Beat Behavioral Bias: Mental Accounting

A lot of economic and financial research deals with behavioral biases, those occasions where the mind plays tricks with us and leads even very intelligent people down the path of irrational and sub-optimal decisions. Other bloggers have pointed out some of these biases before, see Plan Invest Escape on cognitive biases. Also, Northern Expenditure wrote an interesting post on the temptation of instant gratification over saving for the future. Among all the different biases, Mental Accounting is not that well-known but it’s one of the most fascinating. Mental accounting, sometimes called Framing, shows up in human behavior in the following ways:

Intentionally or unintentionally creating different buckets of money and ignoring the fact that money is fungible; displaying different degrees of risk aversion and/or different propensities to consume out of different buckets.

Quite intriguingly, in personal finance the mental accounting bias is not only committed frequently, sometimes it’s even celebrated as a great innovation. It’s not a defect, it’s a feature! Some of the well-known financial gurus fall for this fallacy and are not even ashamed!

The stereotypical (and relatively benign) Mental Accounting example is this: if you find a $20 bill between the sofa cushions, would you blow this windfall money on something you wouldn’t have bought if the money had been in your checking account all along? If yes, you seem to have different propensities to consume out of different pots of money. But money is fungible. In deciding if and how to spend the $20, it should not matter where the money came from. We gathered some other examples of mental accounting from around the web of how human behavior might display Mental Accounting:

They are all quite interesting and entertaining to read but still leave the impression that this bias is relatively benign, more of a little quirk that causes irrational behavior concerning trivial amounts like the $10 movie ticket or $50 of money blown on a casino visit. Who cares?

We should care because if you look carefully, Mental Accounting is everywhere in the personal finance world. We committed it many times ourselves. But when applied to large portfolios, potentially in the multi-millions, the damage is no longer trivial. Here are some examples:

1: Emergency fund

We are no friends of emergency funds. For us, our entire investment portfolio is our emergency fund, as we described in an earlier post. Having an outsized emergency fund in a money market account could be a sign of Mental Accounting. The one way to rationalize the emergency fund held in cash would be the risk mitigation argument, though it’s not that convincing. Specifically, take an example of someone with $100,000 in financial assets, $80,000 in stocks and $20,000 in an emergency fund held in a money market account. I can fully understand that someone is uncomfortable with the roughly 15% annualized risk a 100% equity portfolio would bring. So, if someone wants 12% risk and not a single basis point more one could argue that 20% cash is one way of getting there. But it’s not a very efficient way because you could also get to that same expected risk level by mixing in other uncorrelated or even negatively correlated assets (such as longer duration bonds) that have higher expected returns. You will get a higher expected portfolio return at the same risk level.

Side note for the Finance wonks: if you pick any point along the efficient frontier past the tangency point you have to hold zero cash. Thus, risk reduction is never done by raising cash, but rather by keeping cash at 0% and changing the asset mix along the efficient frontier, see the efficient frontier plot below and our analysis here.

EffFront with Emergency Fund
Efficient Frontier and why the Emergency Fund shouldn’t be in a money market account
By artificially constraining your portfolio into one risky and one risk-less bucket for emergencies, you likely get lower returns than having optimized one large portfolio with the same overall average risk target. So all the financial gurus out there who preach that the emergency fund can’t be invested in anything risky are making this mistake, violating Finance 101. And everybody goes “ooh” and “aah” about how smart this advice is. It’s not a defect, it’s a feature! But it’s an irrational bias and reduces expected returns.

2: Risk management and portfolio optimization by bucket

The emergency fund fallacy is actually only one facet of a much wider-ranging fallacy; the asset allocation decision in separate buckets. That’s typically sub-optimal and there are many examples of it:

Example 1: Robo-advisers

Robo-advisers are celebrated as the new rocket scientists in Finance but the idea of coming up with one target portfolio for the taxable account and one for the tax-deferred account as is practiced by Wealthfront and Betterment is entirely preposterous. There is a short questionnaire beforehand to determine the likely risk-aversion level, but even for a given risk aversion level, you shouldn’t optimize the taxable and retirement accounts separately. Examples:

  1. For the least risk-averse, Wealthfront recommends a 16% share in REITs in your retirement account and 0% allocation to REITs in the taxable account. How can that be optimal if we don’t know what’s the share of your taxable vs. retirement account? If it’s optional for someone with a 50/50 allocation between taxable and tax-deferred account (thus 8% average allocation to REITs), how can it be optimal for someone with only a tax-deferred account (thus 16% REIT allocation), or someone with only taxable accounts (no REIT allocation at all)? It doesn’t make sense at all!
  2. Both Betterment and Wealthfront recommend Municipal Bonds (tax-free interest) in their taxable portfolio. Betterment recommends even taxable bond ETFs in the taxable portfolio. That may be OK if you have only a taxable account and require some bond allocation. But if you also have a retirement account you’ll likely be better off keeping taxable bonds there and only equities (with qualified dividends) in the taxable portfolio, see diagram below.
  3. We list several more examples detailing how an individual with both taxable and retirement accounts would experience a sub-optimal allocation using the Robo-advisers. See our blog post here.

Mental Accounting Diagram
Portfolio Optimization within different buckets (Mental Accounting) recommend by Robo-advisers is likely inferior to Global Portfolio Optimization
Whether this is ineptitude or laziness on behalf of the Robo-advisers, these are all textbook cases of mental accounting and they all generate sub-optimal allocations. Your portfolio allocation should be done globally over your entire portfolio first. Then you decide where to hold what, e.g., bonds and REITs in tax-advantaged accounts, equities with the lowest dividend yield in the taxable account first, then filling up the remainder of the tax-advantaged accounts.

Example 2: real assets/inflation protection

If I had a dollar for every time I heard things like “you need 10% real assets in your portfolio” or “you need 10% REITs” in your 401(k)” I’d be a rich man. How much you need in inflation protection depends on many factors. Are you a homeowner or renter? Do you own rental property? Will you get a pension with Cost of Living Adjustments (COLA) or a nominal payment or none at all? If you are a public servant with a pension that has COLA (cost of living adjustment), you may not need much inflation protection at all. Your goal shouldn’t be to inflation-hedge you 401(k) account, but to inflation hedge your entire lifetime financial situation, including (but not limited to) your retirement cash flow. 

Example 3: ignoring other issues and risk factors outside of the portfolio

Back to the Betterment/Wealthfront Muni bond allocation: The Muni bond allocation seems independent of the income level (even though there’s a question about the income in the initial interview). Individuals in a low tax bracket may be better off with taxable bonds that tend to have a higher yield. Munis are mostly interesting for folks in the 35% and 39.6% federal tax brackets. Again, maximizing the portfolio allocation within one bucket without regard for the overall financial situation is sub-optimal.

3: Taking budgets too seriously

Having a budget is important, some even find budgets sexy. But taking the budget too literally could set you up for overspending. For example, if by the end of the month we are under budget in one category, there could be a temptation to “blow” the money on this category (or another for that matter). But some categories should always be under budget to hopefully average out unexpected spending shocks. Prime example: owning a home and/or a car one should clearly budget for replacing, repairing and maintaining the major components. If you have a budget of, say, $30 a month for fixing a certain component, and it doesn’t break in any given month, that’s not free money. Save it for when something does break and the $30 budgeted for the repair that month will likely not suffice.

How to avoid mental accounting:

  • Repeat after me “money is fungible”
  • Sometimes, splitting a larger problem into several smaller problems is the smart way to go, but in personal finance, you don’t want to miss the forest for the trees. Risk management and portfolio allocation should be done holistically, rather than separately in each bucket. Since the bucketing is irrational and leads to sub-optimal results, our solution is to have one and only one bucket: our investment portfolio. Of course, it is invested in different assets, but we always look at the portfolio as one pot of money.
  • Careful with budgets. Of course, budgets are important. But if towards the end of the month/quarter/year there’s money left in the budget, don’t succumb to the temptation of overspending the leftovers. Use that as an excuse to lower the budget for that category in the future or save the excess for the inevitable large expenditure on major repairs.

24 thoughts on “Beat Behavioral Bias: Mental Accounting

  1. You bring an interesting behavioural bias. In fact, this bias is big time present at our house and in our budget. We have buckets for everything. And yes, I am awara we are sub optimal.

    The thing is, we have not yet reached the point where we feel comfortable without the buckets. They make us sleep at night, they make it possible for me to explain what we I do to my wife, to spend money guilt free on things. Yes, not optimal, and adding some enjoy now to life.

    After reading the article, i am not sure anymore what to think about it?

      1. Besides the numbers and theory, what did help you to make the switch?

        And how do you fit travel in a bucket free world? I would love to teavel a lot more. Yet, that would ruin my FIRE plans when I do not put a limit. Hence, a travel bucket.

        1. I never had too much cash sitting around. The big stock market crashes in 2001 and 2008/9 made me deploy all available cash because stocks were so wildly undervalued back then. That got me to think about how to run a tight cash management with only an absolute minimum of idle cash sitting around.
          Fast forward to 2016. Now stocks are no longer undervalued. But I don’t want to realize taxable capital gains to raise cash now. But since the big 30% run in 2013 I shifted a lot of new investments into alternative asset classes (option trading, real estate through private equity). There I still see good earnings potential and again I avoid much cash holdings.

          Also, we still budget. But the budget categories are mostly to monitor. We are not very rigid about that budget.

          We travel quite a bit now and the biggest constraint for us not so much the money but time (vacation days at work) and energy (traveling with a 2 year old toddler). Hopefully, in early retirement that will change. We definitely budget more for travel once in retirement. But not that much more. A plane ticket costs the same if you go 5 days or 5 weeks. And AirBNB is not that expensive. For the price of a nice hotel for 5 nights one can easily stay through AirBnB for 2 weeks.

          1. Maybe that is what plays in the back of my head: hoard cash for the next big correction. I just fear I will not have the guts to invest my money. I guess I need to set some rules on when and how much I buy ad stick to that. Otherwise, it might not happen. I could do that easiy for my ETFs and for some dividend stocks I have my eyes on.

            1. Yeah, that’s why I don’t try market timing. Be in the market all the time and solve all these headaches about “should I invest now? Should I wait for a correction?”
              As they say, “ignorance is bliss” and I’m ignorant about where the market goes and life is bliss.
              And by the way, you and I both follow the covered call writing style strategies, so even if we get the timing a little bit off, we still collect the option premium to cushion any fall 🙂

  2. I agree with you wholeheartedly.

    Budgets can set you up not only for overspending but also for underspending. Maximizing one’s utility (or whatever objective function one chooses to use) will, in general, not involve spending a fixed amount $x_i on expense category y_i each and every month. It’s okay to spend a ton of money on something, *so long as doing so is in line with your objective function.* Rigid budgets can be a useful tool, especially for people with spending problems, but (in general) they will not be optimal spending strategy. That being said, can anyone truly claim to be so rational as to be able to faithfully and continuously optimize their spending with respect to their objective function?

    I frequently see the unnecessarily restrictive bucket-based approach to portfolios when people attempt liability matching. E.g., an ultra-safe bucket (e.g. TIPS) to provide the desired retirement income in parallel with a risky aspirational bucket of equities. (I guess you can think of the emergency fund as some sort of attempt at liability matching, too.) There’s nothing wrong with liability matching, but the bucket-based approach causes you to ignore much of the set of possible investment strategies, just as those efficient frontier graphs illustrate.

    Btw, I did get your e-mail, I just haven’t had the chance to reply to it yet.

    1. Yes, very good point. Rigid budgets can also lead to underspending, say, not performing necessary car maintenance/repairs because that bucket is already exhausted for the quarter.

      Liability matching is another good example of mental accounting, as you point out. But if the liability matching is done right it doesn’t have to be. If the optimization is done globally over all assets and all liabilities (and not within each little bucket) that’s a useful exercise. You probably don’t have to keep 8-12% of the portfolio in cash for the next 2-3 years of retirement withdrawals. Mixing in more bonds and moving along the efficient frontier works just as well, without the return drag.

      OK, looking forward to you email reply! 🙂

  3. Very though-provoking. I had never heard about this bias before. We all waste that $20 funny money sometimes, but when this bias extends to large portfolios it gets really dangerous.

    1. Thanks for stopping by! Yeah, same here, guilty as charged: We do the same thing sometimes. Blow the $20 “funny money” but as long as we keep that to small amounts, we should be fine. We’re no Mr. Money Mustache, sorry. But at least for the big picture, the big ticket items, knowing about the Mental Accounting Bias helps you avoid the big blunders.

  4. The found $20 is a great example. When I was maybe 13, I was riding a foot powered scooter looking down, I found a $20 bill in the street after my friends rode right past it on their bikes.

    My friends expected me to blow it on pizza & candy (the 13-year olds’ version of hookers & blow) but I said “No. I’m putting it in the bank.” Maybe that’s why I don’t have friends. 😉

    Actually, I still see those guys every once in awhile, and we’re in a fantasy football league together nearly 30 years later. And I’ve bought them plenty of pizza.

    Applying an asset allocation across all accounts is something that can be hard to understand when you’re just getting started. Implementing a plan can be a little tedious, too, if you’re not used to working with a spreadsheet. But it’s the only sensible way to do it.


    1. Thanks Dr. for sharing that story. The $20 of found money is a great example, not just for Mental Accounting but overall good financial habits, most importantly valuing future benefits over instant gratification. Having the right attitude at an early age is an indicator for future success in real life. Seems to have worked out great for you. 🙂

  5. The hardest part about behavioral biases is that it’s hard to catch it because it’s something that’s happened out of habit or the external environment. However, once you catch it and know how to change it for the better, you can take steps to take it. It’s amazing how much richer that we can get just by changing our perspective. Not because of changing portfolio strategy or looking to save more, but just because of changing the way of thinking.

    1. Exactly! That’s a huge problem. Some of these behavioral biases are ingrained in humans after thousands or millions f years of evolution and they are hard to overcome. The only cure is to constantly remind yourself!

      Thanks for stopping by!!!

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.