After having spent maybe too much time reading and studying financial markets, I thought I have seen it all. At least in terms of simple investing strategies that were not published yesterday.
Value Averaging is a strategy proposed by professor Edleson in 1991 and I discovered it by chance last week. You never stop learning. The issue with spending too much time reading and studying financial markets is that 9 times out of 10 you find strategy flaws faster than FinTwin dunking on Jim Cramer these days.
What I am going to do is really stupid. Instead of thoroughly documenting myself I will let my gut spit whatever reaction I had. Do not try this at home. Like trying a trick on my snowboard for the first time, I am pretty sure I can make it, the worst-case scenario I’ll see people around me laughing (but no broken bones this time).
What is Value Averaging
From Investopedia:
Value averaging (VA) is an investing strategy that works like dollar-cost averaging (DCA) in terms of making steady monthly contributions but differs in its approach to the amount of each monthly contribution. In value averaging, the investor sets a target growth rate or an amount of their asset base or portfolio each month and then adjusts the next month’s contribution according to the relative gain or shortfall made on the original asset base. Therefore, instead of investing a set amount each period, a VA strategy makes investments based on the total size of the portfolio at each interval.
Here is the Wiki page:
For example, an investor may want to have a $3600 investment in 36 months. Using VA, the investor would aim to have a total investment value of $100 at the beginning of the first month, $200 at the beginning of the second month, and so on. Having invested $100 at the beginning of the first month, the investment may be worth $101 at the end of that month. In that case, the investor invests a further $99 to reach the second month’s objective of $200. If at the end of the first month, the investment is worth $205, the investor withdraws $5.
The strategy is conceptually easy but a bit too complicated (and boring) to explain for a Sunday afternoon, if not 100% clear please refer to the above links 🙂
The (non) issue of investing at All-Time Highs
I would bet that 97% of the appeal of this strategy comes from investors’ fear of buying at the top. No one wants to be the idiot who bought just before a market crash, so Edleson is selling you a system to avoid that AND outperforms Dollar Cost Averaging. That sounds like a great deal!
But you are investing in an asset, or a portfolio, that is supposed to go up, that has a positive expected return. This asset will therefore spend a lot of time at all-time highs because…that’s what you are looking for, no? Trying to avoid investing in those moments is not productive.
Nick Maggiuli had a great post about it here, including this graph:
As you can see, each crash/bear market starts from an ATH. But not all ATHs are followed by a crash, quite the opposite. Nick’s post includes also a link to a post by Meb Faber on the same topic, which is almost a classic by now. Both Nick and Meb demonstrate that there is nothing wrong with investing near ATH, quite the opposite. The fear is just in our heads.
Valuations
Another investor’s fear is to invest in a bubble when valuations are high. The problem with this idea is that it works really well in theory but quite poorly in practice. Did you ever see a Google-search blank result page? Try to search for “GMO praises”…poor guys (this might be a joke only for insiders, GMO is a very good asset manager that bases its investment choices on a 7-year valuation model; this means they did not touch any US stock, in particular growth stocks, since ages).
Let’s go back to Nick. A few days ago he Twitted this:
If you read the replies to this, you will see how varied is each person’s idea of ‘fair value’. And Nick’s followers are the best minds out there (less me). Building a market-timing system based on valuations is a fool’s game.
I am a fool too. My stock portfolio is not 100% S&P500, the ‘standard’ benchmark out there however conceptually wrong that is. I have tilts towards the World Index, Emerging Markets and Value, all positions that have lower valuations than the S&P500. God knows how much those tilts cost me in the last five years. Yes, I hope they are going to pay me back in the future but at the moment, I only lost potential gains.
The stock market will eventually mean-revert to a lower valuation, the issue is that no one knows where that lower valuation will be. The (partially) good aspect of this strategy is that, at least, it has a defined plan to put you back into the market.
The Lump Sum Dilemma
All the above can be associated with the more generic ‘lump sum dilemma’, i.e. what’s the best strategy to invest a lump sum a person receives after an event (bonus, inheritance, etc). This obviously depends on the person’s objective and time horizon.
If you invest in stocks, your time horizon has to be long. If you invest in stocks, the strategy that maximizes your expected gains is to invest all the sum at once, as soon as you can. This is because stocks have a positive expected return, so the sooner you put your money to work, the better. High returns from stocks come from high volatility; in fact, stocks in the past have spent decades in between ATHs. This means that, from time to time, an investor going all-in on a particular day might spend the following ten or more years cursing against their poor judgment. Any strategy that diverges from going all-in day 1 is a way to manage this scenario. The investor is minimizing regret, they are not maximizing returns. This is a very important difference that should be front and center for anyone.
Internal Rate of Return (IRR)
Edleson demonstrates that VA is better than DCA by comparing the IRR of the two strategies. In doing so, he uses a trick that is widely loved by Private Equity and Venture Capital funds: ignoring capital held on the sideline.
Let’s say you want to invest £1m in a PE or a VC fund. At inception, you commit £1m to the investment but you do not transfer the whole sum to the manager. The manager has a multi-year window during which they can call amounts from you until they reach the total of £1m. This is because the manager only needs the funds when he finds the right investment opportunity; a VC cannot invest in a start-up at will, they have to wait for the start-up to open a round of financing and be accepted as part of it.
So the PE/VC manager will calculate their performance, typically as an IRR, based on when they called each tranche of funds, not since the date the investor committed the whole amount. While this is might be seen as technically correct, the investor can do whatever they want with the funds before they are called, it is just a way to juice the fund performance. The investor can park the funds only in very liquid investments, that typically pay low low yields (for the retail investor out there, this is why you cannot consider Bondora Go&Grow as a liquidity investment…because its liquidity is driven by Bondora, not by the investor needs). For the investor, the return on cash parked on the side-line contributes to the fund, seen as an investment, overall return.
When Edleson calculates VA IRR, he ignores as well the return on cash kept on the side-line. If included, VA IRR would be lower than DCA. This is also quite intuitive to understand: stocks go up (again), so the higher the amount invested, the better the investor return.
Cash on the side-line
From a pure behavioral point of view, telling someone “hey, this month your investment pot generated enough returns that you have to contribute 0”, can be fine in the short term but might lead to perverse incentives in the long run. According to the strategy, the amount you do not invest is supposed to go into a ‘reserve’ and not stay in your pocket. But:
- after several months of growing the reserve, someone might be tempted to use those funds for something else. I mean, the strategy worked so well in the recent past, why not finally adding that pool in your garden your kids are asking you since so long?
- FOMO: all your friends are making millions having exposure to the market, what are you doing with all that cash sitting there???
- *vomit emoticon*: you are supposed to use your reserve and buy stocks when everyone else is shouting in your face that the world is going to end. The market lost 30% last year and you are supposed to catch that falling knife.
You cannot even automatize this rule. From a personal finance point of view, I think what you get in ‘market-timing peace of mind’ you more than lose it for the need of added discipline; the investor has to calculate their portfolio performance each month, while with DCA it does not matter.
When the investor portfolio starts to be relevant, in the million region, the cash balance can be higher than the insured amount in case the bank holding it goes bust. Would the investor remember it? A fringe case but…unnecessary risk?
Another risk scenario is when the market drops so violently that the investor does not have all the necessary funds to bring the overall portfolio back to the set amount for that month. If the investor has to contribute more than their set monthly saving rate, it would be behaviourally challenging (if not materially impossible) to dump even more money in the stock market while it crashes. It reminds me of the Martingale System, which assumes access to an unlimited supply of money, albeit with less risk of going broke.
The initial assumption on your portfolio growth rate
If you want to use the Value Averaging system, you have to set a target amount for your portfolio, like $1 million in 20 years. Or you can set a target yield for your investments, like 7% a year. They are different ways to express the same thing.
This present at least two issues:
- if you are a non-pro, choosing the right rate can be unnecessarily stressful. Should you take the historical rate? But for which period, 10, 20, 100 years? Real or nominal? What I read in that article was real or nominal? What if inflation explodes and stay persistently high? What if in five years we will have deflation instead? What is the correct inflation rate to use? Where do I get it? I can go on all day.
- Why should you cap your gains? I want to retire with at least £1 million, not exactly £1m. What if the market grows at a higher rate than I was expecting, why should I not compound at a higher rate? You want to get as much exposure to positive convexity as possible. What if you reach your target as planned after 30 years but I tell you could have done it in 20, given the particular market environment?
I feel confusion
I did not read Edleson’s original paper because I did not find it for free. On the Wiki page, they say he was suggesting a period of three years but then I see others that associate this as a retirement strategy. This is actually a crucial point.
If you use it for retirement, at a certain point your pot will be so big that its movements will be driven by the market and not by your contributions. But the point of this strategy is that every month the portfolio will move by the same amount, i.e. contributions+market performance. So either when you start you set your monthly contributions to a monster number or your target pot is relatively small. Try it in Excel and you will see what I mean: if you want to reach £1m in 30 years, you need to employ for this strategy a bit less than £2800/month. Good luck saving that amount in the first year.
Conversely, you can use the strategy in three-year intervals…but this has its own set of issues. The biggest of which is that markets do not mean-revert in such a short span. Markets actually tend to trend. Why use a strategy that is meant to exploit mean reversion for periods so short? I do not understand.
Going back to the growth rate point, are you then supposed to use the same rate irrespectively if stocks are in a bull or a bear market? Edleson apparently used 16% as a growth rate, which sounds already a push in bull markets…
Conclusion
I think that the best scenario where to consider VA is if you have to invest a lump sum and you are scared about deploying it at once. In that case, choosing a growth rate is quite simple because you already have all the capital to employ at inception and have to simply divide that amount by your preferred investing horizon.
Other than that, it is definitely more complicated than DCA and I do not think the additional complications (and trading costs) justify the effort. DCA into a 60/40 portfolio mimics the same market-timing effect of VA: if stocks go up, you buy more bonds; if stocks go down, you buy more stocks. Portfolio rebalancing is the part of the strategy that systematically pushes you to buy low and sell high.
Would you agree? Did you have a different experience with this strategy to share? Write it in the comments!
What I am reading now:
Follow me on Twitter @nprotasoni