Do Stocks Dance to the Dollar's Tune?

S&P finds no correlation between movements in the greenback and the performance of the S&P 500 since 1990

by Sam Stovall From Standard & Poor's Equity Research

S&P 500 Foreign Sales

Month-End Trade-Weighted U.S. Dollar

S&P Benchmark Correlations with the U.S. Dollar: 1990-2007

Story Tools

• post a comment
• e-mail this story
• print this story
• order a reprint
• digg this
• save to del.icio.us

Year to date through the end of September, the dollar index, representing the value of the U.S. dollar vs. a trade-weighted basket of foreign currencies, closed near 77 (a discount of 23% to 100, or parity, with the basket), registering a 6-percentage-point fall from the 83 level at the end of 2006, and equal to the forecasted average value of the dollar for all of 2007, according to Standard & Poor's Chief Economist David Wyss. He predicts the dollar will fall further in the year ahead, averaging 71 in 2008 and bottoming at 68 in 2011.

For American tourists venturing abroad, that's not good news. For U.S. exporters, however, that's very good news. Indeed, Wyss expects U.S. exports to rise 9.5% in 2008, as compared with the projected gain of 7.3% in 2007. Exports should increase another 8.4% in 2009, says Wyss, which could aid earnings per share for U.S. multinationals that don't hedge their currency exposure.

Slight Influence

Specifically, S&P Index Services estimates that 45% of 2006 sales for companies in the S&P 500-stock index came from overseas (using the 236 member firms that provided global revenue breakdowns). This figure is up substantially from the 32% estimated in 2001. Companies in the S&P 500 Energy and Information Technology sectors showed the greatest overseas revenue exposure at 56% each, followed closely by the Consumer Staples group at 47%. The sectors with the least exposure included Financials at 30% and Consumer Discretionary at 32%. No data was available for companies in the S&P 500 Telecommunications Services group.

Investors might wonder whether the price changes of these standout sectors were helped or hurt by a falling dollar. In other words, have the S&P 500 and its sectors and subindustries showed any strong correlation with movements in the dollar? Since Dec. 31, 1989 (which is as far back as S&P 500 sector data extend), the monthly percentage change in the S&P 500 index posted a 0.03 correlation with the monthly change in the value of the trade-weighted dollar, indicating that the dollar and "500" had almost no correlation over this extended period.

Selecting five periods over the past 17 years of identifiable weakness in the dollar (6/91–8/92, 12/93–4/95, 7/98–10/98, 2/02–12/04, and 12/05–9/07), however, the correlation was more pronounced at –0.27, indicating that the market's rise during these periods was, at least partly, explained by the dollar's weakness, in our view.

Info tech surprise

In addition to the overall market showing an increase in correlation to the dollar during these selected periods, the S&P 500's growth and value components also saw more pronounced negative correlations, as each fell to or below –0.2 from a prior reading equal to or above zero. What's more, small-cap stocks (as seen in the Russell 2000 for 1990-94 and the S&P SmallCap 600 index since 1995), also showed rising negative correlations.

On a sector level, the results were fairly similar to those for the overall market. The correlation between monthly performances for S&P 500 sectors and the dollar since 1990 were highest for Financials and Consumer Discretionary (indicating a slightly positive correlation with the dollar—meaning as the dollar fell, so did they) and lowest for Energy. All other sector correlations appeared too small to be of consequence, in our opinion. Surprising, however, was Info Tech's relatively high correlation, in light of its large overseas revenue exposure.

During the selected periods, however, even though all 10 S&P 500 sectors registered negative correlations, the S&P 500 Energy, Utilities, Materials, Consumer Staples, and Industrials sectors showed the highest negative correlations, at –0.46, –0.34, –0.29, –0.21, and –0.21, respectively, while the Info Tech and Consumer Discretionary sectors registered the lowest correlations, at –0.07 and –0.08.

Conclusions

We think subindustry leadership corroborated sector-level patterns. Six of the eight S&P 500 subindustries with the highest negative correlations (Oil & Gas Drilling, Oil & Gas Exploration & Production, Oil & Gas Equipment & Services, Integrated Oil & Gas, Electric, and Gas Utilities) were from the Energy and Utilities sectors. Yet only 6 of 81 subindustries that participated in all five periods registered positive correlations, with three of these six coming from the S&P 500 Consumer Discretionary sector. Only three of the six had correlations that were above 0.10: Movies & Entertainment (0.14), Hotels & Cruise Lines (0.12), Office Services & Supplies (0.10).