Friday, October 10, 2014

Limits to Growth One More Time

Not much time this morning, but I’d like to respond to the dustup between Mark Buchanan and Paul Krugman over whether energy (and other resource) use can be decoupled from GDP growth.

1. I get the impression that Buchanan identifies GDP with “stuff”, at least subconsciously.  But GDP is value, what people are willing to pay for.  When I teach an econ class, that’s a component of GDP.  If an extra student shows up, that’s GDP growth.  Of course, a lot of GDP really is stuff, but as economies develop they tend to become less stuffy.  Overall not enough, but how unstuffy they could become is an empirical, not a theoretical matter.

2. But I agree with Buchanan that increases in energy efficiency alone are unlikely to accomplish what we need to contain climate change.  Absent changes on other fronts, the growth of demand for energy services will simply swamp the effect of greater efficiency.  This has been true in the past and any realistic projection puts it in our future as well.

3. And Buchanan is also right that there is no historical precedent for the kind of decoupling between economic growth and fossil fuel use (let’s be specific here) that we would need to meet both economic and climate goals.  The notion of foregoing most of our remaining supplies of extremely energy-dense minerals flies in the face of all of human history.  That’s why it will be a big challenge to bring it off.  The challenge begins with putting in place a policy that prohibits most fossil fuel development.  You can discuss the particulars, but there is no getting around the need for a binding constraint.  The reason is exactly the one that Buchanan pinpoints: increases in efficiency and even increases in renewable energy sources alone will not be sufficient by themselves to offset the energy demands stemming from global GDP growth.

Can we compel most fossil fuels to stay in the ground and still have economic growth?  Since this is about growth in value and not necessarily stuff, the answer still seems to be yes.  But we won’t have a sufficient shift away from stuffiness without measures that prohibit dangerous levels of fossil fuel extraction.  An unprecedented change in the trajectory of economic growth requires unprecedented policies.

Postscript: I’m not interested in whether “unlimited” economic growth at some distant future date is incompatible with resource constraints.  It’s also true that economic growth can’t continue after the universe collapses in on itself.  We’ll let distant future people, if they still exist, worry about this.

12 comments:

Sandwichman said...

I agree with two of the main points here, Peter:

1. "no historical precedent for the kind of decoupling between economic growth and fossil fuel use" that we would need.

2. "An unprecedented change in the trajectory of economic growth requires unprecedented policies."

Having waded through some of the very latest, up-to-datest published research on climate economy modeling, it seems to me that some economists (besides the usual heterodox ecological econ suspects) are not so blind about the limits as Buchanan supposes.

In particular, the responses to the Weitzman "dismal theorem" are encouraging (intellectually, that is). It almost seems to me that we are on the cusp of a breakthrough to constructive dialogue! See the excerpts and abstracts at the end of my D.I.C.E. post.

Thornton Hall said...

It's interesting to think about how many critiques of the economics discipline get dismissed out of hand because the critic makes an error of the sort "GDP growth = more stuff".

Errors like this tend to make Krugman uncivil, but as you demonstrate in this case, such errors are frequently not as fatal as Krugman's juicy adjectives would have you believe.

Notice too that this error belongs to a class of "elementary mistakes" that may not be mistakes at all if you leave the fact free zone known as academic economics. True, some of the GDP value is services, but you can't build a massive pile of services in Florida and shield it from bankruptcy. Growing inequality drives demand for the things rich people buy: assets and conspicuous consumption.

Thus, lowering the top marginal rates in 1981 set the stage for asset bubbles, especially second houses in Florida. This is obvious when you know that the rich need to consume stuff, but totally obscured by the nonsense that fills Krugman's head.

Sandwichman said...

I think there is an element of "Kekule's benzene dream" in the notion that GDP growth is about "value" and not stuff. This is absolutely true. But it becomes rather nebulous when we try to pin down what we mean by value. Yes, I know, NIPA and all that.

The big Oroboros enters the dream in the form of financial services. There is, hypothetically, no limit to the "value" of financial assets, consequently the "services" provided by the FIRE sector can also expand indefinitely with very little additional energy consumption. To what end?

This is where the meaning of value becomes important. Is the financial asset Oroboros a Bubble? And what is the enduring utility of all those financial services after the Bubble has burst?

This line of inquiry could extend to the role of the discount rate in climate economy models like D.I.C.E.

blissex said...

«I get the impression that Buchanan identifies GDP with “stuff”, at least subconsciously. But GDP is value, what people are willing to pay for. When I teach an econ class, that’s a component of GDP. If an extra student shows up, that’s GDP growth.»

To me that sounds like a distorted version of the usual neo-liberal propaganda, because GDP properly speaking is not value, it is entirely *stuff*, as in (final, gross) "Domestic Production" of goods and services.

If teaching is a component of GDP, which almost surely it should not be (because it is arguably alike to R&D which used to be correctly considered an intermediate input), what gets added to GDP is not the "value" of the teaching (bizarre concept!) or the price paid for the teaching, but the *hours* of teaching, which is indeed "stuff", even intangible "stuff".

GDP is a vector of physical quantities, like cars per model per year, tons of cement by type, passenger kilometres for train travel, hours of movies watched in cinemas, tons of garbage collected per year, hours of lawyer work, etc.

Perhaps when you acritically repeat the bizarre idea that «GDP is value, what people are willing to pay for», then as to «willing to pay for» part (the "value" part is just weird) you are talking about *nominal* GDP, which is not *GDP*, but an index related to GDP, computed by taking some vector of prices and multiplying it by the vector of quantities that is GDP.

The vector of prices used to compute the nominal GDP index is not part of GDP, because prices can arbitrarily differ from time to time and place to place, making nominal GDP essentially meaningless except across comparisons with small differences in time and place.

Consider the difficulty with computing yearly "nominal" GDP: while yearly GDP is a conceptually clear vector, which vector of prices should be used to compute "nominal" GDP? That at beginning of the year? At the end of the year? In between? That in another country with stable prices? Whatever?

Therefore often *nominal* GDP is an index computed with different prices vectors, giving different results, but GDP stays the same, because the (gross, final) Domestic Product does not depend on the vagaries of prices.

Except as redefined by the artful dissembling of neo-liberal Economists.

Also, «what people are willing to pay for» is not «value», except perhaps in the ignorant fantasies of neoclassical (that is, mostly neo-liberal) Economists.

To the point that in computing "real" GDP some neo-liberal national statisticians adjust prices by quality, precisely on the basis that «what people are willing to pay for» is not "value", but "value" is price adjusted by quality, in their view.

Neo-liberal propagandists love to spread the novel and ignorant attitude that GDP really means "nominal GDP", the index computed by averaging GDP with a (largely arbitrary) price vector, because nominal GDP being an index based on somewhat arbitrary prices makes it easier to "massage".

blissex said...

«makes an error of the sort "GDP growth = more stuff" [ ... ] True, some of the GDP value is services,»

GDP growth is indeed more "stuff", if "stuff" is properly defined to include *physical* quantities rather than just *material* quantities.

It used to be that "real socialist" economies denied that services were "production" (from their interpretation of what Marx wrote), so they computed *Material* Product, by counting only the goods final output from the primary and secondary sectors. But that is not what we are talking about here.

«There is, hypothetically, no limit to the "value" of financial assets, consequently the "services" provided by the FIRE sector can also expand indefinitely with very little additional energy consumption.»

OMG! You have figured out that there is a banker behind the curtain in the land of Oz! ;-)

Or more precisely that the neo-liberal Economists have been using as part of their propaganda the confusion between GDP and nominal GDP as mentioned previously. Because with certain services as you say "price"/"value" is nearly arbitrary, boosting *nominal* GDP at will.

There is also another matter that is far more fundamental: that GDP is supposed to measure something like value added, that is final production. For a lot of services it is quite hard to define value added, and it is both difficult conceptually and hard to measure.

While for example for cars one counts the production of cars and then subtracts from the relevant components of the GDP vector the tons of steel, paint, etc. used to manufacture them.

Even worse there is a question as to many services whether they produce any value added; for example a lot of contemporary finance turnover is pure gambling ("derivative" is a synonym of "bet") and whether it produces any value added *as an economic activity* is rather questionable.

Similar questions apply to service activities like education and R&D, because in general people get trained or do research and development for the purpose of increasing productivity, not for their own sake, and therefore their output already shows up, *if any*, in the output of other industries, and should not be double counted.

Otherwise the argument is that finance, education and R&D should be counted as final output regardless because they are *entertainment* activities like going to dance, and thus are final output.

But while I don't doubt that some or even many people enjoy trading derivatives, or being educated, or doing R&D purely for their own sake, just like some people would manufacture car axles for the sheer joy of it even if there were no cars to put them in, I tend to agree with the past practice of counting R&D as an intermediate rather than final product.

But double counting intermediate activities as if they were final production is all the rage nowadays that actual GDP and GDP growth are a bit harder to come by.

blissex said...

For a supporting authority, here is the usual, rather understated, view from Stiglitz:

http://host.madison.com/ct/article_71fad514-9caa-11de-9a00-001cc4c03286.html
«For example, while GDP is supposed to measure the value of output of goods and services, in one key sector – government – we typically have no way of doing it, so we often measure the output simply by the inputs. If government spends more – even if inefficiently – output goes up. In the last 60 years, the share of government output in GDP has increased from 21.4 percent to 38.6 percent in the U.S., from 27.6 percent to 52.7 percent in France, from 34.2 percent to 47.6 percent in the United Kingdom, and from 30.4 percent to 44 percent in Germany. So what was a relatively minor problem has now become a major one.
Likewise, quality improvements – better cars rather than just more cars – account for much of the increase in GDP nowadays. But assessing quality improvements is difficult. Health care exemplifies this problem: Much of medicine is publicly provided, and much of the advances are in quality.»

Note that even here he repeats the funny view that GDP is defined as «the value of output of goods and services» (thus continuing the popular and clever confusion between GDP and nominal GDP), and does not mention that the production of the private service sector has much the same issues he mentions for the government sector, and taken together that means that over 80% of GDP is rather a "problem" rather than the 38-50% he reports just for

Sandwichman said...

"double counting intermediate activities as if they were final production is all the rage nowaday"

As Simon Kuznets complained it was back in the days when the Commerce Dept. introduced NIPA. Comm. Dept. replied it would be impractical to sort it all out, which it probably would be. And so, we rely on a 70-year old stale- dated makeshift that had a shelf life of 20 years.

I know, I know, There Is No Alternative and besides there are things that GDP is still useful for, even if it isn't a good proxy for welfare. BUT that's beside the question of whether climate policy should be yoked to the IMPERATIVE of GDP growth (or, for that matter to shrinkage or stationarity).

If we do the right things and growth happens, so be it. But if we DEFINE what are the right things to do by whether or not they promote GDP growth, we're fucked (and fucked in the head).

Thomas L. Hutcheson said...

"And Buchanan is also right that there is no historical precedent for the kind of decoupling between economic growth and fossil fuel use." But there has never yet been any effort to put a price on the damage done by CO2 accumulation and let that price percolate through the economy changing not only energy efficiency of production but the mix between fossil and non-fossil sources of energy, the fossil fuel intensity of the consumption basket, incentives to sequester
co2, and the encouragement to innovate in all the margins of the economy to reduce co2 emissions.

blissex said...

«As Simon Kuznets complained it was back in the days when the Commerce Dept. introduced NIPA. Comm. Dept. replied it would be impractical to sort it all out, which it probably would be.»

Sure, as I wrote defining never mind measuring the net output of some sectors is hard.

But as Stiglitz says defining never mind measuring "quality improvements" is also quite hard, but that has not stopped the BEA to do so with zeal, presumably precisely because they "account for much of the increase in GDP".

Important note: the BEA applies hedonics to *nominal* GDP, not to inflation-adjusted ("real") GDP, as documented in detail here:
http://globaleconomicanalysis.blogspot.co.uk/2005/05/grossly-distorted-procedures.html

«there are things that GDP is still useful for, even if it isn't a good proxy for welfare»

The really critical aspect of nominal GDP seems to me that it is market-moving, that is higher GDP drives stock prices up, and as the Fed Board actions of the past 30 years show, driving up stock prices is probably the second most important policy goal of the USA political system.

blissex said...

«But there has never yet been any effort to put a price on the damage done by CO2 accumulation and let that price percolate through the economy»

In a discussion on limits to growth this is an important topic, but the present discussion is about strictly growth of GDP, and criticizing GDP for not including «the damage done by CO2 accumulation» is totally unfair, because as the name says GDP is "Gross" of depreciation.

And environmental damage is depreciation of capital, so it is not measured by GDP.

So the issue is that using GDP to assess long term growth is a very bad idea, because ignoring capital depreciation in the long term is absurd.

But GDP was designed as a tool to measure *production* in the short term over the cycle, for short term policy's sake, not for strategic management of the economy.

Using gross domestic product instead of net domestic product to drive strategy is more or less suicidal as it encourages fraud and asset stripping.

Accordingly the Chinese government have changed their metrics for regional party chiefs:

http://blogs.wsj.com/chinarealtime/2014/04/30/cutting-the-fudge-chinas-bloated-local-gdp-numbers-go-on-a-diet/
http://ftalphaville.ft.com/2014/08/19/1936122/reform-vs-seasonality-in-chinese-gdp-stats/

Nothing of the sort will happen in the USA, because fraud and asset-stripping seem to be strategic technologies in pushing the stock market up, which seems to be the strategy goal of the USA political system, as good old Grover put it:

«And that is, in 2002, on the investor class stuff … you could have said, just drop $7 trillion in stock market value with the collapse of the bubble … $7 trillion, trillions with a T … Americans had $7 trillion less than they used to have, you can expect them to be very irritated and in trouble. [ ... ]
They were mad at having lower stock prices and 401(k)s, but they didn’t say Bush did this and that caused this. Secondly, the Democratic solution was to sic the trial lawyers on Enron and finish it off. No no no no no.
We want our market caps to go back up, not low. The 1930s rhetoric was bash business — only a handful of bankers thought that meant them.
Now if you say we’re going to smash the big corporations, 60-plus percent of voters say “That’s my retirement you’re messing with. I don’t appreciate that”. And the Democrats have spent 50 years explaining that Republicans will pollute the earth and kill baby seals to get market caps higher.
And in 2002, voters said, “We’re sorry about the seals and everything but we really got to get the stock market up.»

blissex said...

«"double counting intermediate activities as if they were final production is all the rage nowadays"

As Simon Kuznets complained it was back in the days when the Commerce Dept. introduced NIPA.»

But the double counting is expanding I suspect because of a very important and pernicious technicality that is often missed.

"other things being equal" double counting does not affect growth rates, because if applied throughout time consistently then past GDP is proportionally raised as much as current GDP and the net effect on growth zeroes out.

But that does not happen if the rate of growth of the input is higher than the rate of growth of the output, especially if the *price* used to compute that bit of nominal GDP grows faster than average prices.

So for example if R&D or education or financial services grow faster than production of final goods and services, and their cost in dollars grow faster than most prices, then treating them as final production as if they were entertainment means boosting not just nominal GDP but also its growth rate.

But if the rate of growth of the input is higher than the rate of growth of the output then the growth rate of final production is actually shrinking, because it takes more inputs to generate the same output.

Such are the wonders that "sell-side" statisticians and Economists can deliver!

Sandwichman said...

"...because fraud and asset-stripping seem to be strategic technologies in pushing the stock market up, which seems to be the strategy goal of the USA political system..."

As Martinez-Alier put it "Externalities are not so much market failures they are cost-shifting successes."

This all goes back to usury and bills of exchange where the secret to wealth was keeping bad things well hidden on the double-entry books.

Except in the old days it was the usury that had to be hidden by bills of exchange. Now it is the social cost of carbon emissions that are being concealed by our old friend compound interest in the form of an arbitrary "discount rate" which makes postponing GHG abatement far into the future optimal for economic growth (i.e., what used to be known as usury).