Objective research to aid investing decisions

Value Investing Strategy (Strategy Overview)

Allocations for February 2020 (Final)
Cash TLT LQD SPY

Momentum Investing Strategy (Strategy Overview)

Allocations for February 2020 (Final)
1st ETF 2nd ETF 3rd ETF

ISM PMI and Stock Market Returns

| | Posted in: Economic Indicators

According to the Institute for Supply Management (ISM), their Manufacturing Report On Business, published since 1931, “is considered by many economists to be the most reliable near-term economic barometer available.” The manufacturing summary component of this report is the Purchasing Managers’ Index (PMI), aggregating monthly inputs from purchasing and supply executives across the U.S. regarding new orders, production, employment, deliveries and inventories. ISM releases PMI for a month at the beginning of the following month. Does PMI predict stock market returns? Using monthly seasonally adjusted PMI data during January 1950 through January 2016 from the Federal Reserve Bank of St. Louis (discontinued and removed) and from press releases thereafter through December 2018, and contemporaneous monthly S&P 500 Index closes (828 months), we find that:

The following chart tracks monthly behaviors of PMI and same-month closing levels of the S&P 500 Index (logarithmic scale) over the full sample period. Average PMI is 53.2, with trend (dashed line) modestly downward over time. While the two series appear to move together much of the time, visual inspection is not helpful in determining whether one series leads the other.

For precision, we relate S&P 500 Index monthly return to level of PMI.

The following scatter plot relates next-month S&P 500 Index return to monthly PMI over the full sample period. The Pearson correlation for the two series is -0.07, and the R-squared statistic is 0.004. While there is a very slight negative relationship, level of PMI explains hardly any of the variation in next-month stock market return.

For greater sensitivity, we relate next-month S&P 500 Index return to monthly change in PMI.

The next scatter plot relates next-month S&P 500 Index return to monthly change in PMI over the full sample period. The Pearson correlation for these two series is 0.00, and the R-squared statistic is 0.000. Monthly changes in PMI offer no information about next-month stock market returns.

To check for meaningful non-linearity, we calculated average stock market return by range of monthly changes in PMI.

The next chart summarizes average next-month S&P 500 Index returns by ranked fifth (quintile) of changes in PMI since 1950 (165 observations per quintile) and since 1990 (69 observations per quintile). Average S&P 500 Index monthly return for all months since 1950 (1990) is 0.69% (0.67%). Lack of systematic progressions and inconsistencies across sample periods undermine belief in a reliable relationship.

Might PMI or change in PMI systematically lead or lag stock market returns at intervals longer than a month?

The next chart relates monthly PMI and change in PMI to S&P 500 Index return for various lead-lag relationships, ranging from stocks lead PMI/change in PMI by 12 months (-12) to PMI/change in PMI leads stocks by 12 months (12). Results suggest that:

  • Stock returns lead PMI with a cumulative positive relationship (higher stock returns indicate higher future PMI) over the intermediate and long terms.
  • PMI leads stocks with a weak negative relationship (higher PMI weakly indicates lower future stock market returns) over the short term and a weak cumulative positive relationship over the longer term.
  • Stock returns lead changes in PMI with a positive relationship by about one to three months.
  • Changes in PMI have practically no relationship with future stock market returns.

A recent subsample since 1990 generally confirms the persistence of the relationship between stock market returns and future changes in PMI, but with weak indication of a positive relationship between change in PMI and stock market returns over the next two months.

Because of the long-term relationship between PMI and stock market return, we look at annual correlations.

The final chart relates annual PMI/change in PMI to S&P 500 Index return for various lead-lag relationships, ranging from stocks lead PMI/change in PMI by four years (-4) to PMI/change in PMI leads stocks by four years (4). Results suggest:

  • A relatively high (low) PMI may indicate relatively weak (strong) stock market returns over the next three years.
  • Annual change in PMI is not interestingly related to stock market returns in future years.

In summary, evidence from simple tests offers little support belief that ISM’s PMI is useful for predicting U.S. stock market returns.

Cautions regarding findings include:

  • ISM’s retroactive annual adjustments of PMI levels to recalibrate seasonality in pre-February 2016 data incorporates look-ahead bias that interferes with backtests of PMI power to predict stock market returns. Data for February 2016 and later are as-released.
  • Analyses are in-sample. An investor operating in real time during the sample period may draw different conclusions at different times. In-sample findings are generally too weak to motivate out-of-sample testing.
  • Findings do not rule out the possibility that surprises in PMI, relative to some measurable expectation, more usefully predict stock market returns.
  • The subsample since 1990 is small in terms of number of economic cycles.
Login
Daily Email Updates
Filter Research
  • Research Categories (select one or more)