue is less than $5mm. They’ve lost money for every year they have been in existence. Vanguard runs at cost, according to Ross’. I’d prefer to let the VC’s subsidize me.

    C. Tax-loss harvesting is too good to be true and benefit is overstated

    1. We need to save in taxable accounts too – Ross argues that “If your nest egg exists entirely in a retirement account, as it does for many Americans, then tax loss harvesting won’t help you at all.” While this is true, I’d argue that it is a big deal that Wealthfront is providing incentive (TLH) to invest outside retirement accounts, given the dire state of savings rates in our generation.

    2. ETFs have capital gains too – With Ross’ argument that “If you practice the kind of investing that Wealthfront itself evangelizes — buy-and-hold, passive, rational, long-term indexing that is rebalanced with new money or in retirement accounts — then you should not be realizing capital gains regularly anyway.” he fails to mention that his alternative also recognizes capital gains regularly. Most ETF’s distribute capital gains at the end of the year and do not harvest losses, so you’re basically paying capital gains taxes in either scenario. Furthermore this cuts against Ross’ argument that tax loss harvesting gains are capped while Wealthfront’s fees are uncapped.

    3. Even rudimentary TLH is enough for Wealthfront to come out on top – Furthermore, Ross’ argument against Tax Loss Harvesting as made by the Kitces article above actually says that 0.20% improvements are based upon lumpy and dumb harvesting (once/year, at the end of the year) versus a much more sophisticated, real-time algorithm employed by Wealthfront.

    4. Ross’ own citation makes a pretty strong case for Roboadvisors – The same author who Ross cites also writes a pretty good article about How Declining Transaction Costs And Robo-Indexing Could Disintermediate Index Mutual Funds And ETFs.

    D. Roboadvisor fee structures are unprecedented.

    No they’re not.

    Conclusion

    Invest your money in low cost ETF’s. If you have a taxable account and don’t want to spend time on it, I recommend Wealthfront. If you don’t have a taxable account and have time, you can mimic Wealthfront’s allocation and rebalance yourself. You can pick Betterment. You can pick Wise Banyan. You can pick Schwab Intelligent Portfolios. You can pick Vanguard Retirement funds. Do something with your money. But please don’t be paralyzed into inaction by rhetoric. The Roboadvisors will also help you avoid destructive cognitive biases.

    Wealthfront vs Schwab Intelligent Portfolios: A Third Take

    Recently Adam Nash, the CEO of Wealthfront published a piece on medium lamenting how Charles Schwab lost its way, citing the launch of the Schwab Intelligent Portfolios (SIPs) as evidence. Soon thereafter, Charles Schwab responded, calling part of Nash’s example “criminal” (a trope used by Nash himself to describe Schwab’s approach). In this article, I’ll breakdown the arguments and illustrate how even in the best case scenario for Charles Schwab, their SIPs are still a bit worse than Wealthfront’s portfolio.

    The Wealthfront Argument:

    1. The SIPs hold sub-optimally large cash balances.
    2. Schwab pays a sub-market return on those cash balances.
    3. SIPs are self-serving because they use a sub-optimal asset allocation to funnel assets into Schwab’s preferred funds.
    4. SIPs preferred “Smart Beta” ETFs have high fees that are not justified by the value smart beta adds.

    The Charles Schwab Argument:

    1. SIPs cash balances are totally reasonable sizes. Good portfolios have cash balances.
    2. Smart beta portfolios are acceptable and outperform traditional market-cap weighted indices.

    The Cash Question

    Nash says SIPs cash balances are too high. Schwab says they’re reasonable. Who do you believe? It turns out the more important fact is Nash’s 2nd argument that Schwab pays a sub-market return on cash balances, a claim unrefuted by Schwab. According to Nash’s figures, Schwab pays 0.87% less on it’s cash balances than a leading competitor bank (0.99% vs 0.12%). SIP’s with a 6% cash balance and a 30% cash balance create a 0.05% and 0.26% drag on the entire portfolio (relative to the leading competitor bank) because of Schwab’s stingy interest paid on cash balances. This is illustrated in the table below.

    Schwab Cash Drag

    If Schwab was truly trying to be consumer friendly, they could have chosen 1 of 2 approaches that would remain consistent with their viewpoint that moderate cash balances are prudent investing strategy (i.e. if you disagree with Nash’s view that low cash balances are better):

    • Schwab could use the Wealthfront approach by keeping minimum cash on hand and encourage clients to leave moderate cash balances (that are currently being held in SIPs in the sweep allocation) in their external accounts with higher interest rates. This would meet Schwab’s “cash balance is prudent” requirement as well as provide optimal cash return to the client.
    • They could just offer a competitive interest rate on cash balances.

    • But they chose to do neither which leads me to conclude that Nash’s explanation probably rings true: “follow the money.”

    Asset Allocation: A story of mis-aligned incentives

    Wealthfront’s fee is charged for the management of a portfolio of ETF’s. The management includes smart asset allocation, optimized rebalancing, and automated tax loss harvesting. To earn it’s management fee, it is in Wealthfront’s best interest to select a low-cost and appropriately diversified set of ETF’s — so the portfolio performs well, clients stay happy and the asset base grows (benefitting both Wealthfront and the client). This is a dream incentive alignment for clients.

    Schwab’s Intelligent portfolios on the other hand, are mired in some incentive alignment issues. While SIPs fee structure also incents asset base growth, there is also a strong incentive to steer clients towards Schwab products, and expensive Schwab products in particular. In fact, SIPs allocate a large portion of assets (I have seen 60% quoted in some places) to Schwab’s own smart beta products. These smart beta products average an expense ratio of ~0.35%, roughly 3 times the cost of Vanguard ETF’s used by Wealthfront.

    The tough question is whether Wealthfront’s asset allocation is better or worse than Schwab’s asset allocation. Schwab makes the claim that their smart beta portfolios do indeed outperform traditional asset mixes, though I have yet to see definitive evidence of this or even a quote on the magnitude of the difference in performance. Some in-depth analysis (which I’d love to see Wealthfront or Schwab actually do) will be required to answer this question. What I can say though is that it seems a little too convenient to me that SIPs utilize Schwab’s high-priced products in such large quantities.

    Conclusion

    In the best case scenario you believe that (1) Schwab is not nudging allocations to their benefit and (2) a reasonable cash balance should be held in a portfolio. Unfortunately, even in this base case scenario, SIPs still slightly underperform a Wealthfront portfolio.

    The Wealthfront portfolio costs 0.25% plus the underlying cost of the ETFs (~0.12%) for a total annual expense ratio of ~0.37%. We’ll assume that we invest 90% of our cash in this portfolio and 10% in cash at a leading competitor bank. This means the fee we pay to Wealthfront is 0.33% (0.37% fee * 90% of portfolio) and we earn a 0.10% contribution to the portfolio from cash (10% balance at 0.99% interest).

    While SIPs have no management fee, they do allocate to underlying funds costing 0.20% – 0.40%. If we take an average SIP, we’ll pay 0.27% in fees (0.30% underlying fund costs * 90% invested) and we will earn a 0.01% contribution to the portfolio from cash (10% balance at 0.12% interest).

    What this nets out to is that the Wealthfront portfolio is 0.06% more expensive in fees, and earns you 0.09% more in cash return, netting out to a 0.03% advantage for Wealthfront in the most bullish scenario for the Schwab Intelligent Portfolios. Plus, don’t forget Wealthfront’s Tax Loss Harvesting on all taxable portfolios (Schwab has a significant minimum balance requirement for TLH).

    Now, at the end of the day, if you’re not investing your money because you can’t decide between Wealthfront, Betterment, Schwab Intelligent Portfolios, or Wise Banyan — you should just roll a damn 1d4 (if you don’t have one, I’m happy to roll mine for you) and pick one, because they’re all far better than your cash sitting in a checking account.

    Tried and True Investing Advice

    I’m often asked to provide investment advice. My advice is hardly revolutionary. It has been repeated by many others (e.g. John C. Bogle) yet is consistently ignored: passive index investing is the way to go. My goal in this article is to change this behavior for at least some people. I plan to do this by (1) laying out my own version of the advice, (2) providing the rationale for this advice, (3) pointing out behavioral reasons why people do not follow the advice, and (4) providing some tangible ways you can prevent yourself from falling into the trap of not following this advice. In my view the last piece is critical because it is a fundamental reasons why asset allocations are so poor.

    Before I go any further, my disclaimer: I am not a licensed financial adviser and I am providing this write-up for informational purposes only as an expression of my opinions and views. Any investment actions you take as a result of reading this article are solely your responsibility. If it works out, feel free to thank me later. In case it does not work out, I take no responsibility. If you do not agree to that, then do not read any further.

    My Advice

    For most individual investors, I recommend a diversified asset allocation consisting of low-cost index funds. Factors to consider when making this allocation are your investment horizon, risk tolerance, and desire for simplicity. Let me briefly define the 5 key terms I have used:

  • diversified asset allocation – Diversification comes in many forms and the more dimensions you are diversified across, the better off you are in the long run. At the highest level, you should be diversified across asset classes (stocks, bonds, commodities, real estate, etc.) Furthermore, it is good to be diversified within each of the asset classes as well (by geography, market capitalization, type of commodity, etc.) At some point, more diversification has diminishing returns and you begin to create a portfolio that violates the final rule of simplicity. Where to draw the line is up to you.

  • low-cost index funds – The long-run impact of fees on investment returns is staggering, so you should seek to own funds with low expense ratios that seek to passively replicate the returns of an index. I will discuss why below. For some calculations on the impact of fees, you can read this post on Moolanomy.

  • investment horizon – This is a fancy term for how long you have before you will want to utilize the money you have invested. Most of this article is related to investment horizons of 5yrs or more. For shorter horizons, the advice is likely going to be different. That is beyond the scope of this article. If people want it, I can post on shorter time horizons at another date.

  • risk tolerance – How much risk are you willing to take? In the investing world, you cannot get a return higher than the risk-free rate unless you take on risk. Generally, the more risk you take on, the higher the potential you have for return.

  • desire for simplicity – We can run down a rabbit hole of complexity, but ultimately, you need to be comfortable managing your portfolio. A portfolio of 5 names is easier to manage than a portfolio of 15 names. We’ll constantly be trying to balance diversification and investment awesomeness with simplicity.

    • The Logic

    It is extremely difficult to beat the markets consistently and most managers cannot do it. Given the difficulty, “good” managers charge high fees. This means they must perform even better to outperform the market net of fees. As a result, I believe there are only a handful of people in the world that are capable of consistently beating the markets – and very few of them are likely to be reading this post. You may not believe me, but if you think you can beat the markets (or know someone who can), you are not only playing against the odds you are also likely on the wrong end of bias about your own abilities (more about this below).

    So you are left with a choice. You can either play the market and pay 0.25% in fees or you can try to beat the market (the expected value vs the market here is zero since for every winner there is a loser) and pay 1.00% in fees. This means that not only do you need to be a winner in your picks, you also need to consistently win big so you can overcome the extra fees.

    This is where people tell me all sorts of things that they think matter (they don’t really matter, trust me):

  • I am smarter than the average investor
  • My financial adviser went to a top MBA program
  • I have all sorts of cool tools from E*Trade to give me the edge
  • My strategy hasn’t lost money in 40 yrs
  • I am a really good stock picker, because I know a lot about the industry
  • My financial adviser has a great track record

    • So why does none of this matter?

    Behavioral Impediments

    There are behavioral impediments that obfuscate the difficulty of consistently beat markets in the long run. First and foremost is the overconfidence effect, which indicates that we are more confident in our ability to accomplish something than is warranted. Second, we attribute excessive weight to past performance because we fail to consider the likelihood of the manager’s performance being random. In other words, if I told you that I flipped a coin heads 5 times in a row, that might be impressive if I only threw the coin 5 or 6 times total. It would not be impressive, however, if I told you I flipped it 500 times. Not knowing “how many coins were tossed,” is a manifestation of survivorship bias. Survivorship bias is the idea that we do not know how many money managers attempted to match your managers track record and failed. Third, we tend to focus on what is sexy and cool, which in the investing world often means equities. For this reason, the equity markets are among the largest, most liquid, and efficient markets in the world – meaning they’re harder to beat. Other markets are far simpler to beat (the currency markets, as an example, includes many players who are not profit-maximizing). That’s not to say that beating the currency markets is easy, it isn’t. I digress…

    If you do not think you can beat markets, the low-cost index fund strategy is a great way to go. If you do think you can beat markets, you are probably wrong, but you can give it a try (consider your negative expected value the price you pay for the fun of spending your time looking at stock screens, reading 10-Ks or reading credentials of potential financial advisers.)

    If you’re an active investment decision-maker, be sure to check out ways to mitigate destructive cognitive biases for investors.

    Overcoming

    Overcoming these cognitive biases is difficult, but I have found these tangible actions and thought exercises have helped me come to grips with the fact that I will not beat the markets (I do what I described above for 90% of my asset allocation. The other 10% is for fun.)

  • Try running a paper portfolio – This is a portfolio with play money where you actually put on trades. This is a little bit more work than just writing down your ideas on a piece of paper, but it keeps you honest and prevents you from selectively ignoring losers because “I would not have actually put that trade on” or fudging the numbers.

  • Fun Allocation – Set aside a small piece of your asset allocation for fun bets (5-10%) that you do your research on and track. Over time, if you really are killing it, you can consider raising your allocation to decision-making.

  • People are lying to you (not maliciously) – Count the number of people who tell you about their good investing outcomes and the ones that tell you about their bad investing outcomes. They should presumably be the same, but they never are. The reason is because those who tell you about their killer investing strategies are conveniently ignoring their losses or they just got lucky (and you won’t know if it was skill or not unless they’re doing it consistently.) This should at least give you comfort that you’re not missing out (i.e. it’s not like everyone is out there killing it in the market except for you.)

  • Your time vs. Others’ time – Consider the time you spend thinking about your investments on a daily basis. Then consider the time spent per investment. Finally, consider that many people have full-time jobs dedicated to understanding the company you are researching better than anyone else. Then consider the vast resources they have to achieve this. Even if you did think you could beat them, is it worth spending all of that time to try?

    • Final Thoughts

    I do not mean to offend anyone by telling them they cannot beat the markets. While it is likely true, I could be wrong in any individual case. The main argument is that with a well-structured plan of saving from a young age, merely following the market passively is easy (very little work) and will provide a very reasonable nest egg for retirement. Trying to beat the markets is time-consuming and likely to fail, so I’m very content with the table stakes.

    Scroll to Top