The Federal Reserve Bank of Philadelphia last week released what it calls its coincident index for each state. The index is an effort to summarize current economic conditions in a single statistic.
The report, which used October data, makes Maine look pretty good, placing it among only 11 states that saw its index score increase by at least 1 percent in the past three months. Thirty-three other states had their index score increase, but by less than 1 percent, while four states saw their score decrease and two states remained stable.
Post continues after the map
Gov. Paul LePage’s office sent out a news release Monday morning touting the report as evidence of his administration’s success on improving the state’s economy.
“When we took office, Maine’s economy was a mess, workers were losing their jobs in droves, and our state budget was a nightmare after years of liberals who advocated for tax increases, out-of-control government spending and the use of one-time money from the federal government,” LePage said in a statement. “We’ve come a long way in the nearly three years since then. Our economy is growing, and over 8,000 more people are working since 2011.”
All that may be true, but this particular index probably isn’t the most reliable one to hang your hat on, according to Glenn Mills, chief economist for the Maine Department of Labor.
Though the Philly Fed has been producing this index since 2005, I had never heard of it before today, so I gave Mills a call to learn more about it.
From the Philly Fed itself, we know the index is based on four state-level metrics: nonfarm payroll employment, average hours worked in manufacturing, the unemployment rate, and wage and salary disbursements deflated by the consumer price index.
So, how reliable are these variables — and, therefore, the index — at measuring economic recovery (or recession)?
Mills, who also wasn’t familiar with the index, took a look at the report and the variables used to create the index score. His conclusion?
“I don’t put a great deal of credence on it primarily because, as we’ve discussed before, the nonfarm payroll job estimates, which go up and down on any given month, is one of the variables,” Mills said.
Mills and I have often discussed the monthly nonfarm payroll job and unemployment figures, and I’ve learned from him that these preliminary estimates are often unreliable in the short term. They do get revised over time, so are much more valuable when looking at long-term trends.
Mills had the same reaction to the other three metrics used to create the index. The unemployment rate has the same reliability factor on a monthly basis as the payroll data, he said.
“The third is manufacturing hours worked, so you’re talking 7 or 8 percent of jobs — our job total is about 600,000, manufacturing is 50,000, and the production workers in manufacturing are, ballpark, 38,000 to 40,000, so you’re already down to 8 percent of all the jobs,” Mills said. “So I’m not sure it’s that meaningful, and hour estimates are not very reliable either. I sound like a broken record.”
What about the fourth metric: wage and salary disbursements?
“That is the worst of all of them,” he said.
It’s a dataset released quarterly by the U.S. Bureau of Economic Analysis (the other three are released by the Bureau of Labor Statistics).
“They do a lot of projecting,” Mills said, adding that those projections are based on the aforementioned nonfarm payroll data. “So they’re doing a projection of income based on not very reliable job estimates.”
Even Maine’s Consensus Economic Forecasting Commission doesn’t really look at the quarterly wage and salary disbursement data because it’s unreliable, Mills said. Though, he said, as with job and unemployment data, it does get better over time.
According to a 2006 paper by Ted Crone, at the time VP of the Philly Fed’s research department, the coincident index helps “more clearly define business cycles at the state level.” Crone goes on:
“We can also learn about the course of the national economy from what is happening in the states. For example, by following the states whose indexes are declining we can trace the spread of national recessions across the country. Finally, by calculating an index based on the number of states in decline versus the number expanding we can get an early signal of national recessions.”
It’s important to point out that Mills isn’t saying the coincident index report is wrong in its estimation that Maine’s economy is improving. He believes it is, and that future revisions will show that Maine has added jobs and that Maine’s 2013 unemployment rate will be the lowest since before the Great Recession.
He just doesn’t think this index is a reliable way to measure it month by month.
After having criticized the index, I had to ask Mills, what’s the point? If an index like this is so unreliable, why even waste the effort?
“It’s actually a mystery to me,” Mills said. “At the various Federal Reserve Banks they put out a lot of this kind of stuff. It’s kind of disappointing because they have a bunch of people with PhDs who are a lot smarter than you or me, but they don’t seem to look too much below the hood. They just take data and use it, not question it.”