Tag Archives: benchmark revisions

Most people have the impression that these various payroll and employment reports just go into the raw data and count up the number of payrolls and how many Americans are employed. Perhaps the BLS taps the IRS database as fellow feds, or ADP as a private company in the same data business of employment just tallies how many payrolls it processes as the largest provider of back-office labor services. That’s just not how it works, though. In fact, sampling and...

Extraordinary times call for extraordinary commitment. I never set out nor imagined that a quarter century after embarking on what I thought would be a career managing portfolios, researching markets, and picking investments, I’d instead have to spend a good amount of my time in the future taking apart how raw economic data is collected, tabulated, and then disseminated. Yet here we are. I’m not saying, nor have I ever alleged, the government is cheating, cooking the...

The Bureau of Labor Statistics (BLS) has begun the process of updating its annual benchmarks. Actually, the process began last year and what’s happening now is that the government is releasing its findings to the public. Up first is the Household Survey, the less-watched, more volatile measure which comes at employment from the other direction. As the name implies, the BLS asks households who in them is working whereas the more closely scrutinized Establishment...

The Chicago Fed’s National Activity Index (NAI) fell to -0.36 in July. That’s down from a +0.10 in June. By itself, the change from positive to negative tells us very little, as does the absolute level below zero. What’s interesting to note about this one measure is the average but more so its rate of change. The index itself is a product of econometric research. Economists had been searching for an alternative to the unemployment rate in order to increase the...

I missed durable goods last month for scheduling reasons, which was a shame given that May is the month each year for benchmark revisions to the series. Since new estimates under the latest revisions were released today, it seems an appropriate time to revisit the topic of data bias, and why that matters. What happens with durable goods (or any data for that matter, the process is largely the same) is that the Census...