Friday, April 30, 2010

Five Centuries of Inflation

Continuing through the Reinhart-Rogoff paper I mentioned yesterday, I came across this amazing figure which shows the average inflation rate across all the countries in their sample back to 1500!  It's great to have datasets that cross over back into the pre-industrial era.

Of course, in the early years in this sample, currency was silver or gold, and inflation was achieved by debasing it (making coins smaller or mixing the silver/gold with less precious metals).  The thing that struck me about the data was it seems to divide into three main eras based on the energy basis of society and the type of currency:

Prior to the early 1800s, the world was largely pre-industrial (the Watt steam engine was invented in 1775, the first practical steamboat came in 1802, and the Stockton and Darlington railway opened in 1825).  Pre-industrial societies produced a very small economic/energy surplus, and grew very slowly.  During this era, inflation was on average a few percent, with broad fluctuations.

Once fossil fuel powered industry became widespread, economies could and did grow a lot faster.  It appears to have been a little difficult to grow the money supply fast enough under the circumstance - inflation is noticeably lower in the nineteenth century: under 2% almost all the time, with several decade-long deflations in the mix.

Then in the twentieth century, we see the widespread use of fiat currency, which can (as long as you don't let all the banks collapse as in the great depression) generally maintain some minimum inflation rate.  The drawback, also clear in the data, is that it allows stressed-out or profligate government to inflate much faster than pre-industrial governments ever could (back in the day that one had to physically melt the old currency down and re-issue the debased version).


kjmclark said...

Or put another way, when societies had to use largely sustainable resources, there was greater scarcity, so inflation was more likely. Malthus supposedly based his early thoughts on that history. Fossil energy reduced the scarcity of most things by providing ample energy.

What a concept though. They went to fiat currencies because they had a harder time maintaining inflation, because energy became ample? It also makes sense because energy could be used to replace labor, and was, in many industries.

Stuart Staniford said...

"energy became ample". I'm guessing it's more like "the availability of energy relative to precious metals" increased.

Gary said...


Any idea of the amount of debt/credit as a percent of real GDP during the time of that data set? My guess is that credit really expands during the "industrial/fiat" period because, after all, thats what we mean by a fiat money system.

I'm beginning to wonder if if a higher target inflation rate might not be a good idea in the years ahead. Here is why:
1) If inflation is generally higher, so will be average interest rates. This gives central banks more headroom to turn things around during recessions without hitting the zero bound.
2) As the economy begins to grow less strongly, or begin to de-grow, because of the cost of energy, the money supply can still continue to increase.
3) Without inflation, in an economy with low or negative growth, repayment of debt becomes a big problem. Inflation provides a simple way to gently sink all boats while still providing the usual capitalist incentives to keep paddling.

Some might question if central banks can control the inflation rate well enough to make such a target meaningful. Lots of "doomers" expect hyperinflation is just around the corner because of the current bout of deficit spending. Paul Volker might not be so worried.

Any thoughts?


Anonymous said...

I'm seeing three huge inflation spikes corresponding to post-WW1, post-WW2, and (1973?)

What caused the period of relatively low inflation in the 50's and 60's?