Inflation - Deflation?

Stephen Roach

The more I ponder the inflation story, the more I become convinced that we need to come up with a new approach. In two earlier essays, I addressed the shifting composition of inflation (see “Inflation Phobia” July 15, 2005) and the cross-border convergence of pricing (see “Inflation Convergence, July 18, 2005). In what follows, I explore some important shifts in the macro relationships that have long been at the heart of the inflation process. What emerges from this trilogy is a strong conviction that increasingly powerful forces of globalization have fundamentally altered the inflation outlook. Barring a setback to globalization or a major policy blunder, low inflation could well be here to stay for the foreseeable future.

Globalization is all about the cross-border integration of economies, markets, trade and financial capital flows, and information. Ultimately, it also entails increased mobility of the factors of production -- capital, labor, and technology -- thereby forcing us to think about the production process and the dissemination of services in an increasingly global context. That means the pricing of goods and services must also be examined in the same broader framework -- in essence, determined by the market-clearing balance between globalized supply and demand curves. The rapid expansion of global trade in recent years underscores the need to accept this analytical leap of faith in assessing inflation risk. By our calculations, global exports will exceed 28% of world GDP in 2005 -- easily a record and more than ten percentage points above the 17% share that prevailed as recently as 1986. As global trade continues to power ahead, the forces of globalization -- and their impacts on real economic and financial market activity -- can only intensify as a result.

It’s easy to be awestruck by the anecdotes of globalization; look no further than Tom Friedman’s latest best-seller, The World Is Flat (Farrar, Strauss and Giroux, 2005). As macro practitioners, however, we need to dig deeper. In particular, it is critical to assess whether cross-border integration has had a major impact on time-honored macro relationships that drive economic growth, employment, income generation, and inflation. I am very sympathetic to that possibility. I first explored such an outcome in the context of shifting global trends in employment and labor income generation (see my 5 October 2003 essay, The Global Labor Arbitrage). More recently, I have generalized this framework to include the cross-border arbitrage of saving and pricing (see my 6 June 2005 essay, The New Macro of Globalization). Some fascinating new research just published by the Bank for International Settlements adds further evidence to this debate. In particular, it sheds considerable light on how globalization is challenging the macro relationships that have long been at the heart of our understanding of the inflation process (see especially Chapter II of the 75th Annual Report of the BIS, June 2005). The BIS research provides a goldmine of evidence in the laboratory of globalization.

Three findings by the BIS strike me as especially noteworthy in revealing the impacts of globalization on inflation (see accompanying table): First is the link between exchange rates and import prices. Currency depreciation has long been perceived as an inflationary development. Unless foreign exporters are willing to compromise their profit margins, it makes sense for them to maintain price targets in home-currency terms -- thereby allowing external pricing to fluctuate with shifts in foreign exchange rates. While that’s still the case to some extent, BIS researchers have found that this relationship has become far less robust as globalization has taken hold. They compare this relationship over two periods -- the modern-day globalization era of 1990 to 2004 and the “pre-globalization” era of 1971-89. An examination of trends in six major developed countries -- the US, Japan, Germany, France, the UK, and Italy -- finds that the sensitivity (elasticity) of import prices to a one percentage point change in the nominal effective exchange rate has diminished sharply between the two periods. In the US, for example, the exchange-rate-import-price elasticity over the most recent 15 years was more than 60% below the elasticity of the preceding 20 years. Declines of varying magnitudes were also evident in the other five countries -- led by France and followed in descending order by Japan, the UK, Italy, and Germany.

Second, the BIS also finds that that the pass-through of import prices into the domestic price structure has been seriously curtailed as globalization has taken root. With the exception of Italy, all of the six countries examined have experienced a dramatic decline in the sensitivity between fluctuations in import prices and domestic prices in the past 15 years. I suspect this underscores the increased power of the global price-setting mechanism: Even if import prices rise due to currency fluctuations or market conditions in foreign economies, the lack of pricing leverage in a world with a hugely-expanded global supply curve now constrains domestic producers from passing through these higher external costs. In all six countries, this constraint has been evident in the form of reduced elasticities as globalization has taken off over the 1990 to 2004 period.

Third, there is also solid evidence of sharply diminished linkages between inflation and the broadest gauges of market pressures. This shows up in the form of reduced sensitivities between fluctuations in core inflation and changes in the so-called output gap -- the difference between actual and “potential,” or full-employment GDP. The UK experience is an exception to this trend, but sharp reductions in this elasticity were evident between the globalization and pre-globalization periods for Japan, France, Italy, the US, and Germany. Not surprisingly, this result fits well with equally-impressive declines in the linkages between unit labor costs and core inflation that I noted in the second installment of this trilogy (see my 19 July dispatch, “Inflation Convergence). If the broadly-based output gap has lost its potency in driving fluctuations in inflation, it stands to reason that a similar result can be expected from the linkage between inflation and the cost pressures that arise from cyclical fluctuations in the labor-market piece of the output gap.

Don’t get me wrong -- this is not ironclad evidence that globalization has repealed the macro rules of inflation. However, there can be no mistaking the evidence of a sharp reduction of the linkages between price setting and several of its key determinants -- namely, currencies, import prices, output gaps, and labor costs. It’s the timing of these diminished linkages that brings globalization into the story. For six major developed economies, the elasticities and correlations have declined during the same period when globalization has burst forth with extraordinary scope and speed. Maybe that’s just a coincidence. After all, there could certainly be other powerful forces at work. Central bankers would like you to believe that they deserve credit for their success as inflation fighters. In addition, the explosion of the Internet points to a new technology of price setting. These developments can hardly be dismissed as inconsequential events on the inflation front. But, in my view, they are dwarfed by the far more powerful market-driven forces of globalization. I do not think it is a coincidence that global inflation convergence has occurred at the same time when the roles of currencies, import prices, and labor costs have all diminished in importance in shaping inflation. Nor do I think it is a coincidence that these developments have all occurred during a period when global trade has soared repeatedly to new records as a share of world GDP.

At work, in my view, is the globalization of disinflation. Our old closed-economy models have been rendered increasingly obsolete by the emergence of far more powerful cross-border influences on pricing. As a result, in making inflation calls, we now need to pay less attention to country-specific shifts in unemployment and capacity utilization rates. Instead, we need to focus more on the global balance between supply and demand that shape the far more open models of globalization. In that broader context, the outlook for inflation remains very constructive, in my view. After all, it’s hard to have bottlenecks without a bottle.