http://themacrotourist.com/images/2017/01/GSFeb0817.png

Years ago I stumbled upon this interesting stock market indicator created by famed strategist Ed Yardeni. I was intrigued because contrary to most equity models, Yardeni’s formula did not include the traditional inputs of earnings or interest rates.

Convinced it was just some sort of statistical fluke, I modeled it up in my Bloomberg and have been following diligently ever since. Every time Ed’s indicator has diverged from the S&P 500 I think to myself the time for the model to stop working has finally come. Yet, strangely, almost as soon as that thought crosses my mind, the series recouple.

Have a look at the indicator’s performance over the past couple of decades:

http://themacrotourist.com/images/2017/01/YardeniFeb0817.png

The relationship is far from perfect, but look closely at the direction of the moves of both series. They track fairly closely.

And unlike most models that failed to confirm the recent move to new highs on Trump’s surprise election win, the Yardeni “fundamental” indicator actually rose along with the S&P 500.

http://themacrotourist.com/images/2017/01/FiveFeb0817.png

At this point you are most likely saying, “so what? There is probably some sort of circular logic where one of the variables is dependent on the output.”

But that’s what’s so interesting. Yardeni’s “fundamental” indicator has the following inputs:

  • the four week average of initial jobless claims
  • the CRB raw industrial commodity index
  • Conference Board Consumer Confidence

That’s it. No EPS estimate. No GDP forecast. No interest rate component.

I know raw industrials often rise with economic activity, and jobless claims are highly correlated to interest rates, and consumer confidence is reflection of how well the economy is doing. But who would have thought you could create a model that follows the S&P 500 so closely from these three inputs? Hats off to the good Doctor Yardeni.

http://themacrotourist.com/images/2017/01/YardFeb0817.jpg

I am not using the model to justify the recent stock market rally. I don’t think the model serves much predictive value.

But it is valuable to understand what would need to happen for the stock market to suffer a large decline (assuming the Yardeni model continues to be valid). Or conversely, what needs to happen to have the current rally extend even higher?

Let’s go through the three inputs one by one.

First up, the four week average of initial jobless claims:

http://themacrotourist.com/images/2017/01/InitialFeb0817.png

Tough to figure how this will go much lower, so if anything the surprise will probably be the other way.

How about the CRB raw industrial commodity index?

http://themacrotourist.com/images/2017/01/RawFeb0817.png

This index has had a decent little run, but is still miles below its 2011 high.

And finally, what about consumer confidence?

http://themacrotourist.com/images/2017/01/ConfidenceFeb0817.png

Consumer confidence has been on a steady march higher since the terrible collapse in 2008, with the recent Trumphoria pushing this index up to a new high.

I don’t have any answers, but it seems to me that it will be more difficult to get this Yardeni “fundamental” indicator higher than lower. Jobless claims are bouncing along the lows while consumer confidence is sitting at highs. Both of those look like the surprise will be from bad news, not the other way round.

Raw industrial commodities are the Obi-wan Kenobi of this indicator. A rally in industrial commodities from a growing global economy is probably the only hope. But I worry. Obi-Wan’s ability to save the galaxy from the evil Siths was never his greatest strength. In fact, let’s face it, Obi-Wan never got much right…

http://themacrotourist.com/images/2017/01/StarWarsFeb0817.jpg

Thanks for reading,
Kevin Muir
the MacroTourist