EconSouth (First Quarter 1999)

  Research Notes

Research Notes highlights some of the research recently published by the Federal Reserve Bank of Atlanta. For complete text of these articles or papers on this Web site, see the links below.

Venture capital is drawn to the Southeast

Venture capital investment throughout the United States and in the Southeast in particular has grown dramatically in recent years. Edgar Parker and Phillip Todd Parker, in a recent article, examine the history, structure and evolution of the national venture capital industry and then focus on current developments in the Southeast, including state policies promoting such investment.

Pension funds, bank holding companies, insurance companies, investment banks and nonfinancial institutions all invest venture capital to pursue high returns and diversify investment risks. But returns from such investment have been mixed over the industry's relatively short history. As more investors pour more assets into venture capital and as state and local governments seek to attract this capital and the industries it fosters, the potential benefits will grow, but not without raising public policy issues.

The authors observe that venture capital, starting from a small base just a few years ago, has become integral to new business formation in the Southeast. They note that clear evidence on the impact of state venture capital support is lacking, and the role of public support of funds and projects may still be questioned. But, they conclude, technological advances, business opportunities and entrepreneurial needs should continue to spur development of the region's venture capital industry.


Data vintages affect forecast performance

The data on economic variables are usually estimates, and these estimates may be revised many times after their initial publication. Most historical forecast evaluation exercises are conducted using the latest available or most revised vintage of historical data — that is, using estimates that may have been unavailable to a forecaster in real time. Such evaluations thus can give a misleading picture of the forecast performance that can be expected in real-time situations. This fact is particularly relevant if a forecasting model's performance is compared to published real-time forecasts' performance.

A practical question is whether using the data set that was available to a forecaster in real time would lead to inferences that differ substantially from those made using the latest available vintage of data. A related question is whether it matters which vintage of data the forecasts are evaluated against.

In a recent article, John C. Robertson and Ellis W. Tallman argue that the choice of data vintage can have both a quantitative and a qualitative influence on forecast and model comparisons, at least over short horizons. This influence is illustrated by examining the performance of the composite index of leading indicators as a forecaster of alternative measures of real output. More research, however, is required to determine whether the results can be generalized to forecasts of other series that are subject to revision, such as the various money aggregate measures.


How does monetary policy affect the economy?

Many of the studies attempting to sort out how monetary policy affects the economy were inspired by the work of Milton Friedman and Anna Schwartz in the early 1960s. Friedman and Schwartz documented the correlation of monetary aggregates with both output and prices and tried to deduce monetary policy's effect from this correlation. James Tobin later cautioned against this practice, and empirical evidence has since established the unreliability of treating monetary aggregates as a monetary policy gauge.

In a recent working paper, Christopher A. Sims and Tao A. Zha offer critical views on the unreasonable assumptions in existing policy work. They contend that careful economic argument is needed about which assumptions can best identify monetary policy's effects.

The authors argue against the notion that models for policy analysis need not fit the data well and that well-fit models necessarily sacrifice economic interpretability. With the model they present in their paper, they show that it is possible to create a good economic model — useful for policy analysis — that fits the data.

Sims and Zha conclude that U.S. monetary policy in the postwar period has been systematically responding to the changing state of the economy and that the real effects of monetary policy are much less certain than is commonly believed.

JULY 1998


During the October-December 1998 period, the dollar declined slightly versus the 15 major currencies tracked by the Atlanta Fed. This decline followed a five-month rise that peaked in August 1998.

Note: The Atlanta Fed's Dollar Index was revised in January 1999, and the chart above is based on the revised index. See the article below for more details about changes in the Atlanta Fed's Dollar Index. For more detailed, monthly updates and historical data on the dollar index, see the Atlanta Fed's Dollar Index.

Atlanta Fed revises dollar index to include euro

The Atlanta Fed recently revised its monthly trade-weighted dollar index to include the euro, the currency of the European Monetary Union (EMU), as well as currencies in several Latin American countries and Malaysia.

Two key areas that have been revised are the trade weights used to calculate the index and the countries included in the index. The original Atlanta Fed dollar index used trade weights from 1984. In order to update the dollar index to reflect more current trading patterns, the revised dollar index is now calculated using average weights from 1995-97.

In addition, to accurately reflect the scope of the euro, five countries included in the first wave of the EMU are included in the revised index. These countries are Austria, Finland, Ireland, Luxembourg and Portugal. Brazil, Malaysia and Mexico have also been added, while Sweden is deleted. With these changes, the revised dollar index includes the United States' top 15 trading partners.

To maintain as much continuity as possible with the previous dollar index, the subindexes of the revised index remain largely the same. In the revised index, however, the Canadian subindex is renamed the Americas subindex to reflect the addition of Brazil and Mexico.

The other subindexes are the European, the Pacific, the Pacific-excluding-Japan and the classic. The European subindex includes the EMU, Switzerland and the United Kingdom. The Pacific subindex includes Australia, China, Hong Kong, Japan, Korea, Malaysia, Singapore and Taiwan. The Pacific-excluding-Japan subindex includes the countries in the Pacific subindex minus Japan.

To provide a basis for comparison with the original dollar index, a new subindex, called the classic, is included in the revised index. The classic subindex includes all of the countries in the original dollar index with the exception of Sweden. This subindex is calculated going back to 1973, giving it the same time range as the original Atlanta Fed index. The countries in the classic subindex are Australia, Canada, China, the EMU, Hong Kong, Japan, Korea, Saudi Arabia, Singapore, Switzerland, Taiwan, and the United Kingdom.

The methodology for calculating the dollar index is unchanged. However, the data set used to calculate the revised dollar index has changed, and the historical data set used to calculate the original index has been truncated. Data used to calculate the revised dollar index include information back to Jan. 3, 1995, and are indexed so that 1995 equals 100.

For information see the Dollar Index

Return to Index