How much is rebalancing worth?

By Felix Salmon
May 18, 2009

A couple of weeks ago, Lance Knobel asked me what I thought of MarketRiders after reading Erick Schonfeld’s write-up. I kept on meaning to get around to it — you know how these things are — but before I could, Ron Lieber beat me to the punch. Ron says he likes the service, which basically keeps track of your ETF portfolio and tells you when it looks like it’s in need of a rebalancing.

I agree that it’s a useful service, but I’m not sure that it’s really worth $120 a year. For most of the people who will be using this service, the base-case scenario is that they simply buy up a bunch of index funds and then forget about them until the next point at which they have some money to invest, when they’ll probably throw that money into whatever pot looks most underweight at the time. Call it poor man’s rebalancing.

So the question is, how much value does active rebalancing add? I asked this a year ago, when three ETFs were launched which have a built-in rebalancing function, for which they charge 25bp per year. That seemed steep to me — but if you have less than $50,000 to invest, then you’d actually be better off paying 25bp a year than you would paying $9.99 a month.

I certainly don’t think that the rebalancing advice is worth more than 25bp a year, which means that I wouldn’t recommend MarketRiders to anybody with less than $50,000 in investable assets. But more generally I’d love to see some empirical data on the value of rebalancing, in basis points per year. Unless and until it becomes clear how much value it adds, I’d be hesitant to pay good money for the service.

More From Felix Salmon
Post Felix
The Piketty pessimist
The most expensive lottery ticket in the world
The problems of HFT, Joe Stiglitz edition
Private equity math, Nuveen edition
Five explanations for Greece’s bond yield
Comments
7 comments so far

Watson Wyatt and Mercer have done a lot of research into the benefits of rebalancing for institutional pension funds, although I confess I can’t recall the conclusions – possibly not entirely clear cut then.

Posted by James | Report as abusive

The main benefit of rebalancing is risk control – you don’t find yourself taking much more (or less) equity risk than you want.

Rebalancing should not be viewed as adding to returns. Over the long run, stocks outperform bonds. Rebalancing out of stocks hurts returns. The main benefit for returns of rebalancing is when you have two assets with similar performance and imperfect correlation (for example, US and non-US equities over some time periods). Rebalancing can improve returns in this case.

Posted by fusion | Report as abusive

I seem to remember (1) studies cited at the Bogleheads forum suggesting that rebalancing 1/yr is about optimum, and (2) and that rebalancing never was a close second.

Dear Felix, it really depends on when you use the model. For example 50% Aust. Accumulation Index 50% Cash from 1970 to 1990 re-balanced at 30 June each year paid slightly under 2% better than shares on their own which, over that period is a serious boatload of money. Of course re-balancing on 30 September brought even better results. Over the last 24 years, however, the share market easily outran the balanced portfolio until quite recently.

Posted by Andrew Taylor | Report as abusive

The more diversified the portfolio, i.e., the lowest aggregate correlations across each pair of assets, the greater the reward for rebalancing. Annual rebalancing should add 100 to 150 basis points of return for a broadly diversified global multi-asset portfolio, but substantially less for just a stock/bond mix.

Posted by Scott Berglund | Report as abusive

Felix:

Thanks for your post. In fact, we are real believers in rebalancing and we have moved the “state of the art” forward. For example, today’s conversation about rebalancing involves “time-based” rebalancing, i.e. when to rebalance — quarterly, yearly, monthly. We use “event-based” rebalancing by using algorithms to measure how out of balance an ETF, an asset class or an overall portfolio is. When you do this, you get better results from rebalancing because you trigger a rebalancing event at the point in time where it should occur. For example, had someone done “time-based” rebalancing on 9/30/08, and then again on 12/31/08, they would not have gotten the same returns as someone who rebalanced when the market was bottoming in November which is where our system sent alerts.

Our partner Ryan Pfenninger is currently working on a new part of http://www.marketriders.com where one can look at the effects of this on different portfolios.

When David Swensen of Yale writes about rebalancing, its “event-based”, not “time-based” and we believe this is where the community needs to focus the conversation.

Mitch Tuchman, CEO
http://www.marketriders.com

Dear Felix,

You mention that an investor with less than $50K investable dollars should avoid the $100 annual MarketRiders’ fee (annual payment provides a discount on the $9.99 monthly subscription). This insight is spot on and is exactly what we state in our FAQs (http://www.marketriders.com/faq#10) under What Is The Minimum Size of a Portfolio. We provide a direct link to a target date fund which we believe is another good solution for such investors.

Thanks,

Stephen Beck
Co-Founder
http://www.marketriders.com

Post Your Comment

We welcome comments that advance the story through relevant opinion, anecdotes, links and data. If you see a comment that you believe is irrelevant or inappropriate, you can flag it to our editors by using the report abuse links. Views expressed in the comments do not represent those of Reuters. For more information on our comment policy, see http://blogs.reuters.com/fulldisclosure/2010/09/27/toward-a-more-thoughtful-conversation-on-stories/