Kass: The Buck Stops Here

This column originally appeared on Real Money Pro at 9:08 a.m. EST on Feb. 27.

NEW YORK ( Real Money) -- Last week Hedgeye's lynx-eyed Keith McCullough opined on CNBC's "The Kudlow Report" that a strong U.S. dollar is a bullish catalyst for U.S. stocks. His comments piqued my interest in reviewing my analysis from last year on whether a strong U.S. dollar is bullish for U.S. stocks.

The results below, with help from Seabreeze Partners analysts Nick Pollari and Kelley Hopkins, are enlightening.

Bottom line: A rising U.S. dollar is not necessarily stock-market-friendly.

Utilizing monthly return data of the Fed's Trade-Weighted Real Broad Dollar Index and the S&P 500 from 1990 to 2012 we have determined that, statistically speaking, an appreciating US Dollar actually leads to reduced stock returns.

Note that we have used the trade-weighted dollar index (TWD) instead of the traditional U.S. dollar index (DXY) due to the heavy skew towards the euro in the DXY. Utilizing data from the DXY, however, will result in the same conclusion.

The data set has been broken into two time periods to provide insight into both the long term and the shorter term -- from January 1990 to December 2012 and from January 2010 to December 2012.

1990 to 2012; monthly return data; 276 observations:

  • The TWD and the S&P 500 have an R-squared of 0.0766 over this time period.
  • The TWD has a coefficient of -0.995, which implies that an appreciating dollar index reduces the return on the S&P 500 by 0.995% for every 1% increase in the TWD.
  • Utilizing the T-Statistic of the TWD, at the 95% confidence level, we have concluded that the coefficient is statistically significant.
  • Utilizing the F-Statistic of the model, at the 95% confidence level, we have concluded that the model in its entirety is statistically significant.

2010 to 2012; monthly return data; 36 observations:

  • The TWD and the S&P 500 have an R-squared of 0.225 over this time period.
  • The TWD has a coefficient of -1.814, which implies that an appreciating dollar index reduces the return on the S&P 500 by 1.814% for every 1% increase in the TWD.
  • Utilizing the T-Statistic of the TWD, at the 95% confidence level, we have concluded that the coefficient is statistically significant.
  • Utilizing the F-Statistic of the model, at the 95% confidence level, we have concluded that the model in its entirety is statistically significant.

Over time, we can visually decipher the relationship between the TWD and the S&P 500.

From 1990 through 1994 the S&P 500 moved independently of the TWD. Only during the time period of 1995 through 2001 can a positive correlation between these two indices be seen.

As the U.S. economy experienced a recession in the early 2000s, the Fed began cutting the federal funds rate. The Fed cut this rate from over 6% at the end of 2000 to the unprecedented level of 1% over time. As this stimulus worked its way through the economy, we began to see the S&P 500 appreciate, culminating in a new era that began in 2003.

Following 2003, we can see a very clear inverse relationship between the S&P 500 and the TWD, and this relationship holds until this day.
At the time of publication, Kass and/or his funds were short SPY, although holdings can change at any time.

Doug Kass is the president of Seabreeze Partners Management Inc. Under no circumstances does this information represent a recommendation to buy, sell or hold any security.

If you liked this article you might like

Market Is on the Straight and (Very) Narrow

Look Back to Go Forward

Stock Observations; Reviewing Equities: Doug Kass' Views

Even North Korea's Kim Jong Un Can't Stop This Epic S&P 500 Stock Rally

Robots Might Be Biggest Obstacle for Stock Market Bears