Crisis and Calm: Demand for U.S. Currency at Home and Abroad from the Fall of the Berlin Wall to 2011

Abstract

U.S. currency has long been a desirable store of value and medium of exchange in times and places where local currency or bank deposits are inferior in one or more respects. Indeed, as noted in earlier work, a substantial share of U.S. currency circulates outside the United States. Although precise measurements of stocks and flows of U.S. currency outside the United States are not available, a variety of data sources and methods have been developed to provide estimates. This paper reviews the raw data available for measuring international banknote flows and presents updates on indirect methods of estimating the stock of currency held abroad: the seasonal method and the biometric method. These methods require some adjustments, but they continue to indicate that a large share of U.S. currency is held abroad, especially in the $100 denomination. In addition to these existing indirect methods, I develop a framework and basic variants of a new method to estimate the share of U.S. currency held abroad. Although the methods and estimates are disparate, they provide support for several hypotheses regarding cross-border dollar stocks and flows. First, once a country or region begins using dollars, subsequent crises result in additional inflows: the dominant sources of international demand over the past decade and a half are the countries and regions that were known to be heavy dollar users in the early to mid-1990s. Second, economic stabilization and modernization appear to result in reversal of these inflows. Specifically, demand for U.S. currency was extremely strong through the 1990s, a period of turmoil for the former Soviet Union and for Argentina, two of the largest overseas users of U.S. currency. Demand eased in the early 2000s as conditions gradually stabilized and as financial institutions developed. However, this trend reversed sharply with the onset of the financial crisis in late 2008 and has continued since then.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 43 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.