Tuesday, August 11, 2009

Economist's Notebook: GDP

In an Op-Ed in Sunday's New York Times, Eric Zencey goes on at some length about why Gross Domestic Product (GDP) is a poor measure of the health of the economy. He notes that beneficial activities like letting the wind dry our clothes would actually cause GDP to fall and then goes on to suggest that the very use of GDP would cause these activities to be officially discouraged. The first point is a fair one, the second is just dumb.

The first point, made in about 3,000 words, is incredibly easy to sum up for those that know some basic economics: GDP does not include externalities. GDP is simply the adding up of the price of all final goods and services produced during a period of time. Note that GDP is a flow variable not a stock variable, it is only looking at new production, not the accumulation of old production. Anyway, externalities can be both positive and negative and GDP does not account for either, precisely because GDP sums up prices and by definition, externalities do not show up in prices (though their effects may - e.g. if a harmful vapor escapes a chemical plant and sickens people, the medical bills are a part of GDP). So if one cleaning services air dries your clothes and another machine dries them and they charge the same amount, the contribution to GDP will be the same even though the machine dry place contributed to carbon emissions which impose a cost on society. On the same token, a person who hires a landscaper to clean up their front yard and, in so doing, increases the enjoyment of all the neighbors will contribute to GDP only through the hiring of the landscaper.

The second point, that the very use of the GDP measure is harmful, is dumb because no one that matters in terms of economic policy believes that GDP is the objective. Externalities are well known and the point of much of economic policy. Still, it is worth noting that almost every outcome we care about is positively correlated with GDP: education, infant mortality, life expectancy and, yes, energy efficiency (among middle and high income countries). So while GDP may not be the ends, it is an important means to the ends. The fact that GDP is of limited usefulness is precisely why other measures of well-being have been created, most notably the Human Development Index that looks at GDP along with measures of health and education. The UNDP is currently going one step farther and trying to develop an even more encompassing measure of economic opportunity.

Despite this, GDP is still a very important measure. Market based economic activity is how almost all households earn their livelihood, and an absence of such activity (or a decrease) means that there is less income being earned and almost all of that has little to do with switching to air drying our clothes. So if we care about the suffering of US households in the current economic downturn, we should care about GDP.

So GDP does measure consumption, but does not encourage consumption. I don't think about how my activities are effecting GDP when I choose what to do, buy, or trade. It is important to note that efficiency is also positively correlated with GDP, so you can't say in one breath that GDP encourages consumption and discourages efficiency. If I start a GPS traffic navigation service that directs people away from congestion in real time, this will save energy, save time lost waiting in traffic, reduce carbon emissions and positively contribute to GDP through the value of the service I create and charge for.

GDP is not the problem and focusing on a metric is a nonsensical approach. Focusing on the real goals and objectives is where our attention should be placed.

2 comments:

Ralph said...

Hi Patrick,

Just a quick question: Are you suggesting GDP should include externalities?

Also, I was under the impression GDP is a products and services sold/consumed, not simply produced. I was a car manufacturer, producing a car that sits on a lot (doesn't sell) doesn't count towards GDP. Is that right?

Thanks.

Patrick Emerson said...

No, I am not suggesting it should include externalities (or at least I was not intending to). I was pointing out that this is a limitation of GDP as a measure of economic activity, but does not, in itself, mean that GDP is useless or bad.

GDP is supposed to be all final goods and services produced in a year but it is generally measured by consumption. So a car made in 2008 is supposed to count towards 2008 GDP, but if it is not bought until 2009, in reality it is generally counted in 2009 GDP. But this type of measurement error is generally not terribly significant. In today's economy, however, unsold cars may not be insignificant...