Factor Timing: Sin Less Than a Little

By Jack Forehand (@practicalquant) —

There is perhaps no issue that is the subject of more debate in the factor investing community than factor timing. We are all trained to buy low and sell high, and it is tempting to conclude that we can do the same thing with factors. For example, if value stocks have struggled for a long period of time and they are cheap relative to their history, it makes intuitive sense to add exposure to them and to try to take advantage of mean reversion when they bounce back. It also makes sense in theory to reduce exposure to a factor after a long run of positive relative performance. As is the case with many things in investing, though, the actual implementation of that concept proves much more difficult in the real world than it seems in theory.

This is a very difficult topic to write about because there really is no right answer. Whenever Rob Arnott of Research Affiliates and Cliff Asness of AQR, two widely followed and hugely successful investors in the field of quantitative investing, disagree on a topic, as they do on this one, you can safely assume it is very complex and that the answer is not clear.

In his 2016 paper “Timing “Smart Beta” Strategies? Of Course! Buy Low, Sell High!” , Arnott looked at factor timing using both the best and worst performing factors and compared the result to a portfolio that is equal weighted across all the factors. The trend-chasing and contrarian portfolios he built each held the top 3 factors that either performed best or worst using an average of four historical periods. The equal weighted portfolio held an equal exposure to all eight factors they follow. As you would probably expect, the strategy of chasing the hot factor did not work. But the contrarian strategy of buying the out of favor factor did produce more alpha than the equally weighted approach, but that result did come with higher risk due to less factor diversification, so the risk adjusted return was lower than the equally weighted portfolio. This result offers evidence that factor timing is possible, but also supports the thesis that it is very difficult to do in practice.

https://www.researchaffiliates.com/en_us/publications/articles/541_timing_smart_beta_strategies_of_course_buy_low_sell_high.html

Cliff Asness countered that paper with one of his own titled “Contrarian Factor Timing is Deceptively Difficult”. He summarized the paper in a separate post this way:

In our latest paper, we again show that factor timing is likely even harder than market timing. First, the long-short factors in question have higher turnover than the market, making long-run predictability, the possible savior of market timing, a much dodgier proposition for factor timing. Second, unlike timing the market, timing factors using just valuation must contend with contrarian factor timing being already implicitly (but strongly) present in the value factor itself. Despite the rhetoric calling factor timing simple common sense, it isn’t at all obvious that one should value-time factors, at least not to any significant extent (and if one is saying to do just a tiny bit, then let’s not argue about the insignificant!), and certainly not while they’re currently within historical bounds as they are today.

So what should an average investor do when two of the best minds in factor investing disagree on a topic? I think the best thing to do is probably to steer clear of it. If the best investors can’t agree on if something works, the odds of a regular investor successfully implementing it in their portfolio are very low. Having said that, I do tend to agree with Rob Arnott that when done properly, factor timing can enhance returns in a portfolio. For those who do try to implement factor timing, I think there are a few rules of thumb that can help.

You Will Be Early and Your Emotions Will Likely Get the Best of You.

It is one thing to say that when investing styles like value go out of favor, they always come back. It is another to take advantage of that in a real-world investing strategy. Investing factors can go out of favor for very long periods of time and timing their exact turn is impossible. If you are a believer in factor timing, you have likely been adding exposure to value throughout its most recent decade of underperformance. That has only hurt you so far as value has continued to underperform. You will likely eventually be proven right and the excess returns you will get at that point may very well exceed the losses you incurred by being early, but getting to that point will take a level of discipline that most people just don’t have. As Cliff Asness said in the quote above, timing the market is hard, but timing factors is harder. Only those who are willing to be wrong for long periods of time should even attempt it.

Sin Less Than a Little

AQR’s famous paper of market timing concluded that investors could “sin a little” when it comes to timing the market. Given that factor timing is harder than market timing, it makes sense to sin even less than that. In other words, make small moves over time with the understanding that you will likely be wrong for a while before you are right. Investors tend to want to make binary decisions and go “all in” with a factor that is out of favor. That is a recipe for a disaster when it comes to factor timing. If you had gone all in on value years ago, there is little chance you would have been able to stay the course long enough to see its eventual return to prominence. Small, incremental changes offer a much better chance of success.

Have a System

Just like with all aspects of investing, your emotions can be a killer with trying to time factors. If you are going to do it, you want to have a disciplined system that you follow without deviation. As with all timing decisions, you have to get both the buy and sell right to be successful, and that is very hard to do. For example, if you were in momentum stocks through 2017 and early 2018, you did very well. But then it all fell apart in late 2018. Figuring out whether that change is a short-term blip or a long-term shift is very difficult and if you don’t get it right, you can easily lose any gains you got by shifting into momentum in the first place. A disciplined system removes the risk of biases tainting your decisions.

One common way to time factors is to use factor valuation spreads. Spreads measure the difference between the most expensive stocks using the factor and the least expensive ones. It is tempting to want to add exposure to a factor when it is underperforming, but sometimes that underperformance is justified. For example, if value stocks are struggling, but that struggle is due to deteriorating fundamentals, then it might not be a good idea to add exposure to them. Factor spreads would help you pick up on that. In addition to spreads, some investors use things like where we are in the economic cycle, or momentum to select factors. We haven’t seen much success with those, but others have seen some positive results. Whatever you do, though, it is very important to have a disciplined system.

Understand that Sometimes This Time is Different

It is commonly said that “this time is different” is the most dangerous phrase in investing. Most of the time, this time is in fact not different and mean reversion usually wins out. But “this time is never different” can also be dangerous. Sometimes investing strategies that have worked for long periods of time stop working. For example, many believe that in a world dominated by technology and service companies with significant intangible assets, the price/book ratio is no longer a valid measure of value. That debate is too involved to get into here, but if they are correct, trying to factor time the price/book ratio is going to be a disaster because as it struggles you will continue adding exposure to something that no longer works. And given how long it takes to definitively say that a factor no longer works, by the time you figure it out, it will be too late. So if you are going to try to time factors, you need to be confident you aren’t betting on one that is permanently impaired.

In the end, I certainly won’t be the one who settles the factor timing debate. Much smarter people than me disagree of the topic and I can’t add anything to it beyond what they have. If there is one area everyone agrees on, though, it is that factor timing is extremely difficult. That is reason enough for most investors to steer clear of even attempting it. So unless you have an iron will and the ability to be wrong for long periods of time before you are right, there may be better places to look for market outperformance than factor timing.