Factor Premium/Discount & Market Timing

twitterlinkedinrssmail

Smart beta is the new new thing, I even wrote about it here. There has been some interesting writing on it recently that shares some concerns with the popularity of this approach which I will attempt to summarize: Factor investing can work but it can become relatively ‘cheap’ and ‘expensive’ due to performance chasing and affect future returns.

Here is a chart of relative returns of the Value Factor:

Great visual explanation of the time varying returns of the value factor versus investing in the market. If I am an allocator then I  can simply look at the chart and overweight value below -2000 Bps and underweight above 2000 Bps and now I am adding value! Management fee earned.

Here is another chart of rolling returns:

 

Looks pretty similar doesn’t it? Nice stationary pattern, might even add an RSI indicator to it? I don’t mean to say it’s a bad idea to look at relative premium when applying a factor approach but it certainly looks to be adding an element of timing. And if historically the industry is bad at market timing, why would they excel at factor based timing?

While this isn’t a defense of performance chasing behaviour on the part of allocators, but if it’s an inevitable artifact of the process then how can one excel at performance chasing? A momentum based approach with quick stop losses and pyramid allocation to winners might work? Liquidity is obviously a concern with that approach but perhaps this explains the inflows to multi-manager firms as they can divide the capital into smaller and more manageable momentum positions.

If at the end of the day, we continue to come back to the similar conclusion that our greatest edge becomes duration, moderate diversification, strategic discipline, and low cost then what do we need all this engineering for again?

 

 

 

 

twittergoogle_plusredditpinterestlinkedinmail

Be the first to comment on "Factor Premium/Discount & Market Timing"

Leave a comment

Your email address will not be published.


*