In simple textbook descriptions, monetary policy actions have a uniform national effect. In reality, the nation is made up of diverse regions that are linked but that respond differently to changing economic circumstances. For example, the large declines in crude oil prices in the mid-1980s affected energy-producing regions very differently from energy-consuming regions. Indeed, the notions of a "rolling recovery" and of a "bi-coastal recession" have already entered the business vocabulary and suggest that the timing and perhaps the magnitude of ups and downs in economic activity vary across regions. The idea that monetary policy can have varied effects across regions is a short and logical next step. In fact, almost 40 years ago, Walter Isard, founder of the Regional Science Association, stated that "since each of [the nation's] regions has different resource potentials and confronts different obstacles to growth, it follows that monetary policies alone generate both retarding factors for some regions and problem intensifying factors for other regions."
Despite long-standing interest and concern about the issue, there is, at present, little empirical evidence on whether and to what extent monetary policy actions have differential effects on regional economic activity. Monetary policymakers have access to a lot of information about regional economic conditions. They acquire this information through periodic reports from regional Federal Reserve Banks (Beige Book reports) and through the regional business data gathered by the presidents of the 12 Federal Reserve Banks, who attend meetings of the Federal Open Market committee (FOMC). Additional information on differences in the impact of monetary policy across regions may aid policymakers in their consideration of regional developments in the formulation of national monetary policy.
Economic theory suggests at least three reasons why monetary policy might have differential effects across regions: regional differences in the mix of interest-sensitive industries, in the ability of banks to alter their balance sheets, and in the mix of large and small borrowers.
Regional Differences in the Mix of Interest-Sensitive Industries. Different industries respond quite differently to changes in interest rates. These different responses may interact with the different mix of industries across regions, and this interaction may result in differential regional effects of monetary policy. For example, changes in interest rates are likely to have profound effects on people's ability to buy houses and on businesses' willingness to acquire new structures. Construction accounted for almost 8 percent of real gross state product (GSP) in the Rocky Mountain region in 1980, but a little more than 4 percent of New England's real GSP. In addition, manufacturing, another interest-sensitive sector, accounted for just over 30 percent of GSP in the Great Lakes region in 1980, but less than 12 percent of the Rocky Mountain region's real GSP. Compounding these differences are interregional trade relationships, which can transmit localized responses differently across regions.
Regional Differences in the Ability of Banks to Alter Their Balance Sheets. Fed policy actions can have varied effects on different banks' ability to make loans. During periods of tight monetary policy when bank reserves are restricted, some banks can find alternative sources of funding for loans more cheaply and more easily than others, for example, by issuing large denomination CDs. Such banks' lending will be less sensitive to monetary policy changes. Anil Kashyap and Jeremy Stein propose that bank size largely explains differences in financing abilities: large banks have more funding options than small banks. Thus, regions in which a disproportionately large share of bank loans are made by small banks might respond more to monetary policy actions than regions in which a large share of loans are made by the nation's large banks.
One way that Kashyap and Stein define small banks is those with total assets at or below the 90th percentile. Alternatively, John Boyd and Mark Gertler classify a bank as small if its assets are less than $300 million. Since the asset size of the 90th percentile was just under $300 million in 1994, Kashyap and Stein's definition is equivalent to Boyd and Gertler's classification. Whether we look at all small banks or only small banks that are not members of a bank holding company, the regional distribution of loans made by small banks is highly unequal, suggesting that monetary policy could have differential regional effects.
The effect of the differences in regions' reliance on small banks will be diluted if bank-dependent borrowers can obtain credit from sources outside their own regions. However, there is evidence that banking markets tend to be segmented along regional lines. Craig Moore and Joanne Hill note that since banks can identify and monitor local investment projects more efficiently than banks and investors in other regions, it will be less costly for households and small firms to borrow from local banks.
Regional Differences in the Mix of Large and Small Borrowers. Regional differences in the proportion of large and small borrowers and the sources of credit available to each also could lead to different regional responses to monetary policy. According to the credit view of monetary policy, Fed actions affect economic activity by altering banks' ability to provide loans. To the extent that some borrowers are constrained to obtain credit from banks, monetary policy changes will substantially affect their ability to spend. Large borrowers usually have greater access to alternative, nonbank sources of funds, such as the issuance of corporate stocks and bonds or commercial paper. By contrast, small borrowers, such as individuals and small businesses, typically have banks as their sole sources of credit. Consequently, activity in a region that has a high concentration of small borrowers could be especially sensitive to changes in Fed policy.
The percentage of small firms (defined as regional firms with fewer than 500 employees) varies widely across regions. It ranges from a low of 66 to 67 percent in the New England, Mideast, and Great Lakes regions to a high of about 82 percent in the Rocky Mountain region.
While Fed policy actions can conceivably affect regional economies differently, little is known about the actual impact of Fed policy in different areas of the country. Some attention has been paid to the effects of monetary policy on region-specific banking flows as opposed to economic activity. Studies by Randall Miller and Peter Bias have found that Federal Reserve policy actions do affect regional banking flows differently. More typically, evidence has been collected about the effects of monetary policy actions on nominal income in particular regions. Results from these studies suggest that monetary policy has substantially different impacts in different regions.
However, this research contains a notable shortcoming. Existing studies measure the impact of monetary policy region by region without accounting for feedback effects among regions (i.e., monetary policy can directly affect the New England region, but because New England trades with the Mideast region, monetary policy indirectly affects the Mideast region and vice versa).
In an earlier study, we documented the importance of feedback effects among U.S. regions using a statistical technique known as vector autoregression (VAR). Our VAR included eight equations, one for real income growth in each region. For each equation, a region's real income growth depended on past values of its own and the other regions' real income growth.
By considering the system as a whole, rather than one equation at a time, the model allowed us to trace the effects of a change in a particular region's real income growth on real income growth in all other regions. For example, if income growth in New England rises, income growth in all other regions will be affected, since developments in New England will eventually affect other regions. Moreover, after the initial effect, continuing feedback effects on all other regions will occur, with the subsequent effects becoming smaller and smaller.
In a follow-up to that study, we used a VAR to estimate both the direct effects of changes in monetary policy on real personal income growth at the regional level and the spillover effects on income growth among regions. The variables in our model included real personal income growth in each of the eight major regions defined by the Bureau of Economic Analysis, the change in the relative price of energy (to account for the effects of oil price shocks), and the change in the federal funds rate (as a measure of changes in monetary policy). The study employed quarterly data for the period 1958-92.
A typical way to summarize the impacts of policy on personal income growth, and one that captures all interregional dynamics, is the cumulative impulse response. The cumulative impulse response shows how the level of real personal income in a region changes over time because of a monetary policy surprise. Monetary policy surprises are measured by unanticipated changes in the federal funds rate. For example, in fall 1994, Fed actions raised the federal funds rate 0.75 percentage point. Shortly before that time, forecasters were publicly predicting an increase of 0.25 percentage point. Thus, the additional 0.50 percentage point was a policy surprise or innovation.
We found that an unexpected one-percentage-point increase in the federal funds rate reduces real growth temporarily and, thus, leaves the level of real personal income below what it otherwise would have been for about two years. The model treats tightening and easing of the fed funds rate symmetrically, so that an unexpected cut in the funds rate temporarily raises real personal income relative to what it would have been otherwise. Since there are both unanticipated increases and decreases in the federal funds rate over time, we should not conclude that monetary policy lowers real personal income on average.
Interestingly, not all regions respond by the same magnitude. Several regions generally respond to monetary policy surprises with a magnitude and timing similar to those of the national economy. Specifically, the responses of income in five regions (New England, Mideast, Plains, Southeast, and Far West), called core regions, mirror the national response. In those regions, income ultimately falls about 1 percent (compared with what it would have been) subsequent to the one-percentage-point increase in the federal funds rate. The core regions accounted for 68 percent of aggregate 1980 gross state product (GSP) in the United States and for 70 percent of total U.S population.
Other regions (Great Lakes, Southwest, and Rocky Mountain), called noncore regions, show magnitudes of monetary policy effects quite different from the magnitudes for the national economy. The noncore regions accounted for 32 percent of total 1980 GSP in the United States and for 30 percent of U.S population. Personal income in the Great Lakes region showed the largest response to an unexpected increase of one percentage point in the fed funds rate, dropping about half again as much as income in the core regions. The Great Lakes region accounts for 18 percent of total GSP. In two other regions (Rocky Mountain and Southwest), personal income is much less responsive to an unanticipated one-percentage-point increase in the fed funds rate than income in the core regions, falling about half as much. Together these two regions account for 14 percent of aggregate GSP and 12 percent of the U.S. population. The Rocky Mountain region is the smallest, accounting for only 3 percent of aggregate GSP and only 3 percent of the nation's population.
After eight to nine quarters, real personal income begins to recover in most regions. Although the forecasted level of real regional personal income appears to remain below the level that would have existed in the absence of the unanticipated increase in the fed funds rate, we cannot conclude that the level of regional personal income will remain permanently lower.
We identified three ways in which monetary policy could have differential regional effects: regional differences in the mix of interest-sensitive industries, in banks' ability to adjust their balance sheets, and in the mix of large and small borrowers. How important are these factors in accounting for the different regional responses to monetary policy innovations?
To answer the question, we analyzed whether cross-regional differences in the size of real income responses are systematically related to variables capturing these three factors. Having so few observations makes it difficult to sort out the various ways in which monetary policy affects the economy of regions, but our findings are suggestive.
The interest sensitivity of a region's industries is likely to rise with the percent of a region's total gross state product accounted for by construction or manufacturing. Studies have shown that consumer spending on housing and manufactured goods, especially durable goods, tends to be interest sensitive. Spending on services, in contrast, tends to vary little with interest rates. Our analysis indicates that manufacturing-intensive regions are more responsive to changes in monetary policy than the more industrially diverse regions. This finding suggests that differences in interest-rate sensitivities are one reason for different regional responses. However, we do not find significant evidence that regions dependent on the construction industry have greater responsiveness to monetary policy initiatives.
The analysis also reveals that regions containing a large concentration of small firms tend to be more responsive to monetary policy shifts than regions containing small concentrations of small firms. This finding lends credence to the credit view of monetary policy, although we cannot distinguish between a bank lending channel and a broad credit channel.
Finally, we found that a region becomes less sensitive to an increase in the federal funds rate as the percentage of small banks in that region increases. This is inconsistent with the view espoused by Anil Kashyap and Jeremy Stein. One possibility for the inconsistency is that a bank's asset size may be a poor indicator of its ability to adjust its balance sheet to monetary policy actions.
Does monetary policy have differential regional effects? The answer is yes. Our research reveals two regions--the Southwest and Rocky Mountain--in which monetary policy has smaller effects on local economic activity than it has on the national economy and one region--the Great Lakes--in which it has a larger effect. The other five, or core, regions respond to monetary policy changes in ways that closely approximate the average response in the United States. The core regions accounted for more than 68 percent of aggregate personal income in the United States in 1980 and for 70 percent of the nation's population.
The existence of disparate responses underscores the complexity of conducting a national monetary policy for countries as large and diverse as the United States. We hope that information from studies such as this will allow regional data to better inform the national policy process.
Ahmed, Shaghil. "Does Money Affect Output?" Federal Reserve Bank of Philadelphia Business Review (July/August 1993).
Bayoumi, Tamil, and Eichengreen, Barry. "Shocking Aspects of Monetary Unification," Working Paper 3949, National Bureau of Economic Research (1992).
Beare, John B. "A Monetarist Model of Regional Business Cycles," Journal of Regional Science 16 (1976), pp. 57-63.
Bennett, Paul. "The Influence of Financial Changes on Interest Rates and Monetary Policy: A Review of Recent Evidence," Federal Reserve Bank of New York Quarterly Review (Spring 1990), pp. 8-30.
Bernanke, Ben. "Monetary Policy Transmission: Through Money or Credit?" Federal Reserve Bank of Philadelphia Business Review (November/December 1988).
Bernanke, Ben. "Credit in the Macroeconomy," Quarterly Review, Federal Reserve Bank of New York (Spring 1993), pp. 50-70.
Bias, Peter V. "Regional Financial Segmentation in the United States," Journal of Regional Science, 32 (1992), pp. 321-34.
Boyd, John H., and Gertler, Mark. "The Role of Large Banks in the Recent U.S. Banking Crisis," Quarterly Review, Federal Reserve Bank of Minneapolis (Winter 1994), pp. 2- 21.
Carlino, Gerald A., and DeFina, Robert. "Regional Income Dynamics," Journal of Urban Economics, 37 (1995a), pp. 88-106.
Carlino, Gerald A., and DeFina, Robert. "The Differential Effects of Monetary Policy Shocks on Regional Economic Activity," Working Paper 95-15, Federal Reserve Bank of Philadelphia (1995b).
Crone, Theodore M. "A Slow Recovery in the Third District: Evidence From New Time-Series Models," Federal Reserve Bank of Philadelphia Business Review (July/August 1992).
Garrison, Charles B., and Chang, Hui S. "The Effects of Monetary Forces in Regional Economic Activity," Journal of Regional Science, 19 (1979), pp. 15-29.
Garrison, Charles B., and Kort, John R. "Regional Impact of Monetary and Fiscal Policy: A Comment," Journal of Regional Science, 23 (1983), pp. 249-61.
Gertler, Mark, and Gilchrist, Simon. "Monetary Policy, Business Cycles, and the Behavior of Small Manufacturing Firms," Quarterly Journal of Economics, 109 (May 1994), pp. 309-40.
Kashyap, Anil K., and Stein, Jeremy C. "The Impact of Monetary Policy on Bank Balance Sheets," Working Paper 4821, National Bureau of Economic Research (August 1994).
Kashyap, Anil K., Stein, Jeremy C., and Wilcox, D. "Monetary Policy and Credit Conditions: Evidence From the Composition of External Finance," American Economic Review 83 (1993), pp. 78-98.
Mathur, Vijay, and Stein, Sheldon. "Regional Impacts of Monetary and Fiscal Policy: An Investigation into the Reduced Form Approach," Journal of Regional Science, 20 (1980), pp. 343-51.
Miller, Randall J. The Regional Impact of Monetary Policy in the United States. (Lexington: D.C. Heath, 1978).
Moore, Craig L., and Hill, Joanne M. "Interregional Arbitrage and the Supply of Loanable Funds," Journal of Regional Science 22 (1982), pp. 499-512.
Oliner, Stephen D., and Rudebusch, Glenn D. "Is There a Bank Lending Channel for Monetary Policy?" Economic Review, Federal Reserve Bank of San Francisco 2 (1995), pp. 2-21.
Peek, Joe, and Rosengren, Eric. "The Capital Crunch: Neither a Borrower Nor a Lender Be," Journal of Money, Credit, and Banking, forthcoming (1995).
Samolyk, Katherine A. "Banking Conditions and Regional Economic Performance: Evidence of a Regional Credit Channel," Journal of Monetary Economics, 34 (1994), pp. 259-78.
Thornton, Daniel L. "Financial Innovations, Deregulations and the "Credit View" of Monetary Policy," Review, Federal Reserve Bank of St. Louis 1 (1994), pp. 31-49.
Toal, William D. "Regional Impacts of Monetary and Fiscal Policies in the Postwar Period: Some Initial Tests," Working Paper, Federal Reserve Bank of Atlanta (June 1977).