The battle between value and momentum wages on.

Last week, there was massive reversal in this year’s trend, which has been for momentum (and low vol) to outperform value (and size). Last week and today look to be a miniature “quantageddon” (a term Goldman coined after a large reversal in the value/momentum spread in early 2016), because the worst performer last week was the hedge fund crowding factor… we shall see how the rest of this week carries out.

Funny enough, just last week I wrote about trend finally catching good performance but that correlations were nearing a near-term bottom with Carry. The sustained rally in trend (going on since March of this year) has been abruptly put to a halt.

I mentioned I would discuss how the term “smart beta” may have affected correlations in factors. This research is still ongoing, though I will put some initial thoughts to paper:

  1. Momentum’s correlation with other factors is usually a good way to see if “crowding” exists
  2. Google’s search term frequency for “smart beta” may be a good proxy for the popularity of the term
  3. There doesn’t appear to be any major form of crowding, using these metrics.

Some rolling 1-year correlations for you:

Image 4
“Stuck” in negative territory
Image 5
No large rise in size over the last 5/6 years. Notice how rises in correlations between momentum and a factor usually spike upwards and melt downwards
Image 6
Quality, anybody?
Image 7
One might be surprised to see these numbers so low

A normalized frequency plot for search terms associated with “Smart Beta”, from Google:

Image 8

So, the rise in search terms for smart beta really began in earnest in 2013/2014. Thoughts ongoing…

Last Week’s Returns

Image 3

Factor Returns: Year-to-date

2019 09 09 Markets Ytd
2019 09 09 Equity Styles Us Ytd
2019 09 09 Equity Styles Global Ytd
2019 09 09 Alt Factors Ytd

Factor Returns: Trailing

2019 09 09 Trailing

Notes to performance

  1. Past performance does not guarantee future returns
  2. The factor returns may not correspond to the factor returns inside the factorE application.
  3. The equity styles are market neutral, meaning these returns are on top of equity market returns
  4. A csv download will be made available for those who want to show trailing returns in another format