THE PROFITABILITY OF NOTE ISSUE DURING THE FREE BANKING ERA

Howard N. Bodenhorn Michael J. Haupert
St. Lawrence University University of Wisconsin

Introduction

One of the remaining conundrums of the national banking era is the low rate of bank note issue by the national banks. Throughout the post bellum period national banks circulated at volumes considerably less than the legal maximum. This underissue is anomalous because, as Cagan (1963) and Cagan and Schwartz (1991) have shown, the national banks failed to exploit a profit opportunity. Profit rates for note issue ranged between 4 and 400 percent, and were apparently higher than alternative uses of the banks' funds. This anomaly continues to perplex economic and monetary historians.

The incentive to issue notes depended, of course, on their profitability versus alternative investments. Prices of state and federal bonds eligible for backing determined the profitability of note issue. The higher the price of the bonds, the less profitable the purchase of bonds as collateral for note issue. Although the prices of eligible bonds were above par throughout the postbellum era, Cagan and Schwartz (1991) show that profits often went unexploited. Even the most thoroughgoing students of the period can offer no better explanation than banks appear to have acted irrationally in their failure to exploit a profit opportunity (Friedman and Schwartz).

Our paper does not deal directly with the postbellum banking experience. Instead we offer a look at the issue of state bank notes during the free banking era in the hope that the free banking experience may shed some new light on the debate over the national bank conundrum. Our study focuses on four disparate free banking systems: New York, New Jersey, Illinois, and Wisconsin. While national banks are alleged to have issued - for whatever reason - too few bank notes, the state banks that preceded them are typically accused of having issued too many. To many, antebellum free banking remains synonymous with wildcat banking. Using the same methods developed to study the national banking era we are able to show that free banks of the earlier era were generally operated just as conservatively as their successors.

Employing the same sort of calculations as those used to gauge the behavior of national banks, we find that free banks, too, apparently failed to exploit all available profit opportunities. Where Cagan (1963) and Cagan and Schwartz (1991) find unexploited profit rates as high as 427 percent, we find that antebellum free banks failed to exploit profit rates on note issue from 17 to 2674 percent.

Free Banking and National Banking

It was the demise of the Second Bank of the United States which ushered in the free banking era. The expiration of the charter of the Second Bank dropped the sole responsibility for the nations banking system into the laps of the individual states. The response by some was a system of banking based on the Second Bank of the United States: a state bank with several branches; for others it was an experiment somewhat misleadingly referred to as "free banking." In all states that passed free banking laws, banking was never truly free. There were certain prerequisites to opening a bank, but anyone who could meet them was free to do so. In this respect it differed drastically from the system in which each bank needed a special act of legislation to get a charter.

President Jackson's veto of the recharter bill in 1832 numbered the days of the Second Bank of the United States. When its charter expired in 1836, the path toward free banking was paved. The first free banking act was passed in Michigan on March 15, 1837. At that time an identical act was being debated in the New York legislature. This act was subsequently passed, and in 1838 New York became the second state to establish a free banking system. The New York free banking system persisted with only minor changes until the National Banking Act rendered all state banking systems obsolete in 1863. Between the passage of the first free banking act in Michigan in 1837 and the passage of the National Bank Act, eighteen states passed free banking laws of some sort.

The death knell of state regulated free banking ended with the passage of the National Bank Act of 1863, and the subsequent legislation which taxed state bank notes out of existence. The initial act was pushed through by Salmon Chase, then Secretary of the Treasury, who needed an active market in United States Treasury securities in order to finance the war. By requiring all national banks to hold U.S. bonds as backing for their note issues, Chase was able to resuscitate the flagging market for treasury securities as well as monetize a large proportion of the government's debt.

The Data

The data used for this paper focuses on the returns available on bonds eligible to be used to back note issue. By focusing on the available returns we show that banks passed up significant profits to be earned on note issue by failing to purchase more eligible bonds.

The bonds that were eligible to back note issue allowed the bank to purchase the bonds and issue notes in quantities up to the face value of the bond, or the purchase price if the bond sold at a discount. The period in question is 1850 -1863 for the states of New York, New Jersey, Illinois, and Wisconsin. This period was chosen because of the availability of the data. Individual bank balance sheet data is reported in much greater detail beginning in 1850.

New York banks were originally allowed to hold bonds of any state approved by the comptroller as well as U.S. bonds and mortgages on property in the state of New York. By 1850 only New York and United States bonds paying or adjusted to 5% or better, along with a limited number of mortgages on New York land, could be used to back note issue. The notes could be issued against these bonds up to 100% of the lesser of par or market price.

In New Jersey, U.S., New Jersey, and Massachusetts bonds could be used to back note issue up to 100% of the minimum of par or market price. IN 1851 New York, Ohio, Kentucky, and Pennsylvania were added. In 1852 Virginia, 1853 Newark and Jersey City bonds, in 1857 Paterson bonds, and in 1858 North Carolina, Louisiana, Tennessee, Missouri, Iowa, Hoboken, and Hudson bonds were added, all at 100% of the lesser of market or par.

In Illinois, U.S. and all states which paid full interest on their bonds were eligible at 100% of the lesser of par or market, and Illinois bonds at 80% of the minimum, beginning in 1851. In 1857 the same bonds were eligible, but all of them were eligible only up to 90% of minimum. In 1861 only Illinois bonds were eligible, but at 100%.

In 1852 Wisconsin allowed U.S., state bonds paying full interest, Wisconsin, and railroad bonds to be used up to 100% of the minimum of par or market. In 1861 that was changed to only U.S. and Wisconsin bonds, which could be issued at 100% providing the bonds had sold above par for the preceding six months, otherwise at 90%.

Banks earned a profit by earning the coupon payments on the bonds relative to the amount of capital they actually had to tie up to hold the bonds after ultimately lending out the notes acquired by purchasing the bonds. The amount of capital tied up in note issue was equal to the price paid for the bond minus the amount of notes issued against the bond. Since most bonds sold at a premium, the amount of capital actually tied up was P-100 for a $100 par value bond. This return is calculated after subtracting a cost of issuing notes equal to $.0625. This figure was calculated by the Comptroller of the Currency for National Banks. These costs included the costs associated with producing and issuing bank notes. The rate of return a bank could earn by purchasing a bond and then issuing notes against it is calculated as: rate of return = (yield - .0625)/(P - 100).

The numerator is the yield to maturity in dollars on a $100 par value bond less the estimated annual cost of $.0625 of producing and issuing notes. The denominator is the amount of capital a bank had to tie up in order to issue the notes: the price paid for the bond less the amount of notes that could be issued against a $100 par value bond. This calculation assumes that the bonds sold at a premium, which was the usual case.

We have calculated the percentage rates of return that banks were earning on note issue by holding selected bonds. These returns ranged from 17 to 2674 percent. These calculations represent the foregone returns that banks could have been earning had they purchased additional bonds and issued notes against them. As Bodenhorn and Haupert (1991) showed, the banks were not forgoing these returns because they were already extended to their legal maximum (bank note issue could not exceed paid-in capital). Free banks issued notes on an average between 24% and 87% of their paid in capital during the decade before the Civil War. The notable exception to this was Illinois, which issued notes in excess of its paid-in capital four times during this decade.

We extended our study to the returns individual banks were actually earning on the notes they were issuing, given the bonds that they were holding to back these note issues. These returns were calculated in the same way as above. Our calculations include maximum and minimum portfolio returns. These upper and lower bound calculations were necessary because for some banks, the balance sheets did not specify which particular bond was held. For example, an entry might list a bank as holding Illinois 5s. If both the rate and maturity date were not given, then the bond could have been one of several different Illinois 5% issues, all of which may have different returns. For this case, since we know the issuer and coupon rate but not the year to maturity, the calculations were made using the Illinois 5% with the highest calculated return to calculate the maximum portfolio return, and vice-versa for the lowest portfolio return. For some of the bonds held by banks we didn't have any price data, so we could not calculate returns to note issue. When calculating bank portfolio returns we assumed a return of zero for all of these bonds in both the maximum and minimum calculations, biasing our results downward.

The New York data presents strong evidence that the New York free banks were not exploiting potential profits during the last decade of the free banking era. Maximum portfolio returns ranged from 19% to over 1000%, exceeding 100% on five occasions. The average minimum returns ranged from a low of 8% to a high of 108%, exceeding 10% in eleven of the twelve observations, and exceeding 20% five times.

The other states presented evidence of foregone profits that is nearly as compelling as that of New York banks. In New Jersey, maximum and minimum returns both ranged from 1% - 49%, exceeding 10% in half of the observations. In Illinois, the maximum and minimum returns ranged from 36% to 50%, with the maximum returns exceeding 50% four times in seven observations, and the minimum returns less than 47% only once. Finally, the Wisconsin returns ranged from 4% - 78%, with the returns exceeding 30% in two-thirds of the observations.

Why Free Banks Might Pass Up Profitable Note Issue

Why would banks ignore such obvious profits on note issue? Why would they choose not to purchase additional bonds and issue more notes when the returns were so great? A preliminary observation will be made here for the purpose of analyzing such behavior and indicating that it was not an anomaly, but rather a rational response by bankers given the banking environment of the time.

Students of the postbellum era have argued that banks failed to issue additional notes because: (1) they were irrational (Friedman and Schwartz), (2) there were costs involved in redeeming additional note issues not captured by the profit calculations proposed by Cagan (Goodhart), (3) banks found it more profitable to make loans by creating deposits rather than issue notes (James), or (4) banks feared a revocation of their note issue privilege, in which case they would be forced to absorb a capital loss, as the price of government bonds was determined largely because of their eligibility as collateral for note issue (Goodhart).

We find the first explanation implausible. Cagan and Schwartz have largely dispelled the second possibility. The most important of any hidden costs would have certainly been redemption costs, and they argued convincingly that redemption costs were not great enough to substantially reduce the calculated profit rates. A lack of data prohibits direct investigation of the redemption costs for the antebellum era, but, based on our knowledge of antebellum banking, we concur with Cagan and Schwartz. James's explanation also seems unlikely given our knowledge of antebellum banking practice. Deposits were still largely a new banking technology prior to the Civil War - especially in rural areas. Although city banks increasingly relied on deposits, deposit to note issue ratios during the antebellum period were always less than unity.

We are left then with the fear of revocation of the note issue privilege. Goodhart argued that bankers operated under a constant threat of having their note issuing privileges revoked. We find evidence of similar fears among free bankers before the Civil War. New York's free banks hung in a legal limbo from enactment of the free banking law in 1838 until passage of the national banking act in 1863 made the question moot. It was never resolved by the courts whether free banks were corporations or associations granted their charter by an administrative arm of the state government. If they were indeed corporations, their existence was unconstitutional. If they were associations, they were safe from constitutional challenge. A series of court cases in the 1840s failed to resolve the question. They were often regarded as corporations in the findings of lower courts, who only saw those decisions repeatedly overturned in appeals courts. All that was required to overturn their constitutional standing was an appeals court justice with a political ax to grind - and the banks knew it. Even as late as 1857, the state banking commissioner remarked that the free banks were "anomalous, objectionable, and of doubtful legality."

Such legal uncertainty no doubt placed the banks in an unsavory position. Their right to engage in a banking business was never completely clear or above question. Should the law receive serious challenge, the possibility of its annulment was very real. A finding that the law was unconstitutional would have forced the banks to unload large quantities of state debts on relatively thin markets and absorb large capital losses. It is not particularly surprising, therefore, that many banks chose not to invest the full amount of their capital in state securities. To do so would have threatened a much greater percentage of the stockholder's equity should the free banking law ever be revoked.

A desire by banks to remain liquid could also explain their failure to tap into available profits. This point would be closely tied to the previous one, that they feared capital losses by investing more heavily in bonds. In addition, they may have been looking to maximize their ability to redeem notes and stay in operation in case of a bank panic, which was not an unheard of phenomenon during the antebellum period. This would add to their long-run profit maximization scheme, and would certainly be an action that would not be taken by a wildcat bank. What would banks have to gain by sacrificing current, substantial, profits in order to avoid possible suspension of specie redemption in the future? The enhancement of their reputation by not having to suspend specie payments in the future could be very profitable. The passing-up of current profit opportunities is rational if the present discounted value of staying open, given the probability of a panic severe enough to cause suspension is greater than the current rate of return available on increased note issue. The fear of capital losses would be irrelevant to wildcatters, who intended to stay in business only in the short-run. Therefore to the extent that this explanation is true, bank behavior mirrored that of responsible banks, and did not resemble wildcatters.

Conclusion

Sharp distinctions are often drawn between the free banking and the national banking eras. This paper shows, however, that there may be striking similarities. The debate over the apparent "underissue" of bank notes by the national banks has yet to reach any consensus, and we do not directly add to that debate. Instead, we offer a new look at the free banking era using the tools employed in the debate over the latter period, as we believe it is time to move beyond the rhetoric of wildcat banking and judge the antebellum banks on the same grounds as banks in other, less emotionally charged, eras. Recent research has concluded that antebellum banks may have out-performed their successors in providing interregional intermediary services (Bodenhorn and Rockoff, 1990). It is quite possible that they may have been equally conservative - on the whole - as the national banks which followed. This paper is a step in the direction of refocusing the argument on more substantive issues.