Peter Bernholz surveys the global-historical experience of inflation, and argues that there are important commonalities in the political and economic sources of moderate (less than 50 percent per month) inflations and of hyperinflations, and in the recipes for successful termination of inflation. What are they? Firstly, all governments have an inflationary bias, and hyperinflations in particular arise from monetizing deficits. Secondly, in a moderate inflation, prices may not rise as fast as the money stock, so real money balances may rise, whereas in a hyperinflation the opposite occurs and real money balances may plummet. Thirdly, under a commodity money system, a hyperinflation was virtually impossible—the only hyperinflation earlier than the twentieth century, being the assignat period in revolutionary France. Finally, moderate inflations can be concluded by adoption of a monetary constitution while hyperinflations require the kind of (economic) “regime change” described by Thomas Sargent (“The End of Four Big Inflations” in Robert Hall, ed., Inflation: Causes and Effects [Chicago: University of Chicago Press, 1983]). This summary will suggest to the reader that the overall picture presented here is not very new, and to an extent I think that is right, but a point form summary does not do justice to the breadth of the evidence that Bernholz brings to his discussion, and the book will be a useful reference for anyone who wants a broad overview of inflation and hyperinflation. In addition, it provides a wealth of source material; the appendix, for example, contains detailed sources for the study of all 29 episodes of hyperinflation.

The book presents a series of event studies linked by a theoretical model. After two brief introductory chapters, there are three chapters describing in turn experiences of moderate inflations under metallic standards and under paper money standards and then hyperinflations. The next chapter presents a theoretical model of the inflation process and the final two chapters discuss the ending of moderate inflation and then hyperinflation.

The model in chapter 6 captures the core of the inflationary process that is uncovered in the earlier chapters. Begin with a small open economy that uses only coins and where the government is monetizing a constant real deficit in a stationary economy: At first, the paper money issued to finance the debt simply replaces coin in the public's hands with a more convenient medium and is a good thing; then the public has “enough” paper so that the paper money starts to substitute for reserves; when the reserves are depleted, nominal money stocks rise, prices rise (not quite as much) and the exchange rate peg is broken; finally, as inflation rises more rapidly the public stops using the paper money, switching to either barter or the good money (possibly ignoring government threats). At this stage, the government has an incentive to stabilize the monetary system.

Bernholz emphasizes the link between the economic and political aspects of inflation, and especially of any attempt to end inflation. Thus he emphasizes that moderate inflation is typically accompanied by undervaluation, generating a short-run boom that is politically popular, so that a stabilization may be unpopular, though such feelings can be mitigated if the reform entrenches an undervaluation, for example by fixing the exchange rate at an undervalued par, or is accompanied by adoption of a monetary constitution that in some way ties the hands of the monetary authority, for example by establishing an independent central bank, or a currency board, or by adopting a fixed exchange rate. In contrast, high inflation, by leading to a breakdown of the monetary system, creates an environment where there is broad support for reform, and a new government that stabilizes the currency will receive popular support. Bernholz suggests a very straightforward recipe for a successful stabilization which focuses on preventing the monetization of deficits and creating confidence amongst the population that this is the case.

A benefit of any study on such a broad canvas—from Roman times to the 1990s, across the globe—is that it is very likely to suggest unanswered questions, and this book suggests intriguing questions for monetary economists and historians. For example, an event study may tell us that all inflations originate in fiscal deficits, but it cannot tell us how often, and under what economic or political circumstances, fiscal deficits lead to inflation. Similarly, the observation that all hyperinflations lead to stabilization because the economy collapses cannot explain the diversity of maximal inflation rates in hyperinflations—the range is from 53 percent per month in Belarus in 1994 to 309 million percent per month in Serbia in 1994.

Today the media and indeed central banks seem to fear deflation rather than inflation, and it might seem that this book is out of step with the times. But as this chronicle makes clear, history is on the author's side: there have been many more disruptive inflations than deflations, and as the author's analysis of the nature of economic and political systems shows, this bias is likely to remain.