Economies are Long-Run Processes

by Don Boudreaux on November 3, 2006

in The Economy

Paul Krugman, among others, has warned that today’s economy is scarily different from economies in the past.  His reasoning, as explained in this June 16, 2006, column in the New York Times, is this:

In fact, the distinctive feature of the current economic expansion –
the reason most Americans are unhappy with the state of the economy, in
spite of good numbers for the gross domestic product and explosive
growth in corporate profits — is the disconnect between rising worker
productivity and stagnant wages.

I’ve never bought the implication that productivity is really disconnected from workers’ compensation.  Capital is too fluid and the American labor market too competitive for such a "disconnect" to be lasting.  The fact that such a "disconnect" shows up in economic reports reveals more about the shortcomings of inferring long-term trends from the happenings during arbitrary (and ususually short) time periods such as "month," "quarter," or "year."

Now today’s Boston Gbobe has this report, entitled "Pay outpaces productivity: inflation feared."  And here’s a letter that I sent to the Globe in response:

Dear Editor:

months we’ve been warned that the current economic recovery differs
ominously from past recoveries because worker pay now is rising more
slowly than worker productivity.  But in today’s paper we read that
"Growth in productivity – the key ingredient for rising living
standards – skidded to a standstill in the late summer while workers’
wages and benefits shot up at the fastest clip in more than two
decades" ("Pay outpaces productivity: inflation feared," Nov. 3).  In
other words, workers’ pay is catching up with their productivity, just
as economics predicts.

Economies are long-run processes; they
should be evaluated as such.  What happens in any arbitrary time period
- a month, a quarter, or even a year – typically is too filled with
short-run distortions and lags to present a reliable picture of an
economy’s long-run trajectory.

Donald J. Boudreaux

By the way, just as we have no real reason to fear that wages will not keep pace with worker productivity, we also — for a variety of reasons — have no reason to fear that rising wages will cause higher inflation.

Be Sociable, Share!



Add a Comment    Share Share    Print    Email


triticale November 3, 2006 at 9:18 pm

Krugman is supposed to be both intelligent and educated in economics. Surely he knows that "stagnant wages" refers only to the statistical artifact "median wage" and not to the wages earned by any given individual. Does he count on his readers not knowint that?

True_Liberal November 4, 2006 at 10:23 am

Whatever happened to the premise that inflation is simply "too few goods and services chasing too many dollars"?

ben November 4, 2006 at 6:15 pm

Krugman is the best example I know of an intelligent person so blinded by religion, in this case modern liberalism, that he is incapable of producing a balanced argument where that religion conflicts with evidence or reason. Very uninspiring.

Mark Brady November 4, 2006 at 11:15 pm

"triticale" writes thus:

'Krugman is supposed to be both intelligent and educated in economics. Surely he knows that "stagnant wages" refers only to the statistical artifact "median wage" and not to the wages earned by any given individual. Does he count on his readers not knowin[g] that?'

I'm sure Krugman does, just as Don Boudreaux and Russ Roberts do whenever they use statistical artifacts like "average wage", "productivity", the Consumer Price Index, the terms of trade, etc., and make comparisons between nominal and real wages and incomes. Of course, all these statistical artifacts have their limitations, but, for most households, the "median wage" and the "median income" are more meaningful than the "average wage" and the "average income".

python November 4, 2006 at 11:50 pm

Mark Brady,

So your point is that it's okay because others do it?

Please cite the last time that either of the 2 Profs on this blog suggested that average income was more meaningful than median.

Morgan November 5, 2006 at 12:29 am

Mark Brady:

I think you missed the point of triticale's "…wages earned by any individual", which probably refers to the fact that people tend to move up the income distribution over time. So the median wage can remain unchanged even if the median lifetime earnings profile improves. In addition, median wage doesn't include benefits, and is impacted by immigration from low wage countries, the age distribution of the workforce, and the number of workers working part time while they pursue education or semi-retirement.

So there are lots of reasons to reject the use of "median wage" as the only measure of change in standard of living over time – just as there are to reject other single measures.

I haven't seen you comment here before, so I presume that you aren't familiar with the hosts' frequent posts regarding the problems with measuring CPI. Like python, I've also never seen them tout "average wage" as a useful statistic.

Russell Nelson November 5, 2006 at 2:06 am

I have earned zero wages for the past 15 years. I am self employed.

Kent Gatewood November 7, 2006 at 8:56 pm

Don't immigrants from low wage countries vote Democratic and socialist when they come to America?

Previous post:

Next post: