Like the housing bubble of the 2000s, this bubble was fueled with debt provided by a government-sponsored enterprise: in this case, the Farm Credit System. As in the 21st century housing bust, the culpable GSE was bailed out by the government, which theoretically was not obligated for the GSE’s debt, but in reality was. With GSEs, the government always pretends that it and the taxpayers are not on the hook for their debt, but they always are.
One of the key factors in this bubble was the out-of-control general inflation of the 1970s, caused by the Federal Reserve under chairman Arthur Burns, whose hand was on the throttle, revving up the money-printing presses. Farmland was thought to be a good hedge against inflation, and it was better than that—until it was not. The shrivel of the farmland bubble in inflation-adjusted terms is much more dramatic: real farmland prices fell 38% from peak to trough, and in 1987 real prices were back to their level of 1972, 15 years before.
What did it seem like after the bubble turned into a shrivel in farmland prices and farm mortgage lending collateral? As one contemporary study, published in 1987, said, “Severe financial problems in agriculture have drawn a great deal of public attention.” (Of course, they have since been largely forgotten by the public, as is usual in financial cycles.)
“Land values have collapsed,” the study went on, “driving many farmers into bankruptcy and compelling foreclosure. The crisis has also heightened the dominant role played by the government in the market for agricultural credit.”
“The threat to agricultural lenders as a whole is enormous,” opined a 1986 article in the South Dakota Law Review. Failures of agricultural banks had increased rapidly by then. The threat also applied to the Farm Credit System, which in 1985 lost $2.7 billion, a record for financial institutions up to that time, and in 1985-86 had a two-year net loss of $4.6 billion, another record. One study reported, “Performing loan volume was dropping rapidly...nonaccrual loans were rising...interest margins were declining” and “the external auditing firm of Price Waterhouse [issued] a qualified opinion on the long run viability of the system.”
How perceptions, including those of regulators, change as one goes from bubble to shrivel. Thomas Stanton observes, “The reports of the Farm Credit Administration on the financial state of the Farm Credit System turned from rosy optimism in 1982 [at the peak] to a report of financial stress in 1984 and a request for taxpayer help in 1985” and then a major taxpayer bailout in 1987.
How had the former rosy optimism developed? As always, bullish thinkers found plausible ways to argue that the inflation of the bubble was based on fundamental developments. Farm income and prices had been rising strongly, and the price optimism was often based on global pessimism.
As discussed by Richard J. Herring and Susan Wachter, the investment literature of the time often displayed a “Malthusian Optimism hypothesis,” which “predicted that the demand for food would grow exponentially, but the supply of food would not...Thus, productive farmland in North America would become increasingly valuable.” This fit with the observation that “farmland prices had risen during the whole postwar era and...provided a rationale for ignoring earlier collapses in farmland prices.” Things were different this time! “The subjective probability of a collapse in farmland prices was treated virtually as if it were zero,” just as the probability of a house price collapse was treated a generation later.
Herring and Wachter go on to judge the Farm Credit System “the most aggressive lenders” in the bubble because they “increased their lending sharply as farmland prices began their ascent and continued to increase their lending three years after the peak.” Since the Farm Credit System is a GSE, “the creditors failed to discipline the risk-taking.” Farm Credit “placed a volume of bonds in the early 1980s second only to the U.S. Treasury,” based of course on the so-called implicit guaranty of the Treasury, which always turns out to be a real guaranty. “Holders of the bonds had little reason to monitor or price the risks taken”—and indeed that was true, for just like holders of Fannie Mae and Freddie Mac bonds in our own day, they were paid every penny of interest and principal due.
One of the most important government actions affecting the farmland bubble goes back to before it was inflated. “The real value of farm real estate began accelerating in 1972,” a 1985 study pointed out, but in the year before that, the Farm Credit Act of 1971 “allowed these banks to lend up to 85% of the purchase price of farm real estate, instead of the previous limit of 65%.” Thus, Congress promoted greater financial leverage in the farm sector, just as it did in housing, doubtless with good intentions.
"Real prices have been rising for 17 years, and if you are 40 years old now, you were only 10 when the last farmland bubble burst."
The farm credit crisis of the 1980s prompted legislation in 1985, 1986, and 1987. The 1985 Farm Credit Act Amendments set up a backstop line of credit for the Farm Credit System from the U.S. Treasury, making the link to the taxpayers clearer, created a “bad bank” to hold nonperforming loans, and restructured the Farm Credit Administration to be more of an arm’s-length regulator. But the crisis rolled on, according to one history: “farmland values continued to fall...losses continued to mount...[good] borrowers who were concerned about the security of their stock left the system.”
This led to the 1986 Farm Credit Act Amendments. Among these were two particularly interesting provisions: the Farm Credit System could defer provisions for loan losses, and it could also defer recognizing interest expenses on some of its high-coupon bonds. Such cooking of the books of course violated proper accounting standards and mirrored similar desperate accounting by insolvent savings and loans in the same decade—in the latter case, under the direction of the government in the form of the Federal Home Loan Bank Board. It did not work in either case.
This brings us to the Agricultural Credit Act of 1987. The bailout could no longer be postponed: the act accepted “the inevitability of Federal financial assistance to the Farm Credit System,” Congress concluded, and it committed $4 billion of taxpayer funds. A Farm Credit System Assistance Corporation was authorized “to issue federally guaranteed bonds and to purchase the preferred stock of system institutions...thereby funneling the federal ‘bail-out’ funds.” Among its other provisions, the act mandated and facilitated the consolidation of the component institutions of the Farm Credit System into fewer, bigger entities.
The restructured Farm Credit System continued as a GSE. Farmland prices, after five years of painfully falling, began to rise in nominal terms again in 1988. From that point, in inflation-adjusted terms they remained basically flat for six years, from 1988 to 1994. By 1994, it had been 12 years since the 1982 peak (and 13 years since the inflation-adjusted peak in 1981) and real farmland prices were still merely at their early 1970s level of more than 20 years before.
Then, in 1995, real farmland prices began a new ascent, with particularly rapid acceleration after 2004. The inflation-adjusted prices have far surpassed their bubble peak of 1981.
Of course, the bull market in farm prices looks even more dramatic when shown in nominal terms. These prices increases were only modestly reduced by the financial crisis of 2007-09 and have since accelerated. Real prices have been rising for 17 years, and if you are 40 years old now, you were only 10 when the last farmland bubble burst. If you are 35, nominal farmland prices have been rising since you were 10. How does anyone under the age of 40 subjectively assess the probability of a price collapse?
Where are we now? It is not quite clear. Is this a new bubble waiting to collapse? Or do these prices reflect some fundamental structural trends and new developments? Each time, it is the same debate.
In November 2011, the Federal Reserve Bank of Chicago held a conference on “Rising Farmland Values: Causes and Cautions” to address “the risks facing agriculture and the banking industry from rising farmland values.” Earlier that year, the Federal Deposit Insurance Corp. had a conference titled “Don’t Bet the Farm: Assessing the Boom in US Farmland Prices.”
Between 2000 and 2011, Farm Credit System mortgage loans grew at a compound rate of about 8%, with double-digit growth in four of those years (2001-02 and 2007-08). This is rapid credit growth, although it does not match the more than 11% compound growth of residential mortgage loans in the bubble years of 2000–07, of which five years (2002-06) were in double digits. Farm Credit System mortgage loan growth did slow to about 3.5% per year in 2010-11, then increased to an annualized 6.7% for the first half of 2012.
As always, the future is the “kingdom of uncertainty” and can be seen though a glass darkly, at best. As reported by Iowa Farmer Today in May 2012, “It’s tough to know if you’re in a bubble,” said the director of the Food and Agriculture Policy Research Institute at the University of Missouri. Economists from the university “gathered at a seminar recently to discuss ‘a possible bubble in farmland prices,’ but they all stopped short of actually calling the current situation a bubble.”
The new Financial Stability Oversight Council, a big committee of financial regulators, has published a 217-page annual report for 2012. This report devotes about a half a page to farmland prices, saying they are being “driven by increasing crop yields, rising commodity prices, favorable crop export conditions, and low interest rates.” The report adds, “Adjusting for commodity prices and improvements in crop yields, agricultural land values have retreated somewhat.” It has a graph that appears to argue that farmland prices are in line with the value of crop income yields.
So, not to worry? The report adds that “forecasts for production and demand are positive” and that “delinquency rates on real estate farm loans at commercial banks declined.” It does not mention that loan delinquencies always decline in a bubble.
We are in a remarkable period when it comes to interest rates. The Federal Reserve has committed itself to an extended period of exceptionally low rates, while it manipulates long-term interest rates lower through large purchases of bonds and mortgages. This is an attempt, among other things, to support house prices, but it has apparently succeeded in, among other things, helping stoke the record inflation of farmland prices.
Thomas Hoenig, the former president of the Kansas City Federal Reserve, now on the board of the FDIC, has pointed out the interest rate risk in farmland prices. “Cheap money [has] artificially boosted farmland values,” he argued. “Hoenig predicts when current low interest rates reverse and trend higher, land values will drop dramatically.” It is certainly true that when interest rates, and thus the discount rates for calculating present values, rise, all asset prices tend to suffer. Hoenig remembers the 1980s bust well: “In the Federal Reserve Bank of Kansas City’s district alone, I was involved in the closing of nearly 350 regional and community banks...Farms were lost, communities were devastated.”
So are we in another bubble or not? The director of the Center for Commercial Agriculture at Purdue University, Brent Gloy, had the following sophisticated reflections when grappling with this question:
“It is very difficult to understand when expectations are misinformed—compounded by the fact that farmland is an infinite life asset.” “Land values appear to reflect current high returns in agriculture [and are] not obviously overvalued.” “Values are dependent on interest rates remaining low and/or sustained growth in agricultural incomes.”
This is well hedged—a balanced and professional display of the difficulty of foresight.
How has farmland compared over the last 10 years with other investment categories sometimes also thought to be hedges against inflation? It is apparent that over the decade farmland prices have escalated far more rapidly and have been, unlike the other two, without a severe correction—so far.
Where does this leave us overall? The story of the farmland bubble and shrivel of the 1970s and 1980s is clear. Does financial history necessarily repeat? “Farmland values have risen sharply over the past five years, though comparisons to the late 1970s seem unfounded,” wrote the chief economist of the Department of Agriculture, Joseph Glauber, in 2011. “Increases,” he continued, “appear to be generally consistent with the rise in farm income and low interest rates.”
Looking at the housing market in 2004, the Federal Reserve Bank of New York concluded, “the marked upturn in home prices is largely attributable to strong market fundamentals: home prices have essentially moved in line with family income and declines in nominal mortgage interest rates.”
Making judgments about future risks of farmland prices involves considering plausible arguments about agricultural productivity, yields, exports, ethanol and other subsidies and incomes; interest rate forecasts; judgments about the psychology of expectations; and investment alternatives and strategies.
It is typical in such circumstances to have a debate about whether we are in a bubble or not. In 2005, the senior vice president in charge of risk policy at Fannie Mae, having considered the escalating risk factors in housing finance, declared, “No one can say a bubble exists until after the fact.” Yes, the future is always the kingdom of uncertainty.
Can the government do anything about this possible bubble? Well, it can try to analyze the relationship between farmland prices and underlying economic factors, as the Department of Agriculture and FSOC have done. It can hold conferences on the risks involved, as the FDIC and the Federal Reserve have done. It could stop subsidies to ethanol, which help inflate farmland prices. It could stop manipulating long-term interest rates through its central bank. But given the uncertainty involved, overall, there does not appear to be much it is likely to do. Besides, if this really is a bubble, it is already too late.