This paper describes New Deal farm mortgage debt relief programs, implemented through the Federal Land Banks and the Land Bank Commissioner. Along with the Home Owners' Loan Corporation, the analogous program for nonfarm residential mortgage borrowers, these were the first large-scale mortgage debt relief programs in US history.
Farm mortgage debt relief is a relatively unexplored area of the New Deal.2 The goal of this paper is to set out some basic facts about these programs. These were the first large-scale mortgage loan modification efforts in US history, along with the Home Owners' Loan Corporation (HOLC), an analogous program for nonfarm residential home mortgage borrowers which has been the subject of some study in recent years.3 These Depression-era programs offer interesting policy precedents, as the principles of loan modification during the 1930s are still quite relevant in the modern day.
The Emergency Farm Mortgage Act, enacted in May 1933, set up two separate but tightly coordinated programs to address a rising wave of farm mortgage loan defaults. One program was run by the Federal Land Banks (FLBs)--a regional set of twelve private but government-sponsored farm mortgage lenders--and the other was run by their regulator, the Land Bank Commissioner (LBC). Together, the footprint of these programs was large, as about two-fifths of the nation's farm mortgage loans were owned by the FLBs and LBC at peak in the 1930s. Financially, funding came mainly from federally-guaranteed bonds, and the Treasury also provided the programs with significant direct subsidies to offset the costs of the modifications.
Table 1 summarizes the loan terms offered by each program. The FLBs modified all their existing loans to carry these terms, and offered refinancing at the same terms to borrowers at other lenders. The LBC, which had no preexisting loan portfolio, also offered refinancing, with a focus on debts that could not qualify for FLB loans because the debts exceeded the FLB's statutory loan-to-value cap or would be secured by junior liens. Table 1 shows that debt payment relief was primarily in the form of low interest rates--reduced to as low as 3.5 percent on FLB loans--and principal payment forbearance. New FLB borrowers also benefitted from the long durations of 30-40 years, but this was not a source of relief to existing FLB borrowers since their loans had always carried those durations. The LBC terms were a bit less concessionary than FLB terms, with the aim of encouraging borrowers to refinance with the FLBs if possible.
|Interest rate||4.5 - 5% (1933-35) 3.5 - 4% (1935-44)||5% (1933-37). 4% (1937-40) 3.5% (1940-44)||5% (1933-1939) 4.5% (1939 onwards)|
|Loan duration||36 years typically||13 years extended to 20+ in late 1930s||15 years extended to 20+ in late 1930s|
|Principal payment forbearance||Until July 1938||3 years from loan origination||Until June 1936|
|Loan-to-value limit||50% of land plus 20% of improvements||75% of land and improvements||80% of land and improvements|
|Appraisal methodology||"Normal" value||"Normal" value||"Normal" value|
|Lien limitation||First liens only||First or second liens||First liens only|
|Authorized lending period||1916-present||1933-1936 extended to 1947||1933-1936|
|Financial support from Treasury||Capital investment Cash subsidies to lower interest rates capital investments to cover principal forbearance Funds raised by federally guaranteed bonds||Capital investment Cash subsidies to lower interest rates Funds raised by federally guaranteed bonds||Capital investment Funds raised by federally guaranteed bonds|
Table 1 also compares both farm programs to the HOLC, the sister relief program in the nonfarm field. The terms of the farm loans were slightly more generous in some dimensions, made possible by significant cash subsidies and capital investments to the FLBs from the Treasury that were not paralleled in the HOLC. In addition, while the HOLC required that borrowers prove they were in distress, such as facing foreclosure, no such requirement appears to have been in place for the farm programs. The lack of such a requirement is particularly evident in the FLBs' extension of relief to all of their existing borrowers.
Despite the relatively concessionary terms on these loans, delinquency rates on FLB and LBC loans ranged from 20-30 percent in the late 1930s, particularly after the periods of principal forbearance ended. Ultimately, a lower fraction of these loans, about 9 percent (or 11 percent by value) ended in foreclosure, with the difference likely due to two factors: additional legislation in the late 1930s that further liberalized loan terms, and higher prices for farm products and land that buoyed many borrowers during World War II.
I elaborate on the institutional background and the history of the FLBs in section 2, describe the scale of lending activity by the FLBs and LBC after 1933 in section 3, and detail the exact terms of the loans offered by each program in section 4. Section 5 describes the outcomes of the program, in terms of delinquencies and foreclosures. Section 6 contains a calculation of the total discounted cost of the two programs to the Treasury, which I estimate amounted to about 6½ percent of assets. This is not a program evaluation, however, as I make no attempt to quantify the benefits of the programs. Finally, section 7 concludes with some thoughts on the principles evident in the design of these Depression mortgage relief programs.
Federal involvement in farm mortgage lending dates to 1916, when the Federal Farm Loan Act set up two systems of federally chartered lenders: the Federal Land Banks and the Joint Stock Land Banks (JSLBs).4 The goal of both systems was to provide affordable mortgage loans, in particular by offering amortization over long terms (30-40 years). Such long terms were not generally available from other lenders, which included mortgage companies, life insurance companies, commercial banks, and noninstitutional lenders such as individuals. The creation of two systems was a political compromise. The JSLBs were privately owned and competed with each other, while the FLBs were cooperatively owned by their member "farm loan associations."5 The FLBs were assigned non-overlapping regions of the country, and did not originate loans themselves but rather had exclusive correspondent relationships with their member farm loan associations, in an arrangement similar to the German landschaften system. Both the JSLBs and the FLBs funded their operations by the issuance of covered mortgage bonds, subsidized by their exclusion from federal income taxes.
Source: (clockwise from the top) Colliers, October 8, 1932, p. 12; Literary Digest February 4, 1933 p. 10; Literary Digest February 11, 1933, p. 8.
Both lenders encountered difficulties in the early 1930s, when reductions in farm land values and farm incomes (which are correlated by nature) led to a wave of farm mortgage loan defaults.6 Figure 1 shows some cartoons that convey farmers' deep discontent. These credit losses led to broad retrenchment by the portfolio lenders, and the complete collapse of the JSLBs after the federal government declined to bail them out, and in 1933 the JSLBs were restricted from issuing any new loans. The FLBs, in contrast, did receive a bail out through a capital investment from the Treasury, authorized by new legislation in 1932. Prior to this, the FLBs had been operating with the implicit--but not fully certain--backing of the federal government during the 1920s, reflected in the higher prices commanded by FLB bonds throughout the decade.7 Altogether, this institutional backdrop has striking similarity to the nonfarm residential mortgage market of recent years, in which portfolio lenders pulled back from the market, private mortgage securitizers failed, and institutions with implicit government support--Fannie Mae and Freddie Mac--were bailed out by the federal government.8
The capital investment by the Treasury in the FLBs during 1932 totaled $125 million, a significant amount equal to twice the FLBs' existing capital ($63 million) or about 9 percent of assets ($1.4 billion). This was the most significant act of the Hoover administration to address farm mortgage credit. (See the appendix for a list of major pieces of legislation regarding farm mortgage credit during the 1930s). Though these investments stabilized the FLBs, they did not require any specific forms of relief to the FLBs' borrowers beyond granting $25 million in principal forbearance. They also naturally did nothing to assist borrowers at other lenders. The collapse of the joint stock land banks also increased demands for refinancing opportunities.
As a result, there was great demand at the beginning of the Roosevelt administration for additional legislation. The Emergency Farm Mortgage Act was passed early in the Roosevelt administration on May 12, 1933, as part of the same law that created Agricultural Adjustment Administration. This act authorized the Land Bank Commissioner--which up to this point had simply been the regulator of the FLBs--to make direct loans to farmers.9 The act also authorized the FLBs to make direct loans without operating through their member farm loan associations, in areas where those associations were incapacitated. Ultimately, however, only about 10 percent of the FLBs' new loans were direct loans of this nature.10 Finally, the legislation dictated the interest rates, principal forbearance, and other terms of the loans that the FLBs and the LBC would make.11
3. Scale of loan activity
At peak in the mid-1930s, the FLBs and the LBC together held about 1.1 million loans totaling roughly $2.9 billion. These holdings represented about 40 percent of all outstanding farm debt by dollar value, as shown by Figure 2. Other important institutional lenders included commercial banks, life insurance companies, and joint stock land banks. The large noninstitutional portion of the figure includes mortgage companies, which were unregulated and therefore without much statistical coverage. Noninstitutional lenders also included a significant number of individuals, however.
Notes: Measured at year-end. Source: Saulnier, Halcrow and Jacoby (1958), pp. 159-161.
The FLBs already held about 400,000 loans in 1932, prior to the New Deal legislation in May 1933. Their holdings grew to roughly 640,000 loans at peak in 1935, as shown by Figure 3. The LBC held no loans prior to May 1933, but with a burst of activity its portfolio contained 450,000 loans at peak in 1937, and over the life of the program the LBC made about 680,000 loans. The LBC's lending authority finally expired in 1947 after having been repeatedly extended from its original expiration date of 1936.12 Prospective borrowers applied for loans through their regional FLBs, which acted as agents for both FLB and LBC loans. As a result, farmers submitted only one application form to apply for a loan from either organization, or from both. It was common for borrowers to have a first mortgage loan from their regional FLB and a second mortgage loan from the LBC. About 65 percent of LBC loans were secured by second mortgages.13
Notes: Annual year-end data. Source: Farm Credit Administration Annual Reports, 1932-1946.
Financing for these loans came from two sources. From May 1933 to April 1934, both programs purchased loans with cash provided by the Treasury or the Reconstruction Finance Corporation.14 After April, 1934, all loans were funded via the exchange of bonds issued by a newly created federal government corporation, the Federal Farm Mortgage Corporation (FFMC).15 That is, those selling loans to the FLBs or LBC received the FFMC bonds as the means of payment. Since FFMC bonds were fully guaranteed as to both principal and interest payments by the federal government, those receiving the bonds likely treated them as tantamount to Treasury securities.
Table 2: Debts Refinanced by Federal Land Banks and the Land Bank Commissioner,May 1933 to September 1934
|Type of debt refinanced: Mortgage debts held by||Percent|
|Life Insurance Companies||19.0|
|Joint Stock Land Banks||7.6|
|Type of debt refinanced: Other debts held by...||Percent|
Source: 1934 Federal Credit Administration Annual Report, Table 7. Farm Credit Administration (1959, p. 17) gives similar figures for the loans originated by the LBC alone after September 1934.
Under both the FLB and LBC programs, most of the new lending activity served to refinance existing debts held by other lenders. Of the loans written between May 1933 and September 1934, 91 percent was refinancing activity.16 These debts had been held by various lenders and tax authorities as detailed in Table 2. The distribution across these lenders broadly matches the amount of debt held by each class at the time.
Applications poured in quickly after the Emergency Farm Mortgage Act was passed in May, 1933. The large majority of applications were submitted from May 1933 to year-end 1935, when farmers submitted 1,068,267 applications, and 68 percent of these applicants were successful in obtaining a loan. Another 354,205 applications were submitted between 1936 and 1941, with almost the exact same acceptance rate. Application activity after 1941 was minimal, particularly for LBC loans, and the LBC's lending authority ultimately expired in 1947. Only individuals could borrow from either program.17
The acceptance rate on FLB and LBC loans was about two-thirds, which in comparison was higher than the HOLC's acceptance rate of about one-half. Unlike the HOLC, FLB and LBC borrowers were not required to necessarily demonstrate they were distressed, such as facing foreclosure. Indeed, all of the roughly 400,000 pre-existing FLB borrowers were charged lower interest payments and given principal payment forbearance without having to ask for it. No applications to the FLBs or LBC appear to have been rejected because of lack of distress, whereas 13 percent of HOLC applicants were rejected for that reason.18
Hervey (1933) provides a limited amount of information on the reasons given for rejected applications, up to the end of 1933, recapitulated in Table 3. The most common reason for rejection involved the valuation of the property itself. While Hervey does not mention negotiation with lenders, this property valuation issue likely reflects an inability to induce lenders to forgive enough debt for the loan to meet the loan-to-value limits. After all, a low property value is not a reason in itself to reject an application; it is only a problem if the proposed debt is too large relative to the value and the lender will not forgive any of it. The financial qualifications also likely included the ability of the farm. Otherwise, fundamental collateral and underwriting issues were the main reasons for rejections, including the suitability of the land for farm production, the maintenance of the land, and its general riskiness.
|Reason for rejection||Percent|
|Lack of necessary financial qualifications arising from the value of the property or the applicant’s net worth||37.5|
|Unsatisfactory physical condition of the property for farming purposes mainly in regard to equipment or state of improvement||23.5|
|Land poor or farm unfavorably located||9.3|
|Personal characteristics of applicant unsatisfactory||8.3|
|Applicant an inexperienced or inefficient farmer||4.5|
Notes: Data from Hervey (1933) covering a sample of applications during 1933. Legal restrictions include, for example, whether the property was technically a farm, or whether the loans were eligible under the various terms of the Federal Farm Loan Act.
4.1 Interest payment relief
FLB and LBC loans carried interest rates that were significantly lower than rates offered by other lenders.19 The average interest rate in 1928 on farm mortgage loans was 6.1 percent (Wickens 1932, p. 63), though there was significant regional variation, and President Roosevelt argued that interest rates "in many instances are so unconscionably high as to be contrary to a sound public policy."20 Figure 4 shows that, by the late 1930s, the average interest rate on FLB and LBC loans was below 4 percent, while rates on private loans still averaged above 5 percent. The difference in interest rates is particularly large given that FLB and LBC loans had much longer durations than commercial bank or life insurance mortgage loans. Two factors account for the relatively lower rates. First, the loans were financed by federally guaranteed bonds issued by the Federal Farm Mortgage Corporation (see section 6). As a result, funding costs were lower compared to other lenders, and these savings were passed on to borrowers through lower loan rates. Second, legislation ordered the FLBs and LBC to not collect contractual interest rates from borrowers, but rather collect lower interest rates, and the Treasury compensated them for the lost income. For example, in 1933 the interest rate collected from borrowers on FLB loans was 4½ percent; in that year, if an FLB borrower had a contractual rate of 5 percent interest, the borrower only paid 4½ percent and the Treasury paid the remaining ½ percent.21
Notes:The average rate for the FLBs and the LBC is measured using the reduced rates actually charged borrowers rather than the contract rates. Data for interest rates on loans made by commercial bank and individuals are missing for 1946. Source: US Department of Agriculture (1949).
Table 4 gives more detail on the contractual interest rates and the actual interest rates collected from borrowers, at year-end from 1933 to 1946. These interest rates were changed at various dates during the 1930s and 1940s as Congress passed additional legislation. LBC interest rates were generally higher than FLB interest rates, as an incentive to route borrowers through the FLBs if eligible for FLB loans. Likewise, interest rates on FLB loans through the normal channel of farm loan associations were lower than interest rates on FLB loans made directly to borrowers.
|Date||LBC: Contract rate||LBC: Rate collected||FLB -- through farm loan associations: Contract rate||FLB -- through farm loan associations:Rate collected||FLB - direct: Contract rate||FLB - direct: Rate collected|
Notes: Rates are as of year-end. Contract rates pertain to loans originated in each year, whereas the reduced rates pertain to all loans active in a given year. Source: 1946 Farm Credit Administration Annual Report, Table 2
The Treasury fully compensated the FLBs for the lost interest income--the difference between the contractual rates and the rates Congress ordered them to collect from borrowers--to the tune of $265 million from 1933 to 1944. Likewise, the Treasury paid $57 million to the LBC from 1937 to 1944. This subsidy was originally designed to last for only a few years after 1933 but was repeatedly extended (see the appendix for a list of legislation). The expense was large and President Roosevelt vetoed its extension in 1937 due to the cost. Nevertheless, Congress overrode the veto.22
4.2 Principal payment forbearance and forgiveness
All FLB and LBC borrowers were granted principal payment forbearance. For FLB loans, no principal payments were required before July, 1938. For LBC loans, no principal payments were required during the first three years of the loan, regardless of when the loan was originated. Roosevelt described this as a "temporary readjustement of amortization, to give sufficient time to farmers to restore to them the hope of ultimate free ownership of their own land."23
Financially, the FLBs were compensated by the Treasury for these delayed payments. This compensation was not in the form of a direct cash payment (as was the case for the interest rate relief) but rather through capital investment in each FLB in the amount of the forborne payments. In this manner, the FLBs did not take incur any financial cost from delayed principal payments, and the government's investment was designed to be paid back when the delayed principal payments were eventually made. The capital subscriptions cumulated to over $190 million by 1941 and more than doubled the federal government's capital investment in the FLBs from the $125 million invested in 1932 under Hoover. The Treasury made no similar arrangement with the LBC, perhaps because such a capital investment would have been meaningless since the LBC program was already entirely publicly owned.
Neither the LBC nor the FLBs forgave debt. However, some lenders voluntarily forgave debt on loans sold to the LBC and FLBs. There is only a limited amount of information on these reductions. From 1933 to 1935, the average debt reduction, across all FLB and LBC borrowers, was about 6-7 percent. This is quite similar to the experience of the HOLC, which also did not forgive any debt itself, but estimated that previous lenders forgave 7 percent of all borrowers' existing debts when selling loans to the HOLC.24 These aggregate figures obscure some underlying variation across borrowers, however, as most borrowers received no debt reductions, but those who did received nontrivial reductions. For example, in 1933, 18 percent of new LBC borrowers received debt reductions from their pervious lenders that averaged 23 percent of those borrowers' debts.25
4.3 Loan-to-value ratios and appraisals
The FLBs were limited from their inception to loaning a maximum of 50 percent of land value plus 20 percent of the value of buildings or other improvements. This was unchanged by the New Deal legislation. The LBC loan limit was higher, at 75 percent of the value of land and improvements together, in a deliberate effort to target borrowers with debts that exceeded FLB loan limits.
The loan-to-value limits for these farm loans critically depended on the appraisal methodology. The May, 1933 legislation specified that the appraisal methodology for FLB or LBC loans should rely on an estimate of the "normal" value of a piece of land rather than the present market value. The 1933 legislation defined normal value by the period before the war, from 1909 to 1914, avoiding the volatile agricultural prices that characterized World War I and continued through the 1920s.26 This is quite similar to the methodology used by the HOLC, though the HOLC policy was a choice by its administrators rather than a requirement of federal law (see Rose 2011).
FLB loan amounts were also restricted to a maximum of $50,000 though it appears the average loan was far lower, around $4,500. LBC loans could not be larger than $7,500.27 These maxima likely covered the large majority of farms in the country, as the average farm contained $5,554 in land and $2,169 in buildings according to the 1930 census.
4.4 Loan durations
The FLBs originated loans with long durations, typically 36 years, as they had since their inception in 1916. Of the loans made between May 1933 and September 1934, 78 percent were amortized over 30-40 years, not including the initial forbearance period which lasted until July 1938.28 These long durations delivered some relief to borrowers who refinanced with the FLBs, since those borrowers' previous loans generally carried much shorter durations (with the exceptions of borrowers from the joint stock land banks, which also wrote long duration loans). For example, a 1924 survey of lending practices by commercial banks, life insurance companies, mortgage companies, and others indicated that the most common loan duration was 5 years, and that durations longer than 10 years were very uncommon (USDA 1933, p. 19). Existing FLB borrowers were given no relief through this channel, though, since their loans were already written with these long durations. LBC loans had shorter durations, as most were initially written for 13 year periods, consisting of the 3-year forbearance period followed by a 10 year period during which the principal would be amortized.29 However, roughly 150,000 of the LBC loans were ultimately recast between 1939 and 1942 into longer terms, typically 20 years starting from the date of the recasting, as discussed in section 6.1.
5.1 Delinquencies and Foreclosures
Delinquencies were a serious problem on FLB and LBC loans during the 1930s. Figure 5 graphs the delinquency rates on FLB and LBC loans respectively from 1935 to 1942. These rates were elevated throughout the second half of the 1930s but fell rapidly during the war years.
The jump in delinquency rates recorded in 1938 is likely due to the expiration of the principal forbearance periods on most of these loans at that time. The forbearance period on FLB loans ended in July, 1938, and by year-end the delinquency rate on FLB loans reached 20 percent, up from an already-elevated 14 percent in the previous year. Likewise, the three-year forbearance period on LBC loans ended for the bulk of those loans in 1937 and 1938. As a result, the delinquency rate on LBC loans jumped to 28 percent at year-end 1938 compared to 22 percent in 1927.
Source: Farm Credit Administration Annual Reports, 1935-1944.
Foreclosures were ultimately significantly less numerous than these delinquency rates portended. In total, the LBC acquired or charged off 48,773 properties, or about 7 percent of all loans it originated, although by value the foreclosures were about 10 percent.30 It would be interesting to know the foreclosure rates on loans of different vintages, i.e. those originated in 1933 versus those originated in 1940, but unfortunately the LBC never published data that would allow such a calculation. It seems likely that the foreclosure rate was higher on the early loans originated between 1933 and 1935, since the largest numbers of foreclosure were completed in the late 1930s.
Surprisingly, the FLB foreclosure rate was a bit higher than the LBC rate, even though the delinquency rate had been higher on LBC loans and even though FLB loans had stricter underwriting standards, lower interest rates, and longer forbearance periods. Between 1934 and 1942, the FLBs acquired 77,713 properties, constituting a foreclosure rate of roughly 10 percent, and 11 percent by value.31 In comparison to the LBC numbers, it is not clear whether this number includes properties charged off but not foreclosed.
Three factors likely account for the improvement in credit quality during the 1940s. First, higher farm product prices during the war buoyed both farm income and farm land prices. Second, in general, servicing protocols do not appear to have called for aggressive pursuit of foreclosure. For example, the Farm Credit Administration (the federal agency overseeing the LBC) described the LBC foreclosure policy this way:
In view of the conditions which gave rise to the making of Commissioner loans and continuing agricultural difficulties, particularly in certain distressed areas, the Corporation has endeavored to administer its collection policies with due regard to the ability of farmer-borrowers to pay, in the belief that a constructive policy of assisting farmers to remain on their farms would, in the long run, best accomplish the Corporation's aims. (Farm Credit Administration, 1939 p. 73).
A third factor behind the improved credit quality is a set of further liberalizations in lending terms put into place in the late 1930s and early 1940s. One element was the lowering of interest rates (from the contract rate) on all LBC loans for the first time in 1937, by new legislation. Such relief had been afforded FLB borrowers since 1933. Another element became important in 1939, when the LBC offered all borrowers the opportunity to reamortize their loans to longer durations of 20 years, as most LBC loans had originally been written for 13 years. Between 1939 and 1942, 150,000 LBC borrowers--about one-third of all LBC borrowers at the time--exercised this option. This reamortization process mechanically decreased the delinquency rate on LBC loans since reamortized loans were considered newly originated loans. Many fewer reamortizations were put in place for FLB loans as most FLB loans were originally written with durations of 30-40 years.
Extensions of delinquent payments were another tool used to delay foreclosure. From 1940 to 1942, about 70,000 extensions were granted. Special forbearance arrangements were also used, in two varieties, depending on how payments were allowed to fluctuate with borrowers' abilities to pay. Those arrangements were put in place for five year periods.32 In the first plan, a farmer's required payments were fixed at a lower rate than originally required, but when farm income exceeded a "normal" amount (defined in coordination with their FLB lender), one half of the excess was devoted to loan payments. In the second plan, a farmer agreed to devote a certain share of their post-tax farm income to their loan payments, with any shortfalls being made up when income improved. This plan was somewhat more appropriate where it was difficult to define "normal" levels of income. Though officials repeatedly touted these options, in practice these plans appear to have been instituted for less than 25,000 loans (combining both FLB and LBC figures).33
Foreclosed farms were sold quickly by these farm programs, in contrast to the HOLC which held onto its foreclosed properties for many years in order to not depress property prices. The farm programs justified this practice by stating their desire to "get [the farms] back into the hands of farm operators" causing many to be sold "at prices that did not cover the total investment" (Farm Credit Administration 1959, p. 12). It is not clear why a rental program of foreclosed farms would not have accomplished the same goal.
6. Financial results
In this section, I calculate the net present value of the costs of the LBC and FLB programs to the Treasury in 1932, the date of the first investment in the FLB.
Government officials typically touted the Federal Farm Mortgage Corporation's dividends as evidence of the programs' profitability. For example, the Farm Credit Administration stated that "The Federal Farm Mortgage Corporation returned substantial earnings to the Government. The entire $200 million or original [Land Bank Commissioner] capital was repaid. Furthermore, a total of $145.9 million was paid to the Government as earnings or dividends." (1961 Annual Report, p. 33).34 The statements are technically accurate, but misleading by virtue of ignoring the larger financial picture. A more complete assessment of the financial aspects of these programs should incorporate time discounting, since it was not costless for the Treasury to raise funds in the early 1930s that were not repaid for a decade or more. (This is a conceptual shortcoming that also undermined government officials' statements about the FFMC's sister in the nonfarm field, the HOLC.) The statements also do not mention the large costs of the interest subsidies, which were larger than the dividends, and also wrongly imply that FFMC dividends were derived only from the LBC program and not the FLB loans as well. In general, it is necessary to analyze the finances of the two programs together, as the FFMC dividends reflected earnings from both the LBC and FLB loan programs.35
Before detailing the net present value calculation, it is important to note that this is not a program evaluation, as I have not made any attempt to account for the benefits that the programs delivered to the economy, via their impact on borrowers, lenders, or farm land markets. Instead, it is simply a discounted cash flow calculation from the point of view of an accountant in the Treasury. Another important caveat is that this calculation does not take into account the unrealized credit risk to which taxpayers were exposed by guaranteeing the bonds issued by the FFMC, or the implicit subsidies of that guarantee.
The Treasury's costs related to these programs involved both capital investments and cash payments to subsidize interest rates. The capital investments can be grouped in three categories: first, the Treasury invested capital in the FLBs in 1932; second, it capitalized the LBC operations in 1933; and third, it provided additional capital to the FLBs between 1933 and 1942 in compensation for delayed income from principal forbearance. The FLBs repaid their capital investments between 1940 and 1947, and between 1941 and 1947 the LBC repaid 99 percent of its capital investment. In addition, the FFMC issued dividends to the Treasury between 1948 and 1962.36
Table 5 displays the financial results of these capital investments, using three different discount rates: 0, 2, and 7 percent. With no discounting, the capital investments clearly returned a profit. However, since the costs of the capital investments were frontloaded but the revenues backloaded, discounting makes it difficult to claim the investments were profitable to the Treasury. A middle-of-the road discount rate of 2 percent shows the programs' capital investments as roughly breaking even, as the dividends just about compensated for the delay in returning the capital. A higher discount rate of 7 percent suggests a small cost, around 5 percent of assets.
|Capital investments:FLBs - general purpose||0||-29.1||-75.7|
|Capital investments:FLBs - principal forbearance||0||-28.2||-67.7|
|Capital investments:Dividends from FFMC||147.5||101.1||38.7|
|Capital investments:Total as a percent of assets||3.60%||0.10%||-4.90%|
|Interest Subsidies :LBC||-57||-48||-31|
|Interest Subsidies :FLBs||-264.8||-230.7||-163.6|
|Interest Subsidies :Total||-321.8||-278.7||-194.6|
|Interest Subsidies :Total as a percent of assets||-7.80%||-6.70%||-4.70%|
|Grand Total: as a percent of assets||-4.20%||-6.60%||-9.60%|
in supporting the LBC and FLB loan programs
Notes: Dollar figures are in millions of dollars, with negative amounts in parentheses. The d refers to the discount rate used in each column. Source: Farm Credit Administration, Annual Report, 1932-1962.
Table 5 also lists the costs of the Treasury's cash payments that subsidized lower interest rates, paid to the FLBs from 1933 to 1944, and to the LBC from 1937 to 1944. These cash payments were quite expensive, particularly the payments to the FLBs. A discount rate of 2 percent implies these payments cost about 6½ percent of assets. In contrast to the capital investments, higher discount rates map into lower costs for these interest rate subsidies, as the costs were not upfront like the capital investments but rather were spread out over time.37
Combining the capital investments and the interest subsidies together, a discount rate of 2 percent implies a cost of about 6½ percent of assets, all coming from the interest rate subsidies. A higher discount rate of 7 percent implies a large cost of about 9½ percent of assets, with the costs split about equally between the capital investments and interest rate subsidies.
The farm debt programs described in this paper were the first major mortgage debt relief efforts in US history, along with the contemporaneous Home Owners' Loan Corporation, which served nonfarm residential mortgage borrowers. These programs have some common features that suggest certain principles used by New Deal officials in designing these programs.
All of the Depression-era programs--the FLBs, LBC, and HOLC--owned or purchased the loans they modified and serviced them post-modification. By owning the loans, the federal government had the means to address the problem of continued delinquencies in the late 1930s. In addition, since the federal government guaranteed the FFMC and HOLC bonds used to finance these loans, it had the incentive to help borrowers avoid default if possible. Indeed, the actions taken to further lower interest rates and lengthen amortization periods during the late 1930s are quite important in understanding the nature of relief delivered by these programs. It is not difficult to imagine a counterfactual in which federal officials declined to double-down on the risky bets of acquiring these loans, and instead pursued an aggressive policy of foreclosure and liquidation on borrowers that redefaulted. Of course, these policies were not the only factor in reducing delinquency rates in the late 1930s, as the rise in land values during the war was also likely a major factor.
A second common feature at each of these programs was that debt forgiveness was not a common tactic by the government in delivering relief to borrowers. This was true both for the borrowers that refinanced with the programs, and for the existing borrowers at the FLB. Instead of debt forgiveness, these programs arranged for principal payment forbearance and for other relief mechanisms--some of which were costly, such as the interest rate subsidies. The lack of debt forgiveness reflects the era, as borrowers were much more likely to retain some equity even after land values fell, as loan-to-value ratios were usually capped at 50 percent on first mortgages and 80 percent on multiple mortgages. But New Deal officials appear to have made a broader strategic decision, betting that land values would rise again and working to produce that inflation with other policies, such as the Agricultural Adjustment Act. This strategic approach is reflected in the appraisal methods used by the farm programs and by the HOLC as well, all of which relied on long-run values rather than current market prices. Ultimately, land values did rise during the 1940s, though it is not likely that anyone in the early 1930s was clairvoyant enough to predict all of the events that contributed to that rise, such as World War II. Importantly, the approach to land appraisals went hand-in-hand with the reluctance by New Deal officials to liquidate the large amounts of delinquent and redefaulted loans in the FLB, LBC, and HOLC portfolios in the late 1930s, as those loans had been made on the basis of long-run values that had not yet materialized. Fundamentally, these Depression programs conceptualized credit risk along different lines than conventional, befitting loan portfolios largely comprised of troubled loans, and implemented servicing policies commensurate with that conceptualization.
Finally, the costs imposed on the Treasury by the farm programs were not trivial, largely as a result of the interest rate subsidies. Here, the farm programs differed with the HOLC, as no similar subsidies were offered to the HOLC's borrowers. I have two speculative hypotheses to explain this difference. First, while I have not engaged in any political economy analysis in this paper, there is reason to believe that farm mortgage borrowers were the greater priority of the Congress and the president. After all, the farm mortgage programs were proposed before the HOLC, enacted before it, retained authority to originate loans for a longer time period, gave less burdensome terms to borrowers, and cost more to the Treasury. Of course, it should be noted that President Roosevelt vetoed the costly enlargement of interest rate subsidies in 1937, but the override of his veto by Congress demonstrates the political difficulties in denying farmers relief of this sort. Second, the state of the farm market was different in 1933 than the nonfarm market. The 30-year loan had largely not arrived in the nonfarm sector, while the FLBs and joint stock land banks had been providing such loans to farm mortgage borrowers. As a result, the HOLC's relief was able to rely more heavily on extending amortization periods, a technique that was not as helpful to large numbers of farm mortgage borrowers who already enjoyed very long amortization periods. It is significant that interest rate reductions were a more important source of relief to farm mortgage borrowers, because lowering interest rates is a more costly method of relief than extending amortization periods. These farm programs, therefore, may be a warning from the Depression of the costliness of delivering effective relief to mortgage borrowers whose existing loans were relatively generously underwritten at the time, even if those loans would be considered fairly conservative by today's standards. A serious program evaluation is needed in order to determine whether the Treasury's costs were justified by the benefits delivered to borrowers, lenders, and farm land markets.
Alston, Lee J. (1983) "Farm Foreclosure in the United States During the Interwar Period," The Journal of Economic History, vol. 43 no. 4, pp. 885-903.
-- (1984) "Farm Foreclosure Moratorium Legislation: A Lesson from the Past," The American Economic Review, vol. 74 no. 3, pp. 445-457.
Case, H. C. M. (1935) "Farm Debt Adjustments," Journal of Farm Economics, vol. 17, no. 2, pp. 290-300.
-- (1960) "Farm Debt Adjustment during the Early 1930s," Agricultural History, vol. 34, no. 4, pp. 173-181.
Courtemanche, Charles and Snowden, Kenneth. (2011) "Repairing a Mortgage Crisis: HOLC Lending and Its Impact on Local Housing Markets," Journal of Economic History vol. 71, no. 2, pp. 307-337.
Farm Credit Administration. (various years) Sixth Annual Report. Washington: US Government Printing Office.
-- (1933) Loans to Farmers by the Farm Commissioner (pamphlet). Washington: US Government Printing Office.
-- (1946) Study of Federal Land Bank and Commissioner Loans Made Through 1945.
-- (1959) Federal Farm Mortgage Corporation: Its Operations and Achievements. Circular E-46. Washington: Farm Credit Administration.
Federal Home Loan Bank Board. (1938) Annual Report. Washington: US Government Printing Office.
Fishback, Price, Shawn Kantor, Alfonso Flores-Lagunes, William Horrace, and Jaret Treber. (2011) "The Influence of the Home Owners' Loan Corporation on Housing Markets During the 1930s," Review of Financial Studies, vol. 24, pp. 1782-1813.
Fishback, Price, Jonathan Rose and Kenneth Snowden. (Forthcoming, 2013) Well Worth Saving: How the New Deal Safeguarded Home Ownership. Chicago: University of Chicago Press.
Harriss, C. Lowell. (1951) History and Policies of the Home Owners' Loan Corporation. New York: National Bureau of Economic Research.
Hervey, George W. (1933). "Reasons Why Some Applications for Loans are Rejected." Washington: Farm Credit Administration.
Horton, Donald C., Harald C. Larsen, and Norman J. Wall. (1942) Farm-Mortgage Credit Facilities in the United States, United States Department of Agriculture, Miscellaneous Publication No. 478.
Jones, Lawrence A. and David Durand. (1954). Mortgage Experience in Agriculture. Princeton: Princeton University Press.
Preston, Howard H., and Victor W. Bennett. (1934) "Agricultural Credit Legislation of 1933," Journal of Political Economy, vol. 42 no.1, pp. 6-33.
Preston, Howard H. (1936) "Our Farm Credit System," Journal of Farm Economics, vol. 18 no.4, pp. 673-684.
Rose, Jonathan D. (2011) "The Incredible HOLC? Mortgage Modification during the Great Depression," Journal of Money Credit and Banking, vol. 43, no. 6, pp. 1173-1108.
Rucker, Randal R. and Lee J. Alston. (1987) "Farm Failures and Government Intervention: A Case Study of the 1930's," American Economic Review, vol. 77 no. 4, pp. 724-530.
Saulnier, Raymond, Harold G. Halcrow, and Neil H. Jacoby. (1958) Federal Lending and Loan Insurance. Princeton: Princeton University Press.
Skilton, Robert. (1943) "Mortgage Moratoria since 1933." University of Pennsylvania Law Review 92: 53-90.
Snowden, Kenneth. (2010) Testimony before the Senate Committee on Banking, Housing and Urban Affairs, "Covered Bonds: Potential Uses and Regulatory Issues."
-- (1995) "Mortgage Securitization in the United States: Twentieth Century Developments in Historical Perspective" in Anglo-American Financial Systems: Institutions and Mkarets in the Twentieth Century, eds. Michael D. Bordo and Richard Sylla. New York: Irwin.
United States Department of Agriculture. (1933) The Farm Debt Problem, 73rd Congress, 1st Session, House Document no. 9. Washington: US Government Printing Office.
-- (1949) Circular No. 821: Farm Mortgage Interest Charges and Interest Rates, 1940-48. Washington: US Government Printing Office.
Wickens, David L. (1932) Farm Mortgage Credit. United States Department of Agriculture, Technical Bulletin no. 288. Washington DC: USDA.
Woodruff, Archibald M. Jr. (1937) Farm Mortgage Loans of Insurance Companies. New Haven: Yale University Press.
Appendix: Farm debt relief legislation during the 1930s
Note: This appendix summarizes federal legislation that created the farm debt relief programs described in this paper. It avoids a general discussion of all aspects of these bills, which extended into many areas of the agricultural industry. In addition, various acts in the late 1930s and 1940s, not listed here, extended the periods of low interest rates on FLB and LBC loans.