PE by market capitalization

I've seen several recent discussions concerning large cap stocks seeming to be relative bargains at current market prices based on price/earning (PE) ratios. Always missing in these types of "talking head" discussions seems to be the data that might inform a reasonable decision.

So first I wanted to see the current relative PE of stocks with different market capitalizations - from the smallest of companies to the largest. Median PEs are shown (median is not the average)

I'm not sure how folks in the industry look at the stats so I've included median PE for all companies (both + and - earnings) and the median PE for only companies with positive earnings. Based on what I'm seeing the smaller company PEs don't really seem out of line with what I might expect.

Next to place the information in historical context I've plotted relative PE ratios over time, going back around 7 years. (Click the image to enlarge).

The green line is representative of the largest companies (over 30 Billion mktcap.). The red line represents the group of the next largest companies (10-30 Billion mktcap), and so forth...

From the graph it's clear that the largest companies (30B+) have seen a significant devaluation of earnings since 7 years ago, falling from a PE of nearly 25 to below 15 currently (40% decline in the value of earnings).

For other market caps it seems the valuation of earnings on a PE basis is fairly constant since about 2003. Even the large 10-30B stocks (red line) have had a fairly steady PE since 2003.

Conclusion: It does seem that the earnings of the largest cap stocks are at a lower valuation than those of stocks of other market caps. What's unclear to me is whether that's the way it should be. The benefit of the bigger companies is the predictability of earnings - in the past year 99% of stocks with $30B+ mkt cap had positive earnings. On the other hand the largest companies don't tend to grow as fast and probably don't deserve as high of a multiple as the historicals suggest.

I guess while I could see a slightly higher PE multiple assigned to the largest stocks, based upon historicals the relative valuations seem only modestly out of line in aggregate. Maybe if some more largely undervalued individual stocks in the largest market cap range could be identified I'd feel better about the statement, but as a group the case isn't compelling to me.

Overall I'm pretty bearish on the markets right now and am more inclined to limiting exposure (I'm over 50% cash in my accounts outside of my 401K), but short of that this might cause me to consider shifting a bit of my 401k more toward the S&P500 vs. the smaller Russell 2000 index where I have most of my 401K funds currently, but I'm not going to do anything drastic.