Forum: Are productivity statistics a load of tosh?

Cyclical effects muddy the measurements

Over the past few months, usually after a beer or two, I’ve been heard to loudly proclaim that the productivity statistics so often used to paint a picture long-term deterioration and failure of the New Zealand economy are a load of bunk.

I didn’t actually use the word “bunk”, but you catch my drift.

The issue about the statistics is that they are often misused as a measure of national efficiency, which they are not. They are also often used as a stick in shallow political argument.

The formula used is GDP divided by the total number of hours worked in the economy as measured in the Household Labour Force Survey. As such, it is a measure of labour productivity, but even that isn’t a measure of efficiency.

Productivity growth was low in the 1990s and took a spike in 1999 as New Zealand rebounded from drought. Productivity growth has been low in the 2000s as well, despite significant investment in productive capacity since 2004.

The major problem with the numbers is they measure, in large part, cyclical changes in the economy, as Westpac senior economist Dominick Stephens writes in a recent report (pdf) called “More for less”.

“Half to three-quarters of the acceleration is due to technical factors that were previously depressing productivity growth and are now temporarily exaggerating it,” he writes.

“The remainder may be a belated payoff from high investment activity over the past few years, and could be more long-lasting,” he writes.

Stephens is cheered by a recent productivity surge (3% over four quarters from around 1% previously) which had been expected but was a long time coming.

Some of that investment, no doubt, is in the area of information technology, but measuring productivity or efficiency growth from such investment has been highly problematic over the years. In the 1990s it was an area of great dispute, taxing minds such as that of Nobel economics prize winner Robert Solow.

His work and the work of others in response had to get to the nub of the issue: what is productivity?

As Stephens points out, the worth of the statistics depends on your goal.

“Are we getting more efficient? A lot of wage productivity statistics fall way short of measuring that,” he says.

“Is it really a productivity slowdown when low-skilled people pour into the workforce? Are we really being less efficient?”

The nub of this issue is that labour enters the economy when it is in growth mode and this can depress productivity growth. It is, he writes, anti-cyclical:

“Rapid employment growth has been a key cause of low productivity growth this decade. To get an idea of magnitudes we reviewed three economic studies that all came to similar conclusions: rapid employment growth from 2000 to 2006 depressed New Zealand’s productivity growth rate by roughly 0.5 percentage points per annum.”

Stephens says a half to three-quarters of the current measures of productivity growth could be cyclical, in the same way that cyclical factors depressed the measures in the past. However, a “small slice” may be a payoff from investment in productive capacity. About 0.2% is directly attributable to growth in oil exports from the Tui well, he says.

So the next time you read these figures trotted out to show our economic failure, or our success, remember they are a very rough yardstick indeed.

Join the newsletter!

Error: Please check your email address.

Tags Westpacproductivity

Show Comments
[]