Alex Tabarrok writes,
Since the great recession ended, growth in real GDP has been much less volatile than in the 1950s to 1980s. Indeed, volatility has been lower even taking into account the great recession.
He goes on to point out problems with many theories that try to explain the continued Great Moderation.
I offered my explanation in 2003, in an essay called The Elastic Economy.
The United States economy has become more diverse and more robust. We are better able to withstand shocks, minimize concentration of economic power, and sustain growth without being hampered by resource constraints. This can be summarized by saying that the economy has become more elastic.
…There are several factors that have caused the economy to become more elastic. They include product diversity, globalization, the Internet, and increased innovation.
There are now more patterns of specialization and trade. That reduces the overall economic significance of any one product. That means that any particular shock causes less overall pain than it would have fifty years ago.
If this view is correct, then we should be less enthusiastic about claiming that policies in 2008-2009 prevented another Great Depression. Because we have a more elastic economy than we did in 1930, a repeat of the Great Depression was never going to happen.
The elastic-economy hypothesis fits in among theories of structural change to explain lower volatility in GDP growth. Alex points out that such theories, including the theory that sectors like health care are less volatile than manufacturing, want to predict further reductions in volatility in recent years, and this has not happened. Perhaps that is because asset markets have become a more important source of volatility than they were 30 years ago.
https://fred.stlouisfed.org/graph/?id=A191RO1Q156NBEA,
Lower growth. The GDP factory is a store, people waiting in line to get a GDP goodie. If the line is long, the line variance is large; and visa versa. If the GDP store is congested, store margins rise but volatility in inventory control increase.
I suspect there is a trade-off between the rate and volatility of growth. I also suspect we’ve gotten better at collecting and adjusting economic data.
The things he cites do make the economy more elastic. But the financial sector engages in a massive amount of risk management, which can absorb minor shocks but can make the economy much more brittle whenever asset values shift down in any significant way.
If this view is correct, then we should be less enthusiastic about claiming that policies in 2008-2009 prevented another Great Depression. Because we have a more elastic economy than we did in 1930, a repeat of the Great Depression was never going to happen.
1) Of course, if late 1930 we had a bank holiday, ended the Gold Standard and ended Prohibition then do we think the Great Depression simply becomes simply another Depression? Most modern people can not comprehend 25% unemployment rates society and the belief capitalism was at all time low.
2) We forget the NYT headlines of how often somebody said the economy was turning around in 1930 and 1931. But it never did.
3) What if several Lehman Brother happened? Citibank was practically dead in 2008 and Merryll Lynch was ~60 days away from bankruptcy. What about General Electric who needed the AIG bailout and the Fed buying Commercial Paper? Throw in Morgan Stanley and even Goldman Sachs was protected by AIG bailout. What if 1M auto manufacturing jobs were lost in 2009 from GM and Chrysler? That means another 2 – 3M lost jobs in 2009 – 2010. And what if continued job losses happened into 2010 and unemployment was hitting 15%+?
That is a lot of ‘What Ifs’
4) The most effective action the government did was simply drawing a line at lost banks versus surviving banks that they did not do until the 1933 bank holiday.
We might consider examining a bit more closely the Input-Output Tables of the BEA as they relate to the factors included (and not “covered”) by the use of a GDP “calculation.”
If we look at those (quarterly) Tables over the past several years, and draw trends, deviations, and comparisons to the GDP elements. We may see a different picture.
Just saying —
I don’t think it’s the number and diversity of sectors.
It’s the particular character of the sectors which have been becoming increasingly important and dominant in our economy.
I’ve previously laid out my “technologically determined” explanation for why the economy has evolved in the direction of the greater allocation of the workforce in hard-to-automate proximate services with low (and non-volatile) labor productivity growth, and why more and more economic activity is concentrated in the “Neo-Ricardian” sectors with inelastic demand, some of which you call the New Commanding Heights.
I’d say the most important of these sectors are health care, education, government, and real estate. In combination those sectors are able to extract large surplus rents from labor productivity. For many people I know, especially those under 40, these sectors eat up most of their budget. Other expenditures usually pale in comparison.
The key to being a Neo-Ricardian sector is that a human has no closely-desirable alternative opportunities and so must pay the piper to have the ability to live and work.
Now, in the past, calories were dear enough that agricultural land could rise to whatever price laborers could bear to keep themselves from starving, because it wasn’t feasible to quickly and affordably expand the supply produced for local markets, at least in some part because of slow, costly transportation options for imports, and lack of open trade. Most of the labor force was allocated to agriculture, and demand was very inelastic for those at the edge of subsistence, but supply was still susceptible to big shocks given the physical nature of the output and the processes to produce it. Sudden unexpected downturns in weather or disease would have large downstream volatility across the whole economy.
Eventually, with improvements to mechanization and intensive techniques, agricultural labor productivity started to skyrocket and increase much faster than the population, and in combination with a slack supply of arable land available to expand calorie supplies, and freer international trade, and cheaper, faster transportation, food costs dwindled to a smaller and smaller portion of budgets and the overall economy. So arable land was no longer the important Ricardian sector extracting rents from growing labor productivity.
There was an intermediate period where the new “expensive necessities” were manufactured capital like labor-augmenting factory machines and automobiles, which dominated the new economy, and the material and energy commodity inputs were also prone to volatile price swings, and that volatility thus also had big downstream effects on the larger economy.
But with further improvements in automation, transportation, and IT, labor productivity in “stuff” sectors is so high now that most people are working in services, most of which, by necessity, have very stable, slowly growing productivity.
The thing about the Neo-Ricardian sectors, however, is that they are more service-like and so removed from the vagaries and volatility of physical stuff inputs. Demand is inelastic – especially because of government intervention – and supply is stable, and price will adjust to whatever rent-extraction levels people can afford.
The combination of most work having no or low real labor productivity growth, and most budgets going to Neo-Ricardian sectors, means that volatility heads down to a kind of low asymptote of a new equilibrium.
So, how about, “The Buffered Economy”? Pure water has a very volatile ph if you add even a small amount of ions. But a buffered solution can take a lot of acid or base without much effect.