Home > Industry Analysis > Research & Analysis > FDIC: Kansas City Regional Outlook, Third Quarter 1999

FDIC: Kansas City Regional Outlook, Third Quarter 1999

Regional Perspectives Agricultural Sector under Stress: The 1980s and Today Today's economic environment differs from the one that led to the agricultural crisis of the 1980s. Interest rates are low and stable, farm debt levels are moderate, and real farmland values have been relatively stable throughout the 1990s. The Region's farm banks are reporting higher capital and loan loss reserve levels than in the 1980s, indicating that they can better absorb an increase in loan losses. Many farm banks have apparently increased their tolerance for risk by increasing lending volume, yet some maintain lower-than-average capital levels. Although conditions are different from those of the 1980s, several factors may pose significant risks to farm banks: continuing low commodity prices, higher loan levels, and uncertainty regarding the future of federal farm programs. The 1980s were marked by a turbulent agricultural economy that saw rapid declines in farm income and real estate values. This situation led to the failure of many farm banks,1 especially in the Kansas City Region. Map 1 shows the location of the 297 farm bank failures that occurred nationwide between 1977 and 1993. As the map shows, the Region was disproportionately affected, with 182 failures, 61 percent of total farm bank failures. Map 1 1 The Federal Deposit Insurance Corporation (FDIC) defines farm banks as FDIC-insured financial institutions that have at least 25 percent of their loans in agricultural production or secured by agricultural real estate. With prices for wheat, corn, soybeans, hogs, and cattle depressed again in 1999, many people are beginning to ask if the agricultural crisis of the 1980s is about to recur. This question has important ramifications for the Kansas City Region, because over half the Region's institutions are farm banks, and over half the nation's farm banks are located in the Region. This article compares the economic conditions that led to the farm crisis of the 1980s with those of today. While evidence indicates that the agricultural crisis is not about to recur, certain factors in the farm sector could cause serious problems for the Region's farmers and farm banks. Agriculture Prospered in the 1970s To understand the 1980s farm crisis, it is necessary to review the conditions of the 1970s, a decade of unprecedented agricultural prosperity. Chart 1 shows the path of farm income in the Region, restated in 1998 dollars. From 1972 through 1975, real net farm income in the Region reached levels never seen before or since and remained high through 1979. Chart 1 Export demand for farm products grew rapidly in the first half of the 1970s, when a number of trends converged to boost U.S. agricultural exports to record levels. Strong income and population growth among importing countries increased demand for U.S. products. At the same time, these importers gained access to external sources of credit, which, together with a weak U.S. dollar, improved their ability to import food and feed from the United States. Finally, a 1972 drought in the Soviet Union led to the enormous 1973 grain deal.2 2 Cochrane, W. The Development of Agriculture—A Historical Perspective. 1993. Minneapolis: University of Minnesota. p. 134. The early 1970s also witnessed tremendous technological changes as many farmers adopted improved machinery, chemicals, and fertilizers that led to rapid production increases. In addition, the federal government shared in the exuberance of the era, as the United States Department of Agriculture (USDA) encouraged farmers to produce as much as possible. It was during this period that Secretary of Agriculture Earl Butz made his famous appeal for farmers to "plant fencerow to fencerow."3 And they did. Between 1972 and 1981, the nation's wheat acreage increased from 55 million to 88 million acres, corn acreage increased from 67 million to 84 million acres, and soybean acreage increased from 47 million to 67 million acres. By 1981, planted acreage of these crops reached levels that have never been matched. 3 Peoples, K., et al. Anatomy of an American Agricultural Credit Crisis. 1992. Lanham, Md.: Rowan and Littlefield Publishers. p. 20. Inflation in the 1970s Led to a Negative Real Interest Rate Macroeconomic developments in the 1970s had far-reaching effects on the agricultural sector. Inflation (measured by the annual change in the Consumer Price Index) had been low throughout the 1950s and 1960s but began to increase somewhat as the nation's involvement in Vietnam and domestic spending for the Great Society programs led to budget deficits. Inflation spiked significantly in 1974, as the Federal Reserve Board attempted to soften the effects of increased energy prices caused by the Arab oil embargo. Inflation continued to escalate in the last half of the 1970s, reaching 13.5 percent by 1980. Chart 2 shows that interest rates charged by agricultural lenders rose significantly and steadily during the 1970s, from an average of 6.7 percent in 1970 to more than 11 percent in 1981. However, nominal interest rates, which reflected bankers' expected inflation rates, did not rise commensurately with actual increases in inflation. As a result, the "real" interest rate (calculated by subtracting the inflation rate from the nominal interest rate) was negative for most of the 1970s. This unusual situation enabled farmers to borrow today and repay tomorrow with fewer real dollars. Chart 2 Historically high farm incomes and the negative real interest rate that prevailed in the 1970s led to a large increase in borrowing by farmers, who increased the scale of their operations by purchasing land and machinery. The Farm Credit System, commercial banks, and insurance companies increased their levels of farm lending in the Region from $14 billion in 1970 to a peak of $57 billion by 1984 (see Chart 3, which shows farm debt adjusted for inflation to 1998 dollars). Chart 3 Farmland Values Rose to New Heights in the 1970s The 1970s also saw a significant escalation in farmland values (see Chart 4, which shows changes in farmland values, adjusted for inflation to 1998 dollars). The gradual upswing in farmland prices is evident for 1960 to 1972, reflecting long-run increases in economic returns to farming, largely because of technological improvements in the industry. Farmland prices then rose dramatically, by 2.4 times in the eastern states of the Region and 2.0 times in the western states between 1970 and 1981 (in inflation-adjusted terms), stimulated by strong farm income and the increased availability of credit. Rapid increases in land values improved farmers' equity positions, allowing them to continue borrowing and expanding. Chart 4 As Chart 5 shows, despite increasing debt, rapidly increasing land values allowed farmers to maintain low debt-to-equity ratios. For example, land prices in Iowa rose faster and to higher levels than in other states in the Region. As a result, the average debt-to-equity ratio of Iowa's farmers actually declined during the 1970s. In this environment, the level of lending seemed sustainable. Chart 5 Prosperity Unraveled in the 1980s The Region's farm income declined appreciably by the late 1970s, as the demand for farm exports began to subside. A combination of factors led to the drop in export demand: improved worldwide production that increased global inventory levels; a strengthening U.S. dollar that diminished the global competitiveness of U.S. products; the emergence of the European Community as an aggressive competitor in world grain markets; and enactment of the 1981 Farm Bill, which increased U.S. grain price supports and made U.S. products more expensive in foreign markets. As a result, in the first half of the 1980s the Region's farm income (in 1998 dollars) averaged only $6 billion annually, compared with $17 billion in the 1970s. Even these modest returns were possible only because of government payments (see Chart 1); excluding government payments, the Region's agricultural sector lost money during drought-ravaged 1983. In addition to farm sector problems, macroeconomic events worked against highly leveraged farmers in the 1980s. As seen in Chart 2, inflation declined dramatically in the 1980s, from over 13 percent in 1980 to less than 2 percent by 1986. In August 1979, the Federal Reserve Board instituted a fundamental change in monetary policy by targeting money supply growth rates rather than interest rates. This shift to a more restrictive monetary policy enabled the Federal Reserve Board to "wring out the demon" of inflation that had characterized the U.S. economy during the 1970s.4 4 Mussa, Michael. "U.S. Monetary Policy in the 1980s," in American Economic Policy in the 1980s. Edited by Martin Feldstein. 1994. Chicago: University of Chicago Press. p. 103. Nominal interest rates again lagged this relatively rapid change in the inflation rate. As a result, the relatively high nominal interest rates that had prevailed in the 1970s persisted into the early 1980s, leading to a rapid increase in the real interest rate farm borrowers paid. Increases in federal borrowing in the first half of the 1980s also contributed to the rise in the real interest rate. Declining farm incomes reduced farmers' ability to service large debts accumulated in the 1970s. The supply of available credit declined in the early 1980s as bankers became more cautious about lending to farmers. During this period, many farmers sold out or went bankrupt, and collateral values were often less than underlying debt because of the decline in real estate values. As debts were charged off from lenders' portfolios, the aggregate level of debt declined appreciably.5 As Chart 3 shows, by 1987 the real value of farm debt in the Region had fallen to 1960s levels. 5 Peoples, et al., p. 38. As Chart 4 shows, by 1987 real farmland values fell to mid-1960s levels and returned to the historical trend of gradual increase. The rapid decline in land values led to a sharp increase in farmers' debt-to-equity ratios (see Chart 5). As noted earlier, the effect was most pronounced in Iowa because that state's land values registered the largest percentage decline. Financial stress was widespread in the Region's agricultural sector during 1984 through 1986 (see Table 1). Iowa had the highest number of distressed farms (more than 12,000), while Minnesota and Missouri each had the highest percentage of stressed farms (24 percent). Missouri had the nation's highest percentage of insolvent farms (12 percent). The aggregate financial ratios only suggest the magnitude of the problem. A USDA survey of Iowa farmers conducted in 1984 showed that farmers with debt-to-asset ratios above 40 percent represented 28 percent of total operators in the state but held 65 percent of the outstanding debt. The most highly leveraged farmers (those with debt-to-asset ratios over 70 percent) represented 10 percent of operators but held 25 percent of the debt. Clearly, the most highly leveraged farmers were most at risk, as their equity positions declined with the fall in land prices. However, farmers' fortunes improved considerably between 1987 and 1990. Aggregate debt levels declined significantly as a result of an increase in loan charge-offs and farmers' attempts to reduce debt levels. According to one estimate, the share of dollar sales devoted to interest expense by financially stressed farmers declined from 25 percent in 1984 to 13 percent in 1990.6 6 Peoples, et al., p. 63. Farmers also benefited greatly from federal government payments during the last part of the 1980s. Payments to farmers averaged nearly $13 billion annually between 1986 and 1990, compared with an average of less than $3 billion per year in the 1970s. Current Difficulties in Agriculture Differ from Those in the 1980s The Region's agricultural sector is again experiencing financial distress. In June 1999, the USDA forecast 1999 national farm net income at $45.1 billion, the third consecutive year of decline. According to the USDA, large stocks of corn, wheat, and soybeans in domestic and international inventories point to continued low prices for these commodities through the year 2000 (see Table 2). A slight improvement in hog prices is forecast for 2000, but not to the 1997 level. Although the USDA forecasts an increase in cattle prices, forecasters have significantly overestimated cattle prices in each of the past two years. In addition, farmland values in the Region have shown the first indication of decline since 1986. In fact, Iowa State University's annual survey of land values showed a 1.8 percent decline in farmland values in Iowa for 1998. The Federal Reserve Bank of Kansas City reported a second consecutive quarter of declining farmland values in Kansas, Nebraska, and Missouri for the fourth quarter of 1998. The Federal Reserve Bank of Minneapolis survey of agricultural credit conditions reports a year-over-year decline in North Dakota land values. Low prices in the major commodity markets are expected to compound the financial distress of many farmers in the Region. Iowa State University economists simulated the financial effects on Iowa's farms if commodity prices remained at the 1998 level through 2000. Using financial data from 1,153 farms in Iowa, the researchers classified the farms into four categories of financial health: strong, stable, weak, and severe. Chart 6 shows the results of this study. Chart 6 Projected average net income declined approximately 60 percent, from $68,000 in 1997 to slightly more than $29,000 in 1998. In particular, net cash flow was negative for the farms in the weak and severe categories. The economists concluded that if the 1998 price levels continue through 2000, more than one-third of Iowa's commercial farms will require restructuring or liquidation. The results of the Iowa State University study may also apply to other states in the Region, such as Minnesota, Nebraska, and Missouri, that depend on similar commodities. As in the 1980s, a significant subset of highly indebted farmers is at risk today. Despite these difficulties, the level of risk in the agricultural sector is considerably lower than in the 1980s. Current debt levels, in real terms, are similar to 1969 levels, before the buildup of the 1970s. Similarly, land values have not increased as dramatically as they did in the 1970s, instead increasing with steady improvements in agricultural productivity. As a result, despite recent small declines, farmland values appear less vulnerable to the precipitous declines seen in the 1980s. Finally, low debt levels and stable farmland values have led to aggregate debt-to-equity ratios in line with those reported over most of the past 40 years. New Federal Farm Policy and New Problems The future of U.S. agricultural policy remains uncertain, and risks are very different from those that farmers faced in the 1980s. U.S. agricultural policy appeared to have shifted with the enactment of the Federal Agricultural Improvement and Reform Act of 1996, which dismantled the system of deficiency payments to farmers that had been in effect since the Great Depression. In return, farmers were promised a series of payments, unrelated to their production decisions, that would decline to zero by the end of 2002. "The Federal Agricultural Improvement and Reform Act of 1996 Increases Risks and Opportunities," Regional Outlook, Third Quarter 1998, argued that this legislation likely will negatively affect states with limited crop alternatives, such as the wheat-growing areas of North Dakota and Kansas. Developments since 1996 have blunted the impact of the intended reform. In 1998, in response to low commodity prices, Congress approved a $6 billion emergency aid package for agriculture. Senators from farm states have discussed reversing the 1996 reforms, but the likelihood of this occurring is unclear. Additionally, this November, World Trade Organization negotiations resume, during which U.S. agricultural policy will be scrutinized. Negotiators are expected to attempt to reduce domestic agricultural supports as part of a free trade policy. Farm Banking—the "Lag Effect" Masks Problems Before comparing farm banking in the 1980s with the present, it is important to understand the "lag effect." The "lag effect" describes the phenomenon whereby problems in the agricultural sector typically do not manifest themselves in farm bank performance measures for two to three years. For example, farm banks' reported conditions in the 1980s did not deteriorate significantly until 1984, three years into the farm crisis. The primary cause of this lag is the carryover debt process, in which loans not repaid in one season are carried over into the next season. The reason this process is more prevalent in farm lending than in commercial lending is that farm income tends to be volatile. In farming, it is common for one or two bad years to be made up entirely by a third healthy year. Other industries are more cyclical, and lenders tend to be less optimistic that the coming year will be strong enough to cover borrowers' current losses. The lag effect is heightened when farm equity levels are strong, because farmers will have more real estate equity to convert carryover debt to term loans. Farm Banking—the Crisis of the 1980s The 1980s farm crisis caused widespread problems for farm banks. Table 3 shows how certain financial performance ratios declined dramatically between year-end 1982 (the first year of depressed farm income) and 1985 (when farm bank financial performance bottomed out). Although 1982 farm income declined 22 percent from a year earlier and 43 percent from 1979, farm banks reported strong aggregate operating results in 1982. The lag effect masked the impact of the emerging farm crisis on farm banks. Equity capital ratios were relatively high, and loan loss reserves compared favorably with the 1 percent benchmark that existed at the time. The aggregate return on assets (ROA) ratio remained high at 1.23 percent, and few banks were unprofitable. Past-due loans were low at 2.8 percent. By 1985, the situation for farm banks had changed significantly. Their financial performance measures fully reflected the magnitude of the agricultural crisis. Past-due loans were up to 7.6 percent, and loan loss reserve levels were much higher to compensate for the increase in problem loans. The aggregate ROA ratio was just 0.29 percent, and more than a quarter of farm banks lost money. It is interesting to note that equity capital levels rose between 1982 and 1985 as farm banks with low levels of capital failed, increasing the aggregate ratio of the remaining institutions. Why Did Some Banks Fail, While Others Survived? Although aggregate farm bank financial performance declined dramatically between 1982 and 1985, all farm banks were not affected equally. As Map 1 shows, a significant number of farm banks in the Region failed in the 1980s. However, despite the agricultural crisis, 93 percent of farm banks in the Region did not fail. Map 1 What distinguished the failures from the survivors? A 1990 study by the Federal Reserve Bank of St. Louis7 attempted to answer this question. First, researchers noted that the majority of failures were in agricultural areas, but beyond that, the failures were not geographically clustered. This finding suggested that local economic conditions were not the primary cause of farm bank failures. In fact, the researchers found that most counties in which a failed farm bank was headquartered also included headquarters of other farm banks that reported sound financial results throughout the crisis. Few counties in agricultural areas experienced more than one farm bank failure between 1984 and 1986. 7 Belongia, Michael T., and R. Alton Gilbert. "The Effects of Management Decisions on Agricultural Bank Failures." American Journal of Agricultural Economics. November 1990. pp. 901–910. Note that the authors' definition of farm banks (those with a ratio of agricultural loans to total loans greater than the national average) differs from the FDIC's. Many studies point to management decisions as the primary cause of farm bank failures. Commonly cited qualitative characteristics of failed banks include relaxed underwriting standards, misconduct by bank officers, high tolerance for risk, and low involvement by bank directors. However, the St. Louis Federal Reserve Bank researchers attempted to determine if there were quantitative factors relating to management decisions that explained why some farm banks failed and others did not. The researchers created a statistical model8 that explained how changes in certain variables affected the probability of farm bank failure. They included only failed farm banks from counties where banks also survived, and they tested only financial ratios that management directly controls. 8 This model was a multiple regression analysis using 145 failed banks and approximately 600 surviving banks. Results shown in this article are for independent variables lagged three years prior to failure. This model best portrays the effect of management decisions before problems surfaced at their institutions. The results of the study were striking: The loan-to-asset (LTA) ratio was the most significant indicator of bank failure. The study found that for every 1 percent increase in a bank's LTA ratio, its chance of failure rose by 3.3 percent. Loans typically carry more risk than other bank investments; as a result, management that opted for increased loan volume may have had a higher tolerance for risk and followed more relaxed underwriting standards. FDIC researchers working on the History of the Eighties project also determined that the LTA ratio was the most significant indicator of farm bank failure in the 1980s.9 They found that farm banks with LTA ratios in the top 20 percent in 1980 failed at a rate of 6.2 percent, or five times the rate of other farm banks. Researchers conducted the test again in 1982 and found that banks with the highest LTA ratios had a failure rate of 10.3 percent, more than six times that of the remaining farm banks. 9 Federal Deposit Insurance Corporation, History of the Eighties—Lessons for the Future, Volume I. 1997. Washington, D.C. pp. 280–282. Concentration in farm loans was also a factor, but not as significant as the LTA ratio. For each 1 percent increase in the ratio of farm loans to total loans, the probability of failure increased 0.9 percent. Because the depressed farm economy was a driver of farm bank problems, higher concentrations in farm lending led to higher rates of failure. However, management decisions about lending volume were more predictive than concentrations in farm loans. Higher capital levels were the most significant factor preventing farm bank failures. For each 1 percent increase in a farm bank's equity capital-to-assets ratio, the institution's probability of failing declined by 13.4 percent. Although most farm banks would find it difficult to diversify their lending portfolios, management can control their institutions' overall risk profile by adjusting their lending volume and equity capital levels. Thus, the study suggests that management prudence in the 1980s helped some farm banks survive. The Present Situation: The 1980s All Over Again? Just as 1982, 1998 could represent the first year of a period of depressed farm income. Comparing the Region's farm banks at year-end 1982 and year-end 1998 illustrates similarities but important differences as well. Despite 1998's depressed farm income, farm banks reported good conditions at year-end. Table 4 shows that, in aggregate, equity capital levels were strong, earnings were high, and past-due loan levels remained low. Important differences are apparent between farm banks today and in 1982. The aggregate LTA ratio has increased significantly from 1982 levels. This fact could raise concern given the prospect for continued low commodity prices and the studies that identified the LTA ratio as a significant correlating factor with farm bank failures in the 1980s. However, farm banks are significantly better capitalized now than they were in 1982. Aggregate equity capital levels are up almost 2 percentage points compared with 1982. In addition, loan loss reserve levels are much higher, suggesting that today's farm banks are better positioned to handle an increase in loan losses. Although farm banks' aggregate ratio of farm loans to total loans has declined over the past 16 years, their aggregate ratio of farm loans to total assets has increased because they have a higher LTA ratio (see Table 4). This higher ratio indicates that farm banks are more concentrated in farm lending than they were in 1982 and may be more susceptible to a continued weak farm economy. In aggregate, management decisions in the 1990s have led to farm banks comparing negatively in some respects and positively in others with their counterparts of the early 1980s. Farm banks have a much higher aggregate LTA ratio than they did before the farm crisis, suggesting higher risk levels. On the other hand, the increased level of risk appears to be offset in the aggregate by higher capital and loan loss reserve levels. However, some individual farm banks in the Region appear to be aggressively increasing their lending volume relative to capital. Of the 1,355 farm banks in the Kansas City Region at year-end 1998: 312 have an LTA ratio over 60 percent and equity capital of less than 9 percent. 77 have an LTA ratio over 70 percent and equity capital of less than 8 percent. If the St. Louis Federal Reserve Bank study results are assumed to be applicable to the next prolonged agricultural downturn, these banks may be more susceptible to failure than those with lower loan levels or higher capital levels. Not surprisingly, the latter 77 institutions are in states reporting relatively strong farm income during the 1990s because bankers are more likely to manage their institutions aggressively in healthy economies. Iowa has 25 of the 77 and Nebraska has 18. South Dakota and North Dakota, which have had more troubled farm sectors, have a total of seven. These 77 banks' potential vulnerability to an agricultural downturn can be seen in their financial statements, which are already beginning to reflect the weakening farm economy. For example, the Region's 1,355 farm banks posted an aggregate ROA of 1.19 percent in 1998, just 5 basis points below 1997's aggregate ROA. By contrast, the 77 banks posted an aggregate ROA of 0.85 percent, down 28 basis points from their 1997 level. This reduction was caused largely by provision expenses necessary to cover increasing charge-offs in light of below-average loan loss reserve levels. Conclusion—No Recurrence of the Farm Crisis, but Problems Persist We do not expect to see a recurrence of the agricultural crisis of the 1980s, which led to the failure of many farm banks. Problems in the farm sector today appear different from those in the 1980s. The macroeconomy is more stable today, export growth has not led to overinvestment in the farm sector, and land values have not increased as dramatically as they did during the 1970s. On the banking side, farm banks are better capitalized and have higher levels of loan loss reserves than they did in the 1980s. Therefore, institutions are more capable of absorbing loan and operating losses. However, concerns persist. Most importantly, if commodity prices continue at their low level, highly leveraged farmers will be at risk of failure. As shown by the Iowa State University study, if low prices persist through 2000, one-third of Iowa farms could be at risk. Farms in parts of North Dakota and Minnesota, plagued by poor wheat yields and low prices over the past few years, would likely be affected more dramatically than the study suggests, as they began 1998 in weaker financial condition than farms in Iowa. In addition, the strong U.S. economy that has continued through nearly all of the 1990s has apparently increased some farm bankers' tolerance for risk, as shown by the aggregate increase in the LTA ratio. Farm banks that have relaxed their underwriting standards to achieve loan growth will likely be more vulnerable to continued low crop and livestock prices. Finally, the safety net for farmers is shrinking. While the political climate is uncertain, federal transition payments are scheduled to expire at year-end 2002. This event is likely to affect more seriously the Region's wheat growers, who do not have the planting choices of corn and soybean growers. Without federal aid, many farms in the Region may not be viable, and banks in the hardest-hit areas could experience serious problems. John M. Anderlik, Senior Financial Analyst Jeffrey W. Walser, Regional Economist Regional Outlook Information Return to Regional Outlook main page