Don't Throw in the Towel on Recovery Just Yet

Every month the ISM (Institute of Supply Management) releases its survey of purchasing managers in the manufacturing sector. Today's PMI number was below expectations, clocking in at 52.6 versus 52.9 the prior month and expectations of something closer to 54. This miss is at least partially responsible for the market hitting the panic button today.

Still, given that the market has come up so fast and so far (almost 60% in just over six months) some retrenchment was probably in the cards, with or without the PMI number. In fact, things really aren't so dire, a number over 50 indicates that more people believe things are getting better than getting worse. A number sustained in this range typically correlates with GDP growth of over 3%.

The PMI is a wonderful and very reliable early indicator of the end of a recession. As I have mentioned many times, this indicator tends to bottom four or five months prior to the recession, which is exactly what happened this time, as I believe the end of the recession will be pegged in either May or June based on the big coincident indicators (retail sales, employment, industrial production, and incomes). This would be about five months after the index bottomed at 32.9 in December.

So after a move from 32.9 to 52.9 over eight months, a one-month setback of 0.3 doesn't seem to be a reason to throw in the towel on an economic recovery just yet. Also, as I have warned in past pieces, the PMI is not as good an indicator on the way down. For example, between the 1990 and 2001 recessions, the indicator moved down sharply and back up three times over those 11 years before correctly predicting another recession on its fourth dip.

The temporary setback in the PMI number today is very typical of the past 10 recessions. On average the PMI index experiences a reversal in its upward trend about four or five months after the end of the recession. In eight out of 10 recessions, that trend resumed its upward movement in just one month, and in three months during the other two recessions. So we won't have to wait long to see if the number is a fluke or the start of a double dip.

I'd also note that most, but not all, recessions are proceeded by at least one reading that is below 50. Now if the trend is down for two more months and one of those readings is below 50, I will need to rethink my optimism.

See More Articles by Robert Johnson