Bob Murphy asks

An economy is chugging along nicely at full employment, and price inflation is within the desired range. Real GDP is $1 trillion. Then geologists think they’ve discovered a humongous deposit of oil that will make Saudi Arabians feel like chumps. The problem is, the oil is buried pretty deep. So the oil industry in this country (not foreigners) spends $150 billion buying new drilling equipment and other necessary infrastructure. At the end of the year, the macroeconomists report that real GDP was $1.1 trillion. There was an extra $150 billion in output in the sectors producing the oil equipment, but that was only partially offset by a $50 billion drop in output elsewhere. The apparent discovery of the oil increased the productivity of the factors in the economy, which is what allowed potential GDP to go up 10% in a single year.

The next year, to their horror, the people in the oil industry realize that it wasn’t an oil deposit at all, but just some empty bottles that John Maynard Keynes had buried back in 1936. The entire drilling apparatus overnight becomes almost completely worthless, because it can’t be easily disassembled and shipped elsewhere.

So: Assuming no other technological discoveries or workers gaining skills, real output at best will drop back to $1 trillion in the new year, and will actually be less because some of the maintenance on other production processes would have been shunted into the oil industry the year before.

My question: How would macroeconomists in the CBO do their graphs? Would they go back and mark down the $1.1 trillion “real output” figure in the previous year, because that was obviously a mistake? Or would they say, “No, real output really was that high last year, and it just collapsed this year”?

The short answer is no. Real output in the year in question was 1.1 Trillion. Gross Domestic Product is our attempt to measure real output and it would reflect that.

I think there are a few important concepts related to Bob’s question. One is Net Domestic Product, which according to Bob’s description did not rise as fast as Gross Domestic Product and could possibly have fallen.

Net domestic production is Gross Domestic Production minus the Consumption of Fixed Capital.

Its not beach reading by any means but for those who are really interested I strongly recommend the BEA Conceptual and Methodological Handbook.

A key chart that may prove useful


Now in fairness to Bob’s point the consumption of fixed capital is measured using the perpetual inventory method. Key in this calculation is economic service life, which must be imputed based on historical measures. Yet, Bob is postulating a strong shift in economic service life.

This would not be picked up immediately in Net Domestic Product calculations but would be picked up as the series was revised over time.

Essentially the BEA must wait until the capital has actually worn out early before going back and saying that it was wearing out at a faster than average rate.


However, the oil in the ground in would not factor into any of these statistics at all.

This is GDP and its cousins are attempting to measure productive output and income. The oil was not produced by human beings. Instead it is a measure of wealth. It would be captured in the Federal Reserve’s Flow of Funds report.

The increase in the value of the land under which the oil was located should be captured in Net Assets of Households and Nonprofits. When the oil was found not to be there then a corresponding negative entry would be entered.


Lastly, neither Flow of Funds or NIPA captures the concept of welfare which is itself distinct from either wealth or production. It depends on taken advantage of areas where the total rather than marginal willingness-to-pay exceeds the total rather than marginal cost.


This is of course also distinct from what I would term social welfare, which involves having some non-income based aggregation method across individuals. Of course nothing like this even remotely exists.