The Microeconomics of New Deal Policy
July 26-27, 2012
Shawn Kantor, University of California at Merced and NBER; Price Fishback; and John Wallis, University of Maryland and NBER
Kantor, Fishback, and Wallis seek to measure the importance of the New Deal in facilitating the Democrats control of the federal government from the 1930s well into the 1960s. From July 1933 to June 1941 (from fiscal year 1934 to fiscal year 1940), the federal government spent over $27 billion on programs cooperatively administered by state and local governments aimed at relieving the unemployed, building the nation's infrastructure, and supporting farmers' incomes, among other programs, which was more than the federal government spent in total prior to the onset of the Depression. The authors test whether long-term differences in the county-level electoral support for Democratic presidential candidates after the 1932 election can be attributed to the New Deal's interventions into local economies, and more narrowly whether voters rewarded Roosevelt from 1932-6 and from 1936-40 for his efforts to stimulate depressed local economies. Their results suggest that voters did reward Roosevelt for his early New Deal strategies to revive the economy: voters were much more likely to enhance their support for the president when their counties received relatively more relief and public works funding. However, this voter approval of the New Deal and the president seems to have waned as the Depression lingered throughout the decade; by the 1940 election, New Deal spending had very little impact on voter behavior. And, the authors find little evidence that Roosevelt's unprecedented policy interventions had a material influence in building solid electoral support for the Democrats after the 1932 critical election.
Robert Fleck, Clemson University
The Democratic Party's electoral success during the 1930s has long intrigued politicians and scholars. To gain new insight into the factors underlying that success, Fleck examines the striking heterogeneity in county-level support for Roosevelt. Even though the Depression's effects and the New Deal's benefits were famously widespread, only some parts of the country responded with large and durable partisan shifts. One reason is that several factors, including pre-New Deal economic hardship, Dust Bowl conditions, and New Deal spending, appear to have had effects that were largely transitory (fading by 1940). A complementary reason is that swing electorates can, and did, swing both ways: compared to other counties, those with a history of greater electoral variability experienced bigger swings toward Roosevelt in 1932 and 1936, and then bigger back swings in 1940. By contrast, several other variables notably economic and demographic factors discussed the previous literature are related to relatively durable shifts: controlling for Roosevelt's 1932 vote share (and other factors), Roosevelt's 1940 vote share was higher in urban and manufacturing-oriented counties and lower in farming counties and counties with large German-born populations. Finally, heterogeneity in marginal responses may have mattered greatly to national level Democratic success, because pre-New Deal economic hardship appears to have had a larger marginal effect on vote shares in the most populous areas (where most voters lived). By demonstrating which factors were transitory and which were more durable, illuminates the New Deal Realignment and, more generally, the influence of economic conditions and distributive policy on voter behavior.
Charles Calomiris, Columbia University and NBER; Joseph Mason, Louisiana State University; Marc Weidenmier, Claremont McKenna College and NBER; and Katherine Bobroff, Harvard University
Calomiris, Mason, Weidenmier, and Bobroff examine the effects of the Reconstruction Finance Corporations (RFC) loan and preferred stock programs on bank failure rates in Michigan during the period 1932-4, which includes the important Michigan banking crisis of early 1933 and its aftermath. Using a new database on Michigan banks, they use probit and survival duration analysis to examine the effectiveness of the RFC's loan program (the policy tool employed before March 1933) and the RFC's preferred stock purchases (the policy tool employed after March 1933) on bank failure rates. Their estimates treat the receipt of RFC assistance as an endogenous variable. They are able to identify apparently valid and powerful instruments (predictors of RFC assistance that are not directly related to failure risk) for analyzing the effects of RFC assistance on bank survival. They find that the loan program had no statistically significant effect on the failure rates of banks during the crisis; point estimates are sometimes positive, sometimes negative, and never estimated precisely. This is consistent with the view that the effectiveness of debt assistance was undermined by some combination of increasing the indebtedness of financial institutions and subordinating bank depositors. They also find that RFC's purchases of preferred stock which did not increase indebtedness or subordinate depositors increased the chances that a bank would survive the financial crisis. Their parallel analysis of the effects of RFC preferred stock assistance on the loan supply of surviving banks is inconclusive owing to the difficulty of identifying valid and powerful instruments (variables that predict RFC preferred stock assistance but which are not directly related to loan supply or loan demand).
Kris James Mitchener, Santa Clara University and NBER, and Gary Richardson, University of California at Irvine and NBER
Mitchener and Richardson examine how the Banking Acts of the 1933 and 1935 and related New Deal legislation influenced risk taking in the financial sector of the U.S. economy. They particularly focus on changes relating to the liability of bank owners, and assess how the removal of contingent liability influenced leverage before and after federal and state legal changes. Using a new panel data set of bank liability laws and bank balance sheets aggregated at the state level, theyy show that leverage ratios are higher in states with limited liabililty for bank owners. Banks in states with contingent liability converted each dollar of capital into fewer loans, and thus could sustain larger loan losses (as a fraction of their portfolio) than banks in limited liability states. The New Deal replaced a regime of contingent liability with stricter balance sheet regulation and increased capital requirements, shifting the onus of risk management from banks to state and federal regulators. By separating investment banks from commercial banks, the Glass-Steagall Act left investment banks to manage their own leverage, a feature of financial regulation that, in part, depended on their partnership structure. .
Nicolas Ziebarth, Christopher Vickers, and Mark Chicu, Northwestern University
Macroeconomists have long debated the aggregate effects of anti-competitive provisions under the "Codes of Fair Conduct" promulgated by the National Industrial Recovery Act (NRA). Despite this disagreement, there is only limited evidence documenting any actual effects at the micro level. Ziebarth, Vickers, and Chicu use a combination of narrative evidence and a novel plant-level dataset from 1929, 1931, 1933, and 1935 to study the effects of the NRA in the cement industry. They develop a test for collusion specific to this particular industry. They find strong evidence that before the NRA, the costs of a plant's nearest neighbor had a positive effect on a plant's own price, suggesting competition. After the NRA, this effect is completely eliminated with no correlation between a plant's own price and its neighbor's cost. They argue that this work provides some of the strongest evidence yet for the collusive effects of the NRA.
Jason Taylor, Central Michigan University, and Todd Neumann, University of Arizona and NBER
During the New Deal markets operated under a number of different institutional regimes, which were marked by executive orders, the passage of various pieces of legislation, and Supreme Court rulings on their constitutionality. Taylor and Neumann break the New Deal period into the following six regimes: the Hundred Days, the President's Reemployment Agreement, the National Recovery Administration Codes of Fair Competition, the Schechter era, the National Labor Relations Act, and the Fair Labor Standards Act. Under these various regimes, industrial markets were subject to different regulations relating to hourly wage rates, hours of work, production, and unionization. Using a VAR framework, the authors analyze data from 11 industries to examine the performance of employment, weekly hours worked, real wage rates, prices, and output during each regime. They use the residuals from these regressions to examine when "unexpected" shocks occur. They find that the timing of these shocks are associated with some of the regime changes, even after controlling for changes in fiscal and monetary policy. They believe that this suggests that the uneven pace of recovery during the 1930s can, at least in part, be explained by important policy and legal changes within the New Deal. For example, Roosevelt's first "Hundred Days" policies brought positive institutional change (financial reforms, devaluation), but the National Industrial Recovery Act slowed the nascent recovery.
Jonathan Rose, Federal Reserve Board, and Kenneth Snowden, University of North Carolina, Greensboro and NBER
The introduction of the direct reduction (fully-amortized) loan contract to the U.S. residential mortgage market is an important instance of financial innovation. Rose and Snowden describe the adoption of this contract within the building and loan (B&L) industry beginning in the 1880s and culminating in the 1930s. A long chain of complementary innovations at B&Ls gradually reduced the costs of adoption, leading to moderate use by the 1920s and potential for far greater use. The poor performance of traditional contracts during the crisis of the 1930s radically altered the adoption calculus. At this point, a new system of federal savings-and-loan charters incorporated many of the innovations that had been adopted within the small segment of the B&L industry that had introduced direct reduction lending by the 1920s. FSLIC liability insurance also accommodated the loss of credit risk sharing and mutuality inherent in older contracts. Direct participation in FHA loan insurance was not an important factor for B&Ls, although competition from FHA-insured loans by other lenders did encourage adoption. New Deal policies therefore built upon the ongoing process of financial innovation that brought the familiar, fully-amortized modern loan contract to the conventional residential mortgage loan market.
Trevor Kollmann, La Trobe University
Out of the programs created during the Great Depression, the introduction of public housing has been one of the most controversial. A myriad of recent papers have found that public housing is associated with higher crime rates and worse health outcomes in the modern era, leading to the belief that public housing was largely a failure. However, it is not clear that public housing projects prior to World War 2 were subject to the same outcomes as the high-rise projects constructed after the war. Housing conditions for low-income families during the 19th and the first decades of the 20th centuries were generally poor. Edith Elmer Wood argued that as many as a third of all families across the United States resided in dimly lit, crowded housing with many dwellings dilapidated and prone to fire and disease. Kollmann explores the effect of public housing on surrounding property values and contract rents for six cities in the United States between 1934 and 1940 using a series of hedonic models incorporating "locational heterogeneity", described in Páez, Uchida, and Miyamoto (2002). The dataset was constructed from a set of public housing projects for each city, combined with census tract-level data from the real property inventories and a special census of Chicago, and matched to the 1940 United States Census, as obtained from the National Historical Geographic Information System (NHGIS). The findings suggest that median property values largely rose in census tracts within a one mile radius as a result of public housing construction between 1934 and 1940. Moreover, there is evidence to suggest that public housing increased property values on the periphery of the cities, perhaps as a result of citywide slum clearance and the increasing segregation of low-income families into public housing projects. However, the estimates of the effect of public housing projects on median contract rents are mixed, both within and between cities.
Carl Kitchens, University of Arizona
Between 1930 and 1950, malaria rates in the southeastern United States went from infecting over 30 percent of the population to nearly zero. Several factors have been put forward as causes for the decline in malaria mortality and morbidity rates: increased public works targeting malaria, newly developed insecticides, and out migration of the rural poor to non-endemic areas. Kitchens focuses on two of the primary explanations: public works constructed by the Works Progress Administration throughout the 1930s, and the introduction of DDT following the conclusion of World War II. To estimate the effect of both the WPA programs and DDT spraying, he constructs a panel of county specific malaria rates from Georgia between 1932 and 1947. He finds that the WPA malaria campaigns are responsible for two-thirds of the decline in malaria during the 1930s and that the introduction of DDT after 1945 completely eliminated malaria in Georgia by the 1950s.
Briggs Depew, University of Arizona; Price Fishback; and Paul Rhode, University of Michigan and NBER
The policies of the Agriculture Adjustment Act and mechanization often have been held responsible for the rapid reduction of agriculture tenants and croppers (laborers paid shares of the crop). However, this conclusion has come with little empirical backing. Depew, Fishback, and Rhode estimate the impact of a AAA policy on the displacement of these workers in the cotton south. They motivate the empirical analysis by first analyzing a simple on-the-job search model, which shows how the cotton reduction program of the AAA incentivized landlords to displace tenants. The distribution of AAA funds was not exogenous, as they were distributed through local agents who may have been influenced by composition of farmers in the county. The authors construct an instrument based on the amount of funds that a county was supposed to receive under the AAA rules. Their results suggest that the AAA played a significant role in the displacement of black and white croppers and black managing tenants. These results are of particular interest, given that it was a violation of AAA contracts for landlords to displace these workers.
Douglas Irwin, Dartmouth College and NBER
Irwin explores the impact of the reciprocal trade agreement program and other New Deal initiatives on U.S. exports and imports. U.S. exports to trade agreement partner countries grew more rapidly than exports to other countries (an effect mainly evident in 1939 data, less so for 1938), whereas the sourcing of U.S. imports appears to have been unaffected. However, agricultural acreage reduction programs appear to have significantly reduced exports of important crops such as cotton.