Consider the calendar. It's useful for keeping track of your schedule and your Mom's birthday, and remembering Arbor Day.
But a calendar is also a marvelous tool for prediction. My calendar says there will be a full moon Sept. 9 and no moon Sept. 24. Astronomers have figured out the orbit of the moon so precisely that they can predict its phases years, decades and centuries in advance. And it's printed on your calendar.
I think about that at this time of the year, because August is the month that Purdue's Department of Agricultural Economics does its economic outlook. I'm responsible for the general forecast - gross domestic product, unemployment, inflation and interest rates.
Lately, some of our economic indicators seem to have taken on strange, new orbits.
Take the unemployment rate. From 1961 to 2010, real GDP grew less than 2.5 percent in 12 years, and every time the unemployment rate went up. Over that same period, growth was more than 3 percent in 30 years, and in all but three years, the unemployment rate went down. That was an orbit you could count on.
Then in 2011, real GDP grew slowly, and the unemployment rate dropped. Then it happened again in 2012. And in 2013. And it might happen this year, too, once the final numbers are in.
It's like the moon decided to reverse course.
What's going on? Well, the reason that real GDP has to grow to keep unemployment from rising is that the labor force keeps growing. More people want jobs, so more jobs must be created, or people will be unemployed. From 1961 to 2010 labor force growth averaged 1.6 percent per year. Since then, it's averaged 0.3 percent per year. Since fewer jobs need to be created, unemployment can be reduced with slower real GDP growth.
Some of the slowdown in the labor force is due to discouraged workers. These are folks who gave up looking for work because the Great Recession got so bad. They might return as the economy improves. But some of the slowdown is due to retirement by baby boomers. The oldest boomers turned 65 in 2011. For that reason, the labor force probably will grow more slowly from now on.
In the future, real GDP growth above 1.5 percent will probably be enough to make the unemployment rate fall.
Then there's inflation. Too much money chasing too few goods causes inflation. Everybody knows that. From 1961 to 2008, there were 31 years when the quantity of money grew more than the output of goods and services, measured by real GDP growth. In 19 of those years, the inflation rate topped 3 percent. There's something there, though that's not a really tight fit. In 17 years, money grew less than goods, and in 14 of those years inflation was less than 3 percent. There's your orbit.
Since 2008, the quantity of money has increased an average of 12 percent per year, way more than real GDP growth. Did inflation explode?
Nope. The inflation rate has been less than 3 percent each year. It's another orbit out of whack.
What happened? There's too much money. There's too few goods. But the money isn't chasing. It's sitting around in bank reserves, corporate accounts and household savings. If it's not being lent or spent, it's not out there bidding up prices.
There's an alternate prediction rule for inflation. When the unemployment rate is high, the inflation rate goes down. From 1961 to 2008, there were 19 years when the unemployment rate was greater than 6 percent. In 15 of those years the inflation rate went down. It dropped by an average of 0.6 percent per year.
In 2008, the inflation rate (not counting food and energy) was 2.3 percent. The unemployment rate has been above 6 percent since 2009. So, using the past average, if inflation declined 0.6 percent per year for five years, by now the inflation rate should be negative. Deflation!
Wrong again. Inflation has been low, but it hasn't turned negative. Surprisingly, an older version of this orbit seems to be working better - high unemployment keeps the inflation rate low but steady.
What's life like for economic forecasters these days? The economic moons have changed their orbits, and our economic calendars are wrong.