Does this look like a recession?

Hypothesis: the U.S. economy entered into a recession in December 2007. Question: does the payroll employment data in the four months succeeding that date look like the payroll employment data four months into preceding recessions? The graph below shows the most recent payroll employment data for the last four recessions and the current possible recession, from six months before the NBER peak to six months after. Period 0 is the NBER date for that recession. All the series are normalized to equal one at the NBER peak date.

Answer: four months following the start of the last four recessions, payroll employment was around half a percentage point below its value at the NBER peak. Currently employment has fallen only about a quarter of a percentage point (still, that's about 350,000 jobs).

Problem: employment data are periodically revised. If we drew the same graph using the data that was available to guys like me four months after the start of each of the last four recessions (that is, "vintage" data), what would we see? The graph below shows this.

Revised answer: only in 1990 did the data in real time show a half a percent drop in employment. In 1980 and 1981, the data showed about a third of a percent drop in employment, still a bit larger a drop than what we see now. For the recession that began in March 2001, the data available in July 2001 showed exactly the drop in employment that we see four months into this possible recession.

Conclusion: the data doesn't scream recession. But we shouldn't be lulled into a false sense of security by the mild declines thus far in payroll employment. In past recessions the data have subsequently been revised downward.