A reasonable explanation for the upward tilt of US stock PE ratios

One of my whip-smart readers, Cathy Leow, wrote in a really interesting take to explain the gradual and undeniable upward tilt for the PE ratios of US stocks in recent years. Cathy was formerly an economist and then a grains trader, she’s been analyzing and investing in the equity markets since 2005.

I thought this was really smart and great food for thought, what do you think? Because if she’s right, it might require a rethinking of the where we think secular bear markets need to bottom and where they truly become historically expensive…

***

Josh – you asked a question of a fund manager during Fast Money lunchtime edition he couldn’t answer.  You asked “why should the S&P 500 trade at a premium to its long term historical valuation?”

I believe I have the answer.

We can’t look at long term historical valuations because the tax code has changed dramatically.  Prior to 2003 dividends were taxed at ordinary income levels, so for the highest two or three brackets fixed income was as valuable as the dividend at the same yield. Today, at the same yield, dividends are far more attractive since fixed income is taxed at a higher rate for the upper brackets. Additionally, prior to 1982, capital gains taxes were generally higher (as were the dividend tax rates because marginal tax rates were much higher).

Ergo, because of the tax code there should be a rerating higher of p/e levels over the past 30 years relative to prior to that time. And a further upward boost in 2003  – especially for dividend paying stocks.  Whether the level for the S&P 500 should be 17, 18 or more versus the long term historical average of 15 is something for the statisticians to debate.

Unfortunately there are other factors, such as earnings  growth and interest rates that also influence the multiple and pinpointing what the exact level should be is probably impossible. But 15 looks too cheap when taxes and current returns from fixed income are considered.

I had this epiphany a couple weeks ago while looking at a long term chart of historical p/e’s and noticed the upward bias over the past decade or so and was wondering why this happened.  The proverbial lightbulb went on in my brain when thinking about the changes  in how dividends have been taxed in more recent history.  Frankly I’m surprised this hasn’t gotten more attention.  And it also makes me wonder whether the hated utilities and other defensive sectors are really overpriced, especially as long as fixed income returns are so meager, and are likely to remain that way for awhile longer.

***

Thanks Cathy!

So? Is she onto something?