Chapter 10 of 17 · Free Banking: Theory, History, and a Laissez-Faire Model by Larry J. Sechrest
Chapter 6 AMERICAN FREE BANKING
The previous chapter presented reasons why one should question Lawrence H. White’s popular portrait of Scottish free banking. Scottish banking seems to have been both less free and less successful than White’s work would lead one to believe. Here free banking in the United States will be examined. Once again, reasons will be given for rejecting the popular view, but in the other direction. American free banking will be shown to have been significantly safer and more stable than is usually believed to have been the case.
THE STANDARD STORY
The conventional viewpoint regarding the American free-banking period (1837–1863)1 has been that it was “a failure,” “chaotic,” rife with fraud in the form of “wildcat” banking, and a “hindrance to economic development” (Rockoff 1975, i). Bruce R. Dalgaard describes the time as one in which “significant numbers of unstable banks had an image of ‘fly-by-night’ operations out to obtain quick deposits before closing shop and disappearing with depositors’ funds . . . many banks failed . . . excessive creation of bank notes and unsound bank practices caused or aggravated financial panics during this period” (1987, 95–96). Rollin G. Thomas insists that the “system of free banking led to many abuses. Principally, it facilitated the development of wildcat banking . . . the situation was further disturbed by the fact that counterfeiting of bank notes became popular. The multitude of issues made the practice easy . . . the widespread adoption of free banking led to complete chaos in many areas” (1946, 244–46). Furthermore, Phillip Cagan opines that “the nation could not so easily have achieved its rapid industrial and commercial expansion during the second half of the 19th century with the fragmented currency system it had during the first half” (1963, 20).
The abuses allegedly perpetrated by some of these note-issuing free banks were indeed extreme. Charles J. Bullock claims that often “banks were located in inaccessible places, on some bottomless prairie road, or in the depths of forests, where it would prove as difficult as possible to find the ‘offices’ at which the notes were payable” (1900, 84). He adds that “in Ohio, Indiana, and Missouri, between 1855 and 1859, certain persons who presented notes for redemption were threatened with lynching or a coat of tar and feathers” (1900, 85).2 A. Barton Hepburn asserts in reference to the state of New York from 18383 to 1850 that “thirty-two banks failed, entailing a loss upon note holders of $325,487, some paying as low a percentage as thirty cents on the dollar” (1903, 137–38).
In summary then, the conventional critique of American free banking is that (1) the multiplicity of notes led to fraud, with consumers often accepting at face value notes that were redeemable—if at all—at a fraction of that face value; (2) short-lived “wildcat”4 banks established by unscrupulous persons were commonly encountered; (3) there were high rates of bank failure with concomitantly large losses to noteholders; (4) the system more likely hindered than fostered economic development via the allocation of financial capital; and (5) free banking was both highly inflationary and highly unstable, for example, conducive to indiscriminate bank panics that had negative spill-over effects on real variables.
As will be seen, each of these charges has been subjected to criticism by recent researchers. What emerges is a very different image. Free banking in the United States—though certainly blemished—was far less chaotic than has been believed. Moreover, the imperfections that did exist were largely the result of regulatory restrictions.
SOME BASIC FACTS
To set the context for the discussion that follows, several bits of information must be conveyed. First of all, what were the basic provisions of the free-banking laws? One must realize that these statutes were enacted at the state level; each state made slightly different stipulations. Two things were universal: a bank must redeem its notes in specie on demand of the noteholder and either state or federal bonds must be purchased by the bank and held as security for its outstanding notes.5 Failure to redeem its notes brought about closure of the bank and liquidation of its assets. Noteholders were meant to be protected further by being granted first lien on bank assets. Some states—but not all—also required that free banks have some minimum level of capital. For example, this was $100,000 in New York, and $25,000 in both Wisconsin and Minnesota (Rolnick and Weber 1982a, 21). Other states required a minimum ratio of specie to notes. This seems not to have been considered an essential item, however. New York imposed a 12.5 percent minimum reserve ratio in 1838 but repealed it in 1840, whereas Indiana eliminated a similar requirement in 1855 after having introduced it three years earlier (Rolnick and Weber 1982a, 21). Finally, although the precise conditions varied from state to state, free-bank stockholders faced limited liability. In some cases, this was twice the amount of their investment and might extend beyond the time they sold their stock.
Free banking was not permitted in all states. Some continued to require banks to petition the state legislature for permission to operate. That is, some states continued to require that a bank be chartered. This chartering system was one that had proven susceptible to extensive political abuse, however. Indeed, the favoritism and corruption associated with the chartering of banks was what had prompted the introduction of free banking in the late 1830s (Hepburn 1903, 136).
The extent of free banking may be indicated by the fact that, out of thirty-three states in the Union as of 1860, eighteen had free-banking statutes (Rockoff 1975, 3). Moreover, the expansion of American banking was largely due to the spread of free banking. In 1835, when all banks still had to have charters, there were 704 banks in the United States. In 1845, with two free-banking states, there were still only 707. By 1855—with fourteen free-banking states—there were 1,307 banks (see Table 3).
RECENT RESEARCH
Some of the most thorough and extensive modern research on American free banking has been that by Arthur J. Rolnick and Warren E. Weber. Their motivation is made clear: “The conclusion that when banking was left on its own it failed has very significant implications for regulatory policy and to a great extent influences policy today. Surprisingly, though, this conclusion is based on very casual empiricism” (emphasis added) (Rolnick and Weber 1982b, 10). As will be seen, they find “considerable evidence to question the prevailing views on free banking” (1982b, 10).
Failures, Longevity, and Noteholder Losses
In order to examine some of those “prevailing views,” Rolnick and Weber collected data from the state auditors’ reports for four free-banking states: New York, Minnesota, Indiana, and Wisconsin. These states allowed free banking for various time periods and, as a result, had various numbers of free banks that were formed under their statutes. New York had 449 banks (1838–1863), Minnesota had 16 (1858–1863), Indiana had 104 (1852–1863), and Wisconsin had 140 (1852–1863) (Rolnick and Weber 1982b, 19–20). This gives a total of 709 free banks. Of these, redemption information is available for 678, and the findings? “Close to fifty percent of the free banks in the states we investigated failed, but less than a third of these banks failed to redeem their notes at par” (Rolnick and Weber 1982b, 11).6 Overall, then, less than one-sixth of the free banks in these four states failed to redeem their notes at par value.
Table 3
Number of Banks in the United States, 1815–1860
| Year | Banks |
| 1815 | 208 |
| 1816 | 246 |
| 1820 | 307 |
| 1829 | 329 |
| 1830 | 330 |
| 1834 | 506 |
| 1835 | 704 |
| 1836 | 713 |
| 1837 | 788 |
| 1838 | 829 |
| 1839 | 840 |
| 1840 | 901 |
| 1841 | 784 |
| 1842 | 692 |
| 1843 | 691 |
| 1844 | 696 |
| 1845 | 707 |
| 1846 | 707 |
| 1847 | 715 |
| 1848 | 751 |
| 1849 | 782 |
| 1850 | 824 |
| 1851 | 879 |
| 1853 | 750 |
| 1854 | 1,208 |
| 1855 | 1,307 |
| 1856 | 1,398 |
| 1857 | 1,416 |
| 1858 | 1,422 |
| 1859 | 1,476 |
| 1860 | 1,562 |
Source: A. Barton Hepburn, History of Coinage and Currency in the United states and the Perennial Contest for Sound Money (New York: Macmillan, 1903) 118, 153, 174.
Notes: mean (1815–1849) = 630
mean (1850–1860) = 1,224
standard deviation (1815–1849) = 208.5
standard deviation (1850–1860) = 281.9
coefficient of variation (1815–1849) = 0.331
coefficient of variation (1850–1860) = 0.230
No data are available for 1817–1819, 1821–1828, 1831–1833, or 1852.
Coefficient of variation equals standard deviation divided by the mean.
Pursuing the issue of bank failures, it is interesting to compare free-bank rates of failure with failure rates of American banks during other time periods. For example, during the 1811–1830 period, there were 495 state-chartered/state-regulated banks. One-third of them failed (Gilbart 1967, 48–52), and this was largely a period of peace and prosperity for the country. Closer to the present day is the 1921–1929 period. This was a time of real economic growth, low unemployment, and low rates of inflation7 in which banking was regulated, there was a lender of last resort, and the Federal Reserve was undertaking expansionary open market operations. Nevertheless, out of 29,669 banks, 5,411 failed (Upham and Lamke 1934, 247). In addition, it is well known that in mid-1929, there were about 25,000 U.S. banks. By mid-1933, almost 10,000 of them were gone. It is ironic to note in the context of the foregoing that Hugh Rockoff has referred to a failure rate of 15 percent as “high” (1985, 886).
As for the longevity of free banks, it is clear that many economists share Rockoff’s belief that it was common for a free bank to last for only a month or two (Rockoff 1975, 8–9). Rolnick and Weber find little support for this. They conclude that “our data suggest that the conventional view is overstated. In our free banking states, only sixteen percent of the free banks were in business under a year, while the average number of years in business was over five” (1982b, 13). The longer-term banks were to be found in New York and Wisconsin.8 “New York free banks were in existence a mean of 7.9 years . . . and Wisconsin free banks were in existence a mean of 4.3 years” (Rolnick and Weber 1982b, 14). Furthermore, these figures have a downward bias, since they do not include free banks that, after 1863, continued in business as chartered state or national banks (Rolnick and Weber 1982b, 14).
Perhaps the most fundamental concern for individual consumers is the safety of those banknotes they choose to hold, and the usual portrait of free banking has been one in which such safety was ephemeral to say the least. What have Rolnick and Weber discovered? They “found that the most exaggerated views about free banks were those concerning the safety of the bank notes and the losses to noteholders . . . some have claimed that the losses to noteholders ran into the millions for individual states. Our data, however, tell a much different story” (1982b, 15).
Rolnick and Weber calculated two different measures of note safety. The first is the “expected value of a randomly selected bank note held until 1863,” which they arrive at by multiplying the note circulation of each free bank by the rate at which those notes were ultimately redeemed and then dividing that number by the total circulation of all free banks (1982b, 15). By this measure, the free banks in New York were the safest since their notes’ expected value never fell below $.97 per dollar. In Indiana, the lowest expected value was $.92 in 1853, and in thirteen out of fifteen other years, the expected value was $.98 or better (1982b, 15–16, 22–23). Minnesota fared worst. For the first six months of free banking in that state, the expected value of notes was no better than $.50 per dollar. However, over the succeeding four years, that is, until the end of free banking in 1863, the expected value did not fall below $.81.
Their other measure is an estimate of the total losses imposed on noteholders. This “was obtained by multiplying the last circulation for each bank which failed below par by one minus its redemption rate” (Rolnick and Weber 1982b, 16). They found that “the total losses to noteholders under free banking in the four states for which we have compiled data ranged between $1.6 and 2.1 million” (1982b, 17). In other words, there were significant losses in some cases, but the magnitude of such losses was far less than previously believed. It must be pointed out that Rockoff has challenged the validity of Rolnick and Weber’s work (Rockoff 1985). This, however, seems a weak challenge, since Rockoff offers very little data to support his claims.
Free-Banking Fraud
Very much a part of the traditional picture of free banking has been the allegation of widespread fraud. This allegation usually stems from what is a correct observation about the free-banking laws. It was true that such banks were required to buy state bonds (at market value) and deposit them with the state auditor as security for their notes, and then were allowed to issue notes in an amount equal to the face value of the bonds. The difference between the market and face values was often significantly large because of questionable state fiscal policies that led to a depreciation of the bonds. Supposedly, an unscrupulous banker would begin operations and, for a time, studiously redeem his notes on demand so as to foster public confidence in his bank. Then, when he had obtained an extensive circulation for his notes, he would suddenly cease redemption in specie—something made easy by locating his “office” in some remote area—and disappear. Where lay his gain? To the extent that he had persuaded consumers to exchange their specie for his notes at face value, he will have profited at their expense. This follows from the fact that said banker likely paid less than face value for the bonds. Therefore, his gain was the difference between (1) the dollars in gold he received in exchange for his notes and (2) the dollars in gold he expended in order to acquire the state bonds, less some small operating expenses.
This is essentially the explanation for free-bank failures offered by Rockoff (1975, 94–129). In effect, Rockoff contends that conscious attempts by bankers to defraud consumers were the source of most such failures. This kind of fraud was only possible when the bankers could purchase state bonds at market value but could issue notes based on the face value of those bonds. The obvious implication of Rockoff’s hypothesis is that free-bank failures should have occurred only in states that exhibited this market value/face value differential.
Rolnick and Weber disagree with Rockoff’s analysis. Their counterhypothesis is that free banks failed not because of fraud made possible by the counting of state bonds at face value, but rather because fluctuations in the market prices of those bonds often left free banks severely undercapitalized. They argue that this, coupled with the requirement that free banks redeem their notes and deposit accounts in specie, brought about the failures (Rolnick and Weber 1982a, 15, 19). The implication here is that free-bank failures should be most strongly correlated with declines in the market prices of the state bonds held—regardless of whether note issuance was based on market values or face values.
To test the two hypotheses, Rolnick and Weber offer data on bank failures and bond prices for four states: New York, Wisconsin, Minnesota, and Indiana (1982a, 22–30). They find little support for Rockoff’s position, since the majority of below par failures in those states occurred after the state laws were modified so as to restrict banks to note issues based on the market values of bonds held. For example, “over half of New York’s below par failures occurred after the market valuation restriction had been incorporated into its free banking legislation” (Rolnick and Weber 1982a, 15). Furthermore, “all 37 of Wisconsin’s below par failures occurred in 1860 and 1861 when the market valuation criterion was in effect” (Rolnick and Weber 1982a, 15). Half the Indiana failures were after the statutory change. Only the Minnesota data were ambiguous.
In contrast, they uncover substantial evidence that supports their theory regarding the source of free-bank failures. Plotting bond prices and bank failures, they observe that by “examining the data on below par failures during 1852–1863 for New York, Indiana, and Wisconsin, it can be seen that all but two of the 59 below par failures occur in” periods of substantial declines in the prices of bonds (Rolnick and Weber 1982a, 17). For Minnesota, the figure is two out of eleven below par failures (1982a, 25). Therefore, they conclude that free-bank failures were due not to fraud, but to the legal requirements that notes be backed by state bonds and notes be redeemed on demand in specie. The first imposed an unnecessarily risky portfolio on free banks; the second prevented free banks from passing capital losses on to noteholders (1982a, 19).9 Rolnick and Weber conclude the investigation with a provocative line of thought:
Our results also suggest that the so-called free banking acts were not really free. That is, this experience cannot be interpreted as a test of laissez-faire banking. . . . It is doubtful such a system would have evolved on its own. We suspect banking would have been a much safer business under a truly laissez-faire environment. (1982a, 19–20)
Notes as Mutual Fund Shares
A parallel issue regarding the likelihood of free-banking fraud is that of whether or not notes were typically accepted at face value by consumers. It would seem that the profitability of fraud is directly proportional to the frequency with which such acceptance occurs. The usual argument has been “that the public were either naive or misinformed. They accepted free bank notes at par even when it was apparent that the notes were risky” (Rolnick and Weber 1985, 34). Such gullibility is assumed, for example, by Rockoff (1975, 107–10).
Once again, Rolnick and Weber challenge the conventional argument. They suggest “that free bank notes were demanded because they were priced to reflect the expected value of their backing” (1985, 34). In other words, they hypothesize that some free banks essentially acted as mutual funds, with their notes treated like mutual fund shares, that is, discounted to reflect perceived risk.
It is intriguing to notice in light of this that several recent writers have proposed laissez-faire systems in which mutual fund shares might replace traditional hand-to-hand currency. These writers include Fischer Black (1970), Eugene Fama (1980), and David Glasner (1989, 195–200).
To examine this issue empirically, Rolnick and Weber work from a data set based on free banking in the same four states relied upon previously—New York, Indiana, Wisconsin, and Minnesota. In the first three, the “intended objective of the free banking laws was achieved: free bank notes there were relatively safe and circulated close to par” (Rolnick and Weber 1985, 7). They base this judgment on measurements of expected rates of return (1985, 8). In New York, the loss per annum was only 0.03 percent or less for fifteen out of eighteen years. For Wisconsin, losses were zero except for 1860 and 1861, when all of that state’s bank failures occurred. Indiana’s losses were nearly zero, except for 1853 and 1854, when they were 7.47 percent and 4.95 percent, respectively.
The experience in Minnesota was, however, radically different. Rolnick and Weber estimate the annual rate of loss there to range from a low of 12.43 percent in January of 1860 to a high of 54.38 percent for January of 1859 (1985, 10). They ascribe this difference from the other three states to a peculiar feature of banking in Minnesota. In that state, “railroad bonds” could be used as security for a free bank’s notes (1985, 13–16). These were technically Minnesota state bonds, but, in fact, they were backed only by the railroads’ questionable ability to repay the bondholders. The whole arrangement came into being as an indirect means by which the state legislature could help finance railroad development. The critical point is that these bonds were obviously risky and, thus, so too were any notes backed by them. To illustrate this, Rolnick and Weber point out that these railroad bonds could be sold in Chicago only when discounted by 50 percent and when banks holding these bonds failed, the redemption rates on the notes of such banks were no better than 21.25 cents per dollar (1985, 15).
Obviously, “what needs to be explained for Minnesota is why people were willing to hold notes that were backed by such risky assets” (Rolnick and Weber 1985, 15). Their answer is that such notes were both initially issued and subsequently traded below par. Essentially, then, these banks “were acting as mutual funds by intermediating the large-denomination ($1,000) railroad bonds, with the bank notes serving as shares in the mutual funds” (1985, 15). Rolnick and Weber emphasize that “the evidence from the auditor records, legislative testimony, and newspaper accounts about these banks” supports the idea that the public were well informed regarding the riskiness of the situation (1985, 26).
The lesson from Minnesota’s free-banking period is potentially very important. Here was a newly formed, frontier state with a largely rural populace. Despite this, it proved impossible to circulate free-bank notes at par value when they were backed by risky assets. This suggests that laissez-faire banking is not necessarily plagued by an informational asymmetry that encourages fraud. The significance of such evidence lies in the fact that many apologists for central banking base part of their case on the assumption that such asymmetries are universal and ineradicable. See, for instance, Charles Goodhart (1988, 33–34, 63–66).
Bank Panics
Here Rolnick and Weber try to answer two important questions: What is meant when the claim is made that banking is “inherently unstable,” and were free banks in the United States inherently unstable? They remark that there exists “no agreement on a precise definition of inherent instability in banking” (emphasis in original) (1986, 881). They opt for what they deem to be the conventional view. That is, they take inherent instability to mean that bank panics can occur without economy-wide real shocks. Such panics allegedly occur because of either “intrinsic” or “extrinsic” uncertainty (Rolnick and Weber 1986, 881–82). Intrinsic uncertainty arises when, as a result of some local real shock, the value of some banks’ assets falls and some noteholders demand redemption in specie. If information is distributed unequally among the noteholders, the differences between sound and unsound banks will not be clear, and a general run may result. Extrinsic uncertainty occurs when some pervasive, economy-wide real shock changes individuals’ demand for banknotes relative to specie. The trouble spreads as noteholders begin to doubt the solvency of the banking system.
Rolnick and Weber gathered data for the period 1841–1861 in order to test for instability. First they isolated clusters of bank failures, and noted the presence or absence of any local real shocks so as to answer the question of whether instability stems from intrinsic or extrinsic uncertainty. Their assumption was that, if a local shock coincided with a cluster of failures, such failures should be attributed to intrinsic uncertainty (1986, 883). Fully 85 percent of the failures appeared consistent with the intrinsic uncertainty proposition (1986, 885). They explain that these failures seem to have followed whenever free banks became grossly undercapitalized because of devaluations of the state bonds such banks held as security (Rolnick and Weber 1986, 884–85). These devaluations resulted from default on debt by the states, the onset of the Civil War, or, in the case of Minnesota (discussed earlier), the failure of several railroads.
To see if instability was inherent in free banking, Rolnick and Weber determine whether failures in one state led to failures in other states (the so-called contagion effect). This is the test suggested by Milton Friedman and Anna Schwartz (1963, 308). They find no evidence of contagion (1986, 885–86). Specifically, in the period July–December 1854, one-fourth of all the free banks in Indiana failed, yet only one out of 232 in New York failed and none in Wisconsin. Between June and September 1859, over half of the Minnesota free banks failed, but not one of the 390 such banks in New York, Indiana, and Wisconsin closed its doors. Between June of 1860 and June of 1861, more than a third of the Wisconsin free banks failed; only one free bank in New York and one in Indiana did likewise.
Rolnick and Weber suggest that the lack of contagious panics may be due to the informational nature of the bond requirement. For example, noteholders in New York knew that a drop in Indiana state bond prices would have no appreciable effect on the safety of New York free banks.10 They further propose that any regulatory structure that does not provide information on the market value of a bank’s assets will prove incapable of preventing bank runs (Rolnick and Weber 1986, 889).
The Allocation of Funds
As Rockoff has put it, a “fundamental question concerning free banking might appear to be whether permitting free entry produced a more efficient allocation of capital” (1975, 50). There are indeed several bits of evidence that seem to support the belief that free banking did produce such an allocation. For example, in the West (everything from Ohio to the Pacific Ocean), the source of the people’s enthusiasm for free banking was “that they perceived serious difficulties with allocations of bank capital achieved under state owned or state chartered banking systems” (Rockoff 1975, 50). The states in which such difficulties had manifested themselves included Indiana, Tennessee, Ohio, and Missouri (Rockoff 1975, 55–56).
Qualitive evidence like that just cited is not really sufficient. One would hope for quantitative data on the basis of which one might accept or reject the proposition of free-banking allocative efficiency. Rockoff suggests that one may assume such efficiency if there is evidence of competitive conditions.11 Furthermore, he unearths three kinds of data that may serve to indicate the degree of competitiveness: (1) profit rates, (2) the number of banks per 100,000 inhabitants, and (3) the number of unincorporated banks (1975, 57–64). In Kentucky and Indiana, he finds “high” bank profit rates prior to the introduction of free banking. More telling are the data for Ohio. There profit rates fall significantly as soon as free banking is adopted, suggesting a more competitive market. Figures for bank-to-population density are quite ambiguous. The critical factor seems to be the degree of urbanization rather than the presence of free banking, since “this measure of competition is highest in New England and lowest in the South and West” (Rockoff 1975, 60). The numbers of private (unincorporated) banks are generally greater in free-banking states and lower in the others, suggesting that the level of competition was greater in those free-banking states.12 Rockoff tentatively concludes that “New York City benefited from free banking in its competition with Boston and Philadelphia for financial leadership, and that in the West free banking was, or at least appeared to be, a way of slowing the vexatious problem of how to allocate bank capital in a region of new settlement” (1975, 67).
Noteholder Safety
The degree to which a holder of free-bank notes experienced the risk of loss is, of course, one of the important issues regarding free banking. Earlier it was seen that noteholder losses may have been much smaller than was previously believed to be the case (Rolnick and Weber 1982b). Confirmation of such results may be found in the works of A. Barton Hepburn and Charles Conant.
Hepburn lists the total losses to noteholders in the state of New York from 1838 to 1861 as $398,336 (1903, 137–38). Conant helpfully adds that the total note circulation in New York during that time period was approximately $314 million (1927, 375). This represents an average loss per dollar held of a minuscule 0.127 percent. New York did indeed have one of the best free-banking systems—of that there is no doubt. Nevertheless, such small losses seem utterly incongruous in the context of the usual image of free banking. Moreover, this just deals with nominal balances. Over the 1838–1860 period, the purchasing power of each dollar in the United States rose, on average, by 15.6 percent.13 In other words, in New York noteholders on average experienced a net gain in terms of real balances, despite the losses resulting from free-bank failures. For a modern comparison, one may look at the 1968–1990 period. Federal Reserve noteholders suffered losses in purchasing power of almost 75 percent over that span,14 and that does not take into account the enormous cost to taxpayers of closing insolvent financial institutions.
Pursuing the issue of free banks’ safety, one must ask what factor might be of primary importance to the typical noteholder. Rolnick and Weber are largely concerned with assets held in the form of state bonds. They suggest, as seen earlier, that when the market value of such bonds fell, free banks tended to fail. However, state bonds surely represented secondary reserves to free banks. Their primary reserves consisted of outside money, that is, specie. It was their ability to redeem notes (or demand deposits) in gold or silver coin that was the first line of defense against a bank run.15 Therefore, an examination of reserve ratios should shed some light on the question of free-bank safety.
Table 4 presents data on the reserve ratios for the American banking system for the period 1834–1860. Only four states had adopted free banking before 1850—one in 1837,16 two in 1838, and one in 1849. Thirteen additional states embraced free banking between 1850 and 1858, whereas one (Pennsylvania) did not permit free banking until 1860. Therefore, two subperiods might be recognized. From 1834 to 1849, there was a specie standard and banks often issued their own notes,17 but a bank had to be chartered by the state in which it operated. From 1850 to 1860, freedom of entry and exit was added to the idea of note-issuing banks on a specie standard. This was the heyday of free banking.
Table 4
American Bank Reserve Ratios, 1834–1860 (all banks)
| Year | Reserve Ratio |
| 1834 | 15.48% |
| 1835 | 22.83 |
| 1836 | 14.11 |
| 1837 | 13.52 |
| 1838 | 16.64 |
| 1839 | 20.59 |
| 1840 | 18.04 |
| 1841 | 20 |
| 1842 | 18.87 |
| 1843 | 28.96 |
| 1844 | 30.54 |
| 1845 | 25.11 |
| 1846 | 21.03 |
| 1847 | 18.72 |
| 1848 | 20.56% |
| 1849 | 21.02 |
| 1850 | 18.59 |
| 1851 | 17.3 |
| 1852 | 15.48 |
| 1853 | 17.28 |
| 1854 | 16.5 |
| 1855 | 14.5 |
| 1856 | 14.68 |
| 1857 | 12.9 |
| 1858 | 22.34 |
| 1859 | 23.83 |
| 1860 | 18.12 |
Sources: William Graham Sumner, A History of American Currency (New York: Greenwood Press, 1968) 188. Amasa Walker, The Science of Wealth: A Manual of Political Economy (New York: Kraus Reprint Company, 1969) 162.
Notes: mean (1834–1849) = 20.38%
mean (1850–1860) = 17.41%
standard deviation (1834–1849) = 4.62
standard deviation (1850–1860) = 3.14
coefficient of variation (1834–1849) = 0.227
coefficient of variation (1850–1860) = 0.180
As can be seen, the average reserve ratios for the two subperiods do not differ greatly. They are 20.38 percent for 1834–1849 and 17.41 percent for 1850–1860. Also, one may note that the year-to-year variations (as measured by the coefficient of variation) are smaller for 1850–1860. One might protest, however, that these time-series data combine both free banks and chartered banks. Perhaps a cross-sectional approach would be preferable.
Table 5 provides that sort of comparison. There the states are divided into free-banking and chartered banking categories. One can see that the disparity between the average reserve ratios remains about the same: 14.9 percent for free banking and 17.2 percent for chartered banking. The variations for the free-banking states are slightly greater than those for the chartered banking states.
Table 5
American Bank Reserve Ratios as of January 1, 1860 (by state)
| Free Banking | Chartered Banking | ||
| State | Reserve Ratio | State | Reserve Ratio |
| Louisiana | 38.6% | Missouri | 37.0% |
| Georgia | 23.7 | Kentucky | 23.4 |
| Tennessee | 23.0 | North Carolina | 22.1 |
| Indiana | 22.3 | Maryland | 21.4 |
| Alabama | 22.2 | Pennsylvania* | 21.3 |
| Iowa | 20.0 | Virginia | 16.7 |
| New York | 15.6 | South Carolina | 15.5 |
| Ohio | 15.2 | Maine | 10.0 |
| Massachusetts | 15.1 | Delaware | 9.9 |
| Florida | 10.5 | Rhode Island | 6.3 |
| New Jersey | 8.9 | New Hampshire | 5.7 |
| Connecticut | 7.5 | ||
| Wisconsin | 5.5 | ||
| Vermont | 4.2 | ||
| Michigan | 4.0 | ||
| Illinois | 2.3 | ||
| mean = 14.9% | mean = 17.2% | ||
Source: Amasa Walker, The Science of Wealth: A Manual of Political Economy, 1969, 163.
Notes: standard deviation (free banking) = 9.45
standard deviation (chartered banking) = 8.78
coefficient of variation (free banking) = 0.634
coefficient of variation (chartered banking) = 0.510
*Pennsylvania did not adopt free banking until after this date.
Clearly, chartered banks held slightly greater primary reserves. Whether or not the difference implies that chartered banks were significantly safer may be impossible to resolve. The common belief about free banking has been that the absence of government supervision led to fraud, inflation, and chaos, but if that were so, surely one would find that free banks often held very small specie reserves. Otherwise, how could free banking be (allegedly) so attractive to dishonest operators? Also, how could free banks manage to issue such excessive quantities of notes that serious inflation resulted? Such allegations should be questioned. It would appear plausible that holding a free bank’s notes in antebellum America was no riskier than holding a chartered bank’s notes.
At present, a common measure of a financial intermediary’s prudence is its capital/asset ratio. Table 6 provides information on such a ratio for the period 1834–1860.18 Chartered banks show an average capital ratio of 68.6 percent. Free banking exhibits an average of 57.6 percent. Again, this difference may not be large enough to suggest any significant differences in bank management policies. Furthermore, the ratios under free banking fluctuate noticeably less than they do under chartered banking. The coefficients of variation are 0.081 and 0.144, respectively.
Table 6
American Bank Capital Ratios, 1834–1860 (ratio of capital stock to sum of loans and discounts)
| Year | Capital Ratio |
| 1834 | 1.7% |
| 1835 | 63.3 |
| 1836 | 55.1 |
| 1837 | 55.4 |
| 1838 | 65.4 |
| 1839 | 66.5 |
| 1840 | 77.4 |
| 1841 | 81.1 |
| 1842 | 80.3 |
| 1843 | 89.9 |
| 1844 | 79.6 |
| 1845 | 71.4 |
| 1846 | 63.1 |
| 1847 | 65.4 |
| 1848 | 59.5% |
| 1849 | 62.4 |
| 1850 | 59.7 |
| 1851 | 55.1 |
| 1852 | . . . * |
| 1853 | 50.8 |
| 1854 | 54.1 |
| 1855 | 57.7 |
| 1856 | 54.2 |
| 1857 | 54.2 |
| 1858 | 67.7 |
| 1859 | 61.2 |
| 1860 | 60.9 |
Source: Charles A. Conant, A History of Modern Banks of Issue (New York: G.P. Putnam’s Sons, 1927) 391.
Notes: mean (1834–1849) = 68.6%
mean (1850–1860) = 57.6%
standard deviation (1834–1849) = 9.88
standard deviation (1850–1860) = 4.67
coefficient of variation (1834–1849) = 0.144
coefficient of variation (1850–1860) = 0.081
*No data are available.
The data on reserve ratios and capital ratios may not be conclusive, but they are suggestive, particularly when considered in conjunction with the conclusions of those researchers whose work was reviewed above. That is, free banking does not seem to have been the disaster it is usually thought to have been, despite the larger number of banks and the greater variety of banknotes that accompanied it.
SOME STYLIZED MACROECONOMIC FACTS
Recent research has argued that the traditional image of American free banking is seriously flawed. Both the prevalence of counterfeiting19 and fraud, and the magnitude of noteholder losses seem to have been badly overstated. Furthermore, the deficiencies of free banks are now thought to have stemmed largely from the legal requirement that such banks hold suboptimal asset portfolios of state bonds as security for their note issues and the restrictions on branching (White 1986).20
Such research is wonderfully illuminating, but it is more or less microeconomic in nature. It is neither inferior nor defective; it is simply incomplete.21 It does not address the fundamental macroeconomic question: How stable was the economy as a whole? The traditional story about American free banking suggests a period of rapid and erratic money growth, that is, a monetary series that exhibited both severe fluctuations and a strongly positive secular trend. From that premise, one would be likely to infer that wholesale and consumer prices, interest rates, and stock prices exhibited a high degree of instability as well, and that little real economic growth occurred.22
In contrast to the above, one finds the following provocative assertion by Richard Timberlake:
The banking and monetary system on the eve of the Civil War had enjoyed fifteen years of relative stability . . . records for the period 1845–1860 imply an economic tranquility. . . . Growth in bank credit and the stock of money was as orderly as it had been or would be in any other period of United States financial history. . . . Only the panic of 1857 rippled the surface, and even this event was short-lived and relatively harmless. . . . The Civil War put an end to these idyllic conditions. (1978, 84–85)
Which of these views is more nearly correct? The purpose of this section is to offer a tentative answer to that question.
The Data
Ideally, one would like to test the key implications of the model in Chapters 2 and 3. Those were that (1) nominal national income is maintained at all times, (2) the price level is stable except in the face of per-capita productivity changes, and (3) the market rate of interest remains equal to the natural rate. Sadly, there are insufficient data to test those specific propositions. One must be content with a “second-best” approach that is more general. Namely, the limited data compel one to investigate merely the question of general stability.
In order to examine the issue of relative macroeconomic stability during the free-banking period, data are offered on the monetary base per capita, the money supply per capita, wholesale prices, consumer prices, the commercial paper rate of interest, common stock prices, real income per capita, and industrial output per capita. Data on unemployment rates are not included, because there seem to be no such series that reliably extend back beyond about 1890.23 The stability of each of the first six series is calculated (using three different measures). Average annual rates of growth in income and industrial output are provided. Six different regimes in U.S. banking history are then compared in terms of the stability of monetary growth, commodity prices, interest rates, and stock prices, and the rate of real economic growth. The six regimes are the time of state chartering of banks (1835–1849),24 the heyday of free banking (1850–1860), the Greenback era (1866–1878), the classical gold standard (1879–1914), the early days of the Federal Reserve, that is, prior to the domestic abandonment of the gold standard (1919–1932), and the modern Fed system (1959–1989). The war years 1861–1865 and 1915–191825 are excluded from consideration.
Several empirical issues must be addressed before proceeding to the results. First, some explanation of the three stability measures is in order. The measures used are the coefficient of variation (CV), the mean absolute percentage change (MAPC), and the mean net percentage change (MNPC). The standard deviation (or, alternatively, the variance) of a distribution is perhaps the most commonly used measure of stability. However, the standard deviation is not reliable when one is comparing the relative variation of distributions that have very different mean values. In such a case, one needs the coefficient of variation, which equals the standard deviation divided by the mean.
There is still a potential problem with the CV. It will not always distinguish between a series that has sharp period-to-period fluctuations but little secular trend, and a series with small period-to-period fluctuations but a strong (positive or negative) secular trend. Therefore, to make that distinction, the MAPC and MNPC are used.26 The MAPC sums the absolute values of the percentage changes and divides by the number of periods. This gives a measure of short-run (period-to-period) stability. The MNPC is the absolute value of the sum of the percentage changes (positive changes net of negative changes) divided by the number of periods. It provides a measure of long-run stability in the sense that a mean-reverting series possesses long-run stability. In short, MAPC indicates the magnitude of the fluctuations, whereas MNPC indicates the magnitude of the secular trend.
Second, one must reiterate that the free-banking period is here taken to be 1850–1860 instead of 1837–1863. The justification is twofold: (1) As was mentioned in an earlier section, only four states—Michigan (1837), New York (1838), Georgia (1838), and Alabama (1849)—adopted free-banking statutes prior to 1850, and (2) the years of the Civil War are being excluded for the usual reason that wars grossly distort fundamental economic relationships.27 Since a number of states never adopted free banking, however, on what grounds can one propose that free banking dominated the 1850–1860 period?
In 1860, the United States was composed of thirty-three states. Of those, eighteen had some variant of free banking (Rockoff 1975, 3). Indeed, there were fourteen free-banking states by 1853 (out of thirty-one states altogether). Moreover, several of the most populous states were among those that embraced free banking: New York, Massachusetts, Pennsylvania, Ohio, Illinois, and Indiana, for example (Rockoff 1975, 4).28 To be precise, out of a total resident population of 31,513,000 in 1860, 21,218,000 (67.3 percent) lived in free-banking states.29 Finally, one should not overlook the fact that the prime financial center of the United States—New York30—was in a state that had an excellent free-banking system that had been in operation since 1838. It would seem plausible that free banking did come to dominate the American monetary system between 1850 and 1860.
One will notice that in the accompanying tables, both the money supply and the monetary base are measured in per-capita terms. Considering the span of time covered—150 years—during which there were enormous population changes, a per-capita measure seems preferable to measuring the total money supply.31 Furthermore, these per-capita figures are broken down into both nominal and real balances. In addition, three different estimates for the period from 1835 to 1860 are cited. The desire is to gain a clear and comprehensive view of monetary movements, particularly for the free-banking era.
The commercial paper rate of interest is taken to represent “interest rates,” because it is one of the few interest rate series for which one can find data as far back in time as the 1830s32 and that is still of economic significance today.
Regarding stock prices, the reader may wonder why two different indexes are used. For the periods 1835–1860 and 1866–1878, the index is constructed from railroad common stock prices, whereas for all later periods, the well-known Standard and Poor index is used. The reliance upon railroad stock prices in the early periods is due to the fact that, although a crude index of industrial (mostly mining) stocks could have been constructed as far back as 1857, railroad stocks were of greater economic importance (Macaulay 1938, 139fn). Indeed, “until about 1909 the railroad stock market was the American security market par excellence” (Macaulay 1938, 139).
Data on both consumer prices and wholesale prices are presented. Greater credence is given to the wholesale price figures, however. It has been widely observed that, with regard to the nineteenth century, the extant information on wholesale prices is both more extensive and more reliable than that on consumer prices. Thus, in the tables to follow, all real magnitudes are derived from nominal magnitudes by using the wholesale price index.
Finally, per-capita figures for real national income and industrial production are included in order to compare rates of growth under the various monetary regimes.
The Results
Tables 7–10 provide figures for the money supply per capita (both nominal and real). The nominal money balances are on page 114.
Table 7
Money Supply per Capita, 1835–1860 (currency plus deposits)
| Year | Nominal | Real |
| 1835 | $14.52 | $13.50 |
| 1836 | 14.55* | 11.86 |
| 1837 | 13.93* | 11.26 |
| 1838 | 14.81* | 12.52 |
| 1839 | 12.32* | 10.23 |
| 1840 | 10.65* | 10.42 |
| 1841 | 9.89* | 10 |
| 1842 | 8.12* | 9.21 |
| 1843 | 9.58* | 11.89 |
| 1844 | 10.36* | 12.51 |
| 1845 | 11.24* | 12.6 |
| 1846 | 12.14 | 13.61 |
| 1847 | 12.49 | 12.9 |
| 1848 | $14.55* | $16.49 |
| 1849 | 17.04* | 19.32 |
| 1850 | 15.49 | 17.16 |
| 1851 | 16.96 | 19.01 |
| 1952 | 18.12 | 19.15 |
| 1853 | 19.62 | 18.81 |
| 1854 | 19.15 | 16.5 |
| 1855 | 19.52 | 16.5 |
| 1856 | 20.36 | 18.04 |
| 1857 | 16.43 | 13.76 |
| 1858 | 23.13* | 23.13 |
| 1859 | 21.23* | 20.78 |
| 1860 | 17.61** | 17.61 |
Sources: Milton Friedman and Anna J. Schwartz, Monetary Statistics of the United States (New York: National Bureau of Economic Research, 1970) 220–24. U.S. Bureau of the Census, Historical Statistics of the United States. Colonial Times to 1970 (Washington, D.C.: Government Printing Office, 1975) Series A 6–8.
Notes: Estimates are as of the end of the year unless otherwise indicated.
Currency is taken to be the sum of banknotes and specie, as held by the public.
The nominal data were adjusted using the price indexes from Table 17.
* Represents an average of upper and lower estimates.
** Incorporates estimate of specie as of June 1860.
Nominal Money Balances: |
Real Money Balances: |
| CV (1835–1849) = 0.1913 | CV (1835–1849) = 0.1990 |
| CV (1850–1860) = 0.1132 | CV (1850–1860) = 0.1280 |
| MAPC (1835–1849) = 10.378 | MAPC (1835–1849) = 10.981 |
| MAPC (1850–1860) = 11.607 | MAPC (1850–1860) = 14.847 |
| MNPC (1835–1849) = 1.855 | MNPC (1835–1849) = 3.460 |
| MNPC (1850–1860) = 1.415 | MNPC (1850–1860) = 1.325 |
Table 8
Money Supply per Capita, 1867–1914 (currency plus commercial bank deposits)
| Year | Nominal | Real |
| 1867 | $35.16 | $20.18 |
| 1868 | 32.55 | 19.16 |
| 1869 | 32.37 | 19.93 |
| 1870 | 32.3 | 22.25 |
| 1871 | 33.95 | 24.28 |
| 1872 | 37.72 | 25.8 |
| 1873 | 37.72 | 26.38 |
| 1874 | 36.15 | 26.68 |
| 1875 | 37.61 | 29.64 |
| 1876 | 36.13 | 30.54 |
| 1877 | 34.97 | 30.68 |
| 1878 | 32.53 | 33.26 |
| 1879 | 33.71 | 34.82 |
| 1880 | 39.4 | 36.65 |
| 1881 | 46.16 | 41.66 |
| 1882 | 49.11 | 42.3 |
| 1883 | 51.96 | 47.85 |
| 1884 | 49.91 | 49.91 |
| 1885 | 50.07 | 54.78 |
| 1886 | 53.21 | 60.33 |
| 1887 | 55.37 | 60.58 |
| 1888 | 55.57 | 60.08 |
| 1889 | 57.35 | 65.84 |
| 1890 | 61.98 | 70.27 |
| 1891 | $62.44 | $71.28 |
| 1892 | 68.07 | 83.11 |
| 1893 | 62.54 | 74.54 |
| 1894 | 61.87 | 82.16 |
| 1895 | 64.34 | 83.89 |
| 1896 | 60.18 | 82.44 |
| 1897 | 62.09 | 84.82 |
| 1898 | 69. 90 | 91.85 |
| 1899 | 80.67 | 98.50 |
| 1900 | 83.82 | 95.14 |
| 1901 | 94.56 | 108.81 |
| 1902 | 101.17 | 109.37 |
| 1903 | 105.69 | 112.8 |
| 1904 | 109.34 | 116.57 |
| 1905 | 120.16 | 127.29 |
| 1906 | 127.14 | 130.94 |
| 1907 | 131.13 | 128.06 |
| 1908 | 127.69 | 129.24 |
| 1909 | 138.88 | 130.9 |
| 1910 | 142.85 | 129.28 |
| 1911 | 149.09 | 146.31 |
| 1912 | 157.21 | 144.89 |
| 1913 | 159.9 | 145.76 |
| 1914 | 163.24 | 152.56 |
Sources: Milton Friedman and Anna J. Schwartz, A Monetary History of the United States, 1867–1960 (Princeton: Princeton University Press, 1963)704–8. Historical Statistics of the United States, 1975, Series A 6–8.
Notes: The nominal data were adjusted using the price indexes from Table 17.
Nominal Money Balances: |
Real Money Balances: |
| CV (1867–1878) = 0.0594 | CV (1867–1878) = 0.1748 |
| CV (1879–1914) = 0.4500 | CV (1879–1914) = 0.3832 |
| MAPC (1867–1878) = 4.248 | MAPC (1867–1878) = 5.680 |
| MAPC (1879–1914) = 5.972 | MAPC (1879–1914) = 5.631 |
| MNPC (1867–1878) = 0.566 | MNPC (1867–1878) = 4.761 |
| MNPC (1879–1914) = 4.737 | MNPC (1879–1914) = 4.482 |
Table 9
Money Supply per Capita, 1919–1932 (currency plus commercial bank deposits)
| Year | Nominal | Real |
| 1919 | $207.41 | $ 95.27 |
| 1920 | 220.71 | 91.01 |
| 1921 | 195.27 | 127.38 |
| 1922 | 195.79 | 128.89 |
| 1923 | 202.92 | 128.43 |
| 1924 | 205.91 | 133.62 |
| 1925 | 220.13 | 135.38 |
| 1926 | 220.81 | 140.55 |
| 1927 | 217.30 | 144.96 |
| 1928 | 217.25 | 143.02 |
| 1929 | 216.96 | 144.93 |
| 1930 | 206.47 | 152.15 |
| 1931 | 190.54 | 166.27 |
| 1932 | 165.20 | 162.28 |
Sources: Milton Friedman and Anna J. Schwartz, A Monetary History of the United States. 1867–1960, 1963, 709–13. Historical Statistics of the United States, 1975, Series A 6–8.
Notes: The nominal data were adjusted using the price indexes from Table 17.
Nominal Money Balances:
CV = 0.0728
MAPC = 4.471
MNPC = 1.547
Real Money Balances:
CV = 0.1525
MAPC = 5.971
MNPC = 4.653
Using CV |
Using MAPC |
Using MNPC |
|||
| (1867–1878) | 0.0594 | (1867–1878) | 4.248 | (1867–1878) | 0.566 |
| (1919–1932) | 0.0728 | (1919–1932) | 4.471 | (1850–1860) | 1.415 |
| (1850–1860) | 0.1132 | (1959–1989) | 4.838 | (1919–1932) | 1.547 |
| (1835–1849) | 0.1913 | (1879–1914) | 5.972 | (1835–1849) | 1.855 |
| (1879–1914) | 0.4500 | (1835–1849) | 10.378 | (1959–1989) | 4.731 |
| (1959–1989) | 0.4898 | (1850–1860) | 11.607 | (1879–1914) | 4.737 |
Table 10
Money Supply (M1) per Capita, 1959–1989 (an average of monthly observations)
| Year | Nominal | Real |
| 1959 | $797.93 | $836.40 |
| 1960 | 785.65 | 822.67 |
| 1961 | 788.56 | 829.19 |
| 1962 | 795.78 | 839.43 |
| 1963 | 808.56 | 855.62 |
| 1964 | 828.45 | 874.82 |
| 1965 | 853.20 | 883.23 |
| 1966 | 882.86 | 884.63 |
| 1967 | 909.14 | 909.14 |
| 1968 | 963.52 | 940.02 |
| 1969 | 1010.38 | 948.71 |
| 1970 | 1035.13 | 937.62 |
| 1971 | 1090.32 | 957.26 |
| 1972 | 1154.77 | 969.58 |
| 1973 | 1226.91 | 907.48 |
| 1974 | 1276.39 | 797.25 |
| 1975 | $1322.72 | $756.27 |
| 1976 | 1383.77 | 756.57 |
| 1977 | 1474.64 | 759.34 |
| 1978 | 1578.53 | 754.19 |
| 1979 | 1681.65 | 714.08 |
| 1980 | 1764.94 | 656.6 |
| 1981 | 1869.8 | 637.29 |
| 1982 | 1972.63 | 659.08 |
| 1983 | 2170.2 | 716 |
| 1984 | 2300.82 | 741.48 |
| 1985 | 2487.86 | 805.65 |
| 1986 | 2792.57 | 931.48 |
| 1987 | 3056.29 | 993 .27 |
| 1988 | 3156.88 | 986.22 |
| 1989 | 3154.89 | 939.23 |
Sources: Board of Governors of the Federal Reserve System, Division of Research and Statistics. U.S. Bureau of the Census, Statistical Abstract of the United States (Washington, D.C.: Government Printing Office, 1990) 7.
Notes: The nominal data were adjusted using the price indexes from Table 18.
The current definition of M1 was used throughout the data set.
Nominal Money Balances:
CV = 0.4898
MAPC = 4.838
MNPC = 4.731
Real Money Balances:
CV = 0.1210
MAPC = 3.790
MNPC = 0.525
Free banking performs rather well when stability is measured by either the CV or the MNPC, but very poorly per the MAPC. Actually, the model would lead one to expect something like this. That is, significant period-to-period changes in the money supply were often the means by which nominal national income was kept constant in that model. There should be no strong secular trend, however, and there is not (see the MNPC). The modern Fed does rather poorly, ranking sixth, third, and fifth in terms of the CV, the MAPC, and the MNPC, respectively. A surprise may be the classical gold standard. The traditional argument has been that the growth in money under such a structure must be quite stable because of the inelastic nature of the supply of gold. Yet these figures reveal a definite positive trend as well as some considerable period-to-period variation.
To see how robust these results are for free banking, two other series are offered, one from A. Barton Hepburn and one from Peter Temin. These are Tables 11 and 12, respectively. Once again, the tables show both nominal and real balances, but the summaries which follow are for nominal balances. Per Hepburn:
Using CV |
Using MAPC |
Using MNPC | |||
| (1835–1849) | 0.2253 | (1835–1849) | 11.762 | (1835–1849) | 0.265 |
| (1850–1860) | 0.1069 | (1850–1860) | 9.930 | (1850–1860) | 5.401 |
Per Temin:
Using CV |
Using MAPC |
Using MNPC | |||
| (1835–1849) | 0.1926 | (1835–1849) | 10.762 | (1835–1849) | 0.116 |
| (1850–1858) | 0.0867 | (1850–1858) | 10.460 | (1850–1858) | 4.159 |
With both the alternative series, free banking exhibits greater stability per the CV and the MAPC, and less stability per the MNPC, than with the Friedman and Schwartz data used originally. Overall, one might say that according to the Friedman and Schwartz series, free banking exhibited relatively less short-run stability and relatively more long-run stability than with the Hepburn and Temin series. However, the results remain broadly similar in all three cases.
Tables 13–16 contain figures for the monetary base per capita. As with the money supply data, the following figures are for nominal balances.
Using CV |
Using MAPC |
Using MNPC | |||
| (1919–1932) | 0.0439 | (1867–1878) | 3.187 | (1919–1932) | 0.460 |
| (1850–1860) | 0.0543 | (1919–1932) | 3.572 | (1850–1860) | 1.446 |
| (1867–1878) | 0.1018 | (1879–1914) | 3.900 | (1879–1914) | 2.376 |
| (1879–1914) | 0.2218 | (1850–1860) | 4.625 | (1867–1878) | 3.187 |
| (1835–1849) | 0.2533 | (1959–1989) | 5.420 | (1835–1849) | 3.549 |
| (1959–1989) | 0.5279 | (1835–1849) | 9.453 | (1959–1989) | 5.259 |
Table 11
Money Supply per Capita (Hepburn Estimate), 1835–1860 (currency plus deposits)
| Year | Nominal | Real |
| 1835 | $14.82 | $13.78 |
| 1836 | 20.2 | 16.48 |
| 1837 | 19.31 | 15.62 |
| 1838 | 15.31 | 12.94 |
| 1839 | 15.88 | 13.19 |
| 1840 | 13.38 | 13.09 |
| 1841 | 12.18 | 12.32 |
| 1842 | 10.79 | 12.24 |
| 1843 | 8.97 | 11.13 |
| 1844 | 10.32 | 12.47 |
| 1845 | 11 | 12.33 |
| 1846 | 11.97 | 13.42 |
| 1847 | 13.13 | 13.56 |
| 1848 | $13.17 | $14.93 |
| 1849 | 12.37 | 14.03 |
| 1850 | 14.75 | 16.33 |
| 1851 | 17.02 | 19.08 |
| 1852 | . . . . * | . . . . * |
| 1853 | 17.83 | 17.1 |
| 1854 | 20.9 | 18 |
| 1855 | 20.23 | 17.1 |
| 1856 | 20.56 | 18.21 |
| 1857 | 21.66 | 18.14 |
| 1858 | 17.45 | 17.45 |
| 1859 | 19.36 | 18.94 |
| 1860 | 19.77 | 19.77 |
Sources: A. Barton Hepburn, History of Coinage and Currency in the united States, 1903, 118–20, 153–54, 174. Historical Statistics of the United States. 1975, Series A 6–8.
Notes: Currency is taken to be the sum of banknotes and specie, as held by the public.
The nominal data were adjusted using the price indexes from Table 17.
*Incomplete data.
Nominal Money Balances:
CV (1835–1849) = 0.2253
CV (1850–1860) = 0.1069
MAPC (1835–1849) = 11.762
MAPC (1850–1860) = 9.930
MNPC (1835–1849) = 0.265
MNPC (1850–1860) = 5.401
Real Money Balances:
CV (1835–1849) = 0.1005
CV (1850–1860) = 0.0554
MAPC (1835–1849) = 7.103
MAPC (1850–1860) = 7.747
MNPC (1835–1849) = 0.548
MNPC (1850–1860) = 3.834
Table 12
Money Supply per Capita (Temin Estimate), 1835–1858 (currency plus deposits)
| Year | Nominal | Real |
| 1835 | $16.40 | $15.26 |
| 1836 | 17.9 | 14.6 |
| 1837 | 14.64 | 11.84 |
| 1838 | 14.76 | 12.48 |
| 1839 | 12.89 | 10.71 |
| 1840 | 10.86 | 10.63 |
| 1841 | 9.81 | 9.92 |
| 1842 | 8.61 | 9.76 |
| 1843 | 10.23 | 12.69 |
| 1844 | 10.94 | 13.21 |
| 1845 | 11.94 | 13.39 |
| 1846 | 12.84 | 14.39 |
| 1847 | $13.13 | $13.56 |
| 1848 | 12.13 | 13.75 |
| 1849 | 14.54 | 16.49 |
| 1850 | 17.15 | 18.99 |
| 1851 | . . . . * | . . . . * |
| 1852 | 18.1 | 19.13 |
| 1853 | 21.22 | 20.35 |
| 1854 | 20.29 | 17.48 |
| 1855 | 20.63 | 17.44 |
| 1856 | 21.66 | 19.19 |
| 1857 | 17.15 | 14.36 |
| 1858 | 19.05 | 19.05 |
Sources: Peter Temin, The Jacksonian Economy (New York: W. W. Norton, 1969) 71, 159. Historical Statistics of the United States, 1975, Series A 6–8.
Notes: Currency is taken to be the sum of banknotes and specie, as held by the public.
The nominal data were adjusted using the price indexes from Table 17.
*No data provided.
Nominal Money Balances:
CV (1835–1849) = 0.1926
CV (1850–1858) = 0.0867
MAPC (1835–1849) = 10.762
MAPC (1850–1858) = 10.460
MNPC (1835–1849) = 0.116
MNPC (1850–1858) = 4.159
Real Money Balances:
CV (1835–1849) = 0.1492
CV (1850–1858) = 0.0942
MAPC (1835–1849) = 8.707
MAPC (1850–1858) = 13.059
MNPC (1835–1849) = 1.247
MNPC (1850–1858) = 3.184
Table 13
Monetary Base per Capita, 1835–1860
| Year | Nominal | Real |
| 1835 | $4.33 | $4.03 |
| 1836 | 4.73 | 3.86 |
| 1837 | 5.5 | 4.45 |
| 1838 | 5.36 | 4.53 |
| 1839 | 4.98 | 4.14 |
| 1840 | 4.37 | 4.28 |
| 1841 | 4.53 | 4.58 |
| 1842 | 4.88 | 5.53 |
| 1843 | 5.27 | 6.54 |
| 1844 | 4.92 | 5.94 |
| 1845 | 4.77 | 5.35 |
| 1846 | 5.64 | 6.32 |
| 1847 | 5.09 | 5.26 |
| 1848 | $5.36 | $6.08 |
| 1849 | 6.57 | 7.45 |
| 1850 | 7.58 | 8.39 |
| 1851 | 7.91 | 8.87 |
| 1852 | 8.72 | 9.22 |
| 1853 | 8.54 | 8.19 |
| 1854 | 8.68 | 7.48 |
| 1855 | 8.4 | 7.1 |
| 1856 | 8.54 | 7.56 |
| 1857 | 8.55 | 7.16 |
| 1858 | 8.17 | 8.17 |
| 1859 | 7.58 | 7.42 |
| 1860 | 7.54 | 7.54 |
Sources: Milton Friedman and Anna J. Schwartz, Monetary Statistics of the United States. 1970, 220 24. Historical Statistics of the United States, 1975, Series A 6–8.
Notes: The nominal data were adjusted using the price indexes from Table 17.
Nominal:
CV (1835–1849) = 0.2533
CV (1850–1860) = 0.0543
MAPC (1835–1849) = 9.453
MAPC (1850–1860) = 4.625
MNPC (1835–1849) = 3.549
MNPC (1850–1860) = 1.446
Real:
CV (1835–1849) = 0.1984
CV (1850–1860) = 0.0845
MAPC (1835–1849) = 12.246
MAPC (1850–1860) = 7.625
MNPC (1835–1849) = 5.288
MNPC (1850–1860) = 0.463
Table 14
Monetary Base per Capita, 1867–1914
| Year | Nominal | Real |
| 1867 | $22.80 | $13.09 |
| 1868 | 20.28 | 11.94 |
| 1869 | 19.49 | 12 |
| 1870 | 19 | 13.09 |
| 1871 | 18.96 | 13.56 |
| 1872 | 18.63 | 12.74 |
| 1873 | 18.21 | 12.73 |
| 1874 | 18.19 | 13.42 |
| 1875 | 17.19 | 13.55 |
| 1876 | 16.37 | 13.84 |
| 1877 | 16.08 | 14.11 |
| 1878 | 15.88 | 16.24 |
| 1879 | 15.93 | 16.46 |
| 1880 | 18.6 | 17.3 |
| 1881 | 20.55 | 18.55 |
| 1882 | 21.28 | 18.33 |
| 1883 | 21.89 | 20.16 |
| 1884 | 21.47 | 21.47 |
| 1885 | 21.76 | 23.81 |
| 1886 | 20.63 | 23 . 39 |
| 1887 | 21.23 | 23.23 |
| 1888 | 21.64 | 23.39 |
| 1889 | 21.34 | 24.5 |
| 1890 | 21.66 | 24.56 |
| 1891 | $22.27 | $25.42 |
| 1892 | 23.28 | 28.42 |
| 1893 | 22.61 | 26.95 |
| 1894 | 22.98 | 30.52 |
| 1895 | 21.59 | 28.15 |
| 1896 | 19.74 | 27.04 |
| 1897 | 21.11 | 28.84 |
| 1898 | 23.42 | 30.78 |
| 1899 | 24.14 | 29.47 |
| 1900 | 25.68 | 29.15 |
| 1901 | 26.68 | 30.7 |
| 1902 | 27.07 | 29.26 |
| 1903 | 27.99 | 29.87 |
| 1904 | 29.28 | 31.22 |
| 1905 | 29.47 | 31.22 |
| 1906 | 30.75 | 31.67 |
| 1907 | 32.97 | 32.2 |
| 1908 | 34.88 | 35.3 |
| 1909 | 34.51 | 32.53 |
| 1910 | 34.38 | 31.11 |
| 1911 | 34.93 | 34.28 |
| 1912 | 35.1 | 32.35 |
| 1913 | 35.19 | 32.08 |
| 1914 | 35.64 | 33.31 |
Sources: Milton Friedman and Anna J. Schwartz, A Monetary History of the United States. 1867–1960. 1963, 799–801. Historical Statistics of the United States, 1975, Series A 6–8.
Notes: The nominal data were adjusted using the price indexes from Table 17.
Nominal: |
Real: |
| CV (1867–1878) = 0.1018 | CV (1867–1878) = 0.0808 |
| CV (1879–1914) = 0.2218 | CV (1879–1914) = 0.1831 |
| MAPC (1867–1878) = 3.187 | MAPC (1867–1878) = 4.878 |
| MAPC (1879–1914) = 3.900 | MAPC (1879–1914) = 4.905 |
| MNPC (1867–1878) = 3.187 | MNPC (1867–1878) = 2.168 |
| MNPC (1879–1914) = 2.376 | MNPC (1879–1914) = 2.171 |
Table 15
Monetary Base per Capita, 1919–1932
| Year | Nominal | Real |
| 1919 | $63.34 | $29.10 |
| 1920 | 67.51 | 27.84 |
| 1921 | 60.25 | 39.30 |
| 1922 | 57.09 | 37.58 |
| 1923 | 59.64 | 37.75 |
| 1924 | 60.24 | 39.09 |
| 1925 | 60.66 | 37.31 |
| 1926 | 60.67 | 38. 62 |
| 1927 | 60.25 | 40.19 |
| 1928 | 59.11 | 38.91 |
| 1929 | 58.47 | 39.06 |
| 1930 | 56.36 | 41.53 |
| 1931 | 59.01 | 51.49 |
| 1932 | 62.37 | 61.27 |
Sources: Milton Friedman and Anna J. Schwartz, A Monetary History of the United States, 1867–1960. 1963, 801–4. Historical Statistics of The United States, 1975, Series A 6–8.
Notes: The nominal data were adjusted using the price indexes from Table 17.
Nominal: |
Real: |
| CV = 0.0439 | CV = 0.1982 |
| MAPC = 3.572 | MAPC = 9.144 |
| MNPC = 0.460 | MNPC = 6.614 |
Here free banking does very well in the long run, but only moderately well in the short run. The early Fed exhibits consistent stability (ranking first or second in each case), whereas the modern Fed ranks either fifth or sixth by all three measures.
Some might argue at this point that the long-run stability of the money supply under free banking seems to be attributable to the stability of the commodity base—gold—and thus is not indicative of the monetary stability of free banking per se. Throughout this book, however, the model of free banking that has been defended is that based on a specie standard. It has furthermore been suggested that the two may be inseparable (White 1985, 124–26). The most likely form of free banking in the future—and the only form that has existed in the past33—is that which issues inside money redeemable in some commodity, usually gold. In such a context, the stability of the commodity base is an attribute of free banking itself.
Table 16
Monetary Base per Capita, 1959–1989
| Year | Nominal | Real |
| 1959 | $244.62 | $256.42 |
| 1960 | 240.75 | 252.09 |
| 1961 | 238.75 | 251.05 |
| 1962 | 242.29 | 255.58 |
| 1963 | 247.93 | 262.36 |
| 1964 | 256.77 | 271.14 |
| 1965 | 266.01 | 275.37 |
| 1966 | 276.93 | 277.48 |
| 1967 | 287.45 | 287.45 |
| 1968 | 303.16 | 295.77 |
| 1969 | 317.6 | 298.22 |
| 1970 | 331.01 | 299.83 |
| 1971 | 350.24 | 307.5 |
| 1972 | 370.84 | 311.37 |
| 1973 | 399.28 | 295.33 |
| 1974 | 429.5 | 268.27 |
| 1975 | $453.35 | $259.21 |
| 1976 | 481.42 | 263.21 |
| 1977 | 514.33 | 264.85 |
| 1978 | 555.66 | 265.48 |
| 1979 | 593.72 | 252.11 |
| 1980 | 638.05 | 237.37 |
| 1981 | 671.71 | 228.94 |
| 1982 | 707.25 | 236.3 |
| 1983 | 763.39 | 251.86 |
| 1984 | 819.28 | 264.03 |
| 1985 | 873.73 | 282.94 |
| 1986 | 941.70 | 314.11 |
| 1987 | 1,025.02 | 333.12 |
| 1988 | 1,093.50 | 341.61 |
| 1989 | 1,128.76 | 336.04 |
Sources: Board of Governors of the Federal Reserve System, Division of Research and Statistics. U.S. Bureau of the Census, Statistical Abstract of the United States, 1990, 7.
Notes: The nominal data were adjusted using the price indexes from Table 18.
Nominal: |
Real: |
| CV = 0.5279 | CV = 0.1051 |
| MAPC = 5.420 | MAPC = 3.383 |
| MNPC = 5.259 | MNPC = 1.208 |
Tables 17 and 18 present data on the wholesale price index:
Using CV |
Using MAPC |
Using MNPC | |||
| (1850–1860) | 0.1000 | (1879–1914) | 4.316 | (1879–1914) | 0.387 |
| (1879–1914) | 0.1245 | (1959–1989) | 4.695 | (1835–1849) | 1.095 |
| (1835–1849) | 0.1468 | (1850–1860) | 5.822 | (1850–1860) | 1.425 |
| (1866–1878) | 0.1718 | (1866–1878) | 5.922 | (1959–1989) | 4.399 |
| (1919–1932) | 0.2184 | (1835–1849) | 6.195 | (1919–1932) | 4.876 |
| (1959–1989) | 0.4847 | (1919–1932) | 7.916 | (1866–1878) | 5.159 |
Furthermore, Tables 19 and 20 deal with the consumer price index:
Using CV |
Using MAPC |
Using MNPC | |||
| (1850–1860) | 0.0436 | (1879–1914) | 1.326 | (1879–1914) | 0.128 |
| (1879–1914) | 0.0506 | (1850–1860) | 2.722 | (1850–1860) | 0.775 |
| (1919–1932) | 0.0791 | (1866–1878) | 3.375 | (1835–1849) | 1.433 |
| (1835–1849) | 0.0865 | (1835–1849) | 3.790 | (1919–1932) | 1.584 |
| (1866–1878) | 0.1156 | (1919–1932) | 4.859 | (1866–1878) | 3.375 |
| (1959–1989) | 0.5174 | (1959–1989) | 4.994 | (1959–1989) | 4.994 |
The classical gold standard and the free-banking era clearly dominate the other periods with regard to commodity prices. That such stability characterized the 1879–1914 period has, of course, long been recognized. It is not often noticed that the 1850–1860 period also exhibited a high degree of price stability. As one can see from the above figures, the modern Fed performs very poorly, whereas the early Fed was only marginally better.
Tables 21–24 reveal the quarter-by-quarter movements in the commercial paper rate of interest:
Using CV |
Using MAPC |
Using MNPC | |||
| (1879–1914) | 0.2138 | (1919–1932) | 0.405 | (1850–1860) | 0.002 |
| (1866–1878) | 0.2804 | (1879–1914) | 0.673 | (1879–1914) | 0.017 |
| (1850–1860) | 0.2945 | (1959–1989) | 0.686 | (1959–1989) | 0.019 |
| (1919–1932) | 0.3071 | (1866–1878) | 1.145 | (1836–1849) | 0.022 |
| (1959–1989) | 0.4264 | (1850–1860) | 1.666 | (1866–1878) | 0.041 |
| (1836–1849) | 0.5507 | (1836–1849) | 2.520 | (1919–1932) | 0.065 |
The only period to rank near the top of the lists both short run and long run is the classical gold standard. The time of state chartering of banks does quite badly, whereas free banking exhibits large quarterly fluctuations but by far the smallest secular trend.
Tables 25–28 review common stock prices:
Using CV |
Using MAPC |
Using MNPC | |||
| (1866–1878) | 0.1379 | (1866–1878) | 7.251 | (1866–1878) | 0.222 |
| (1835–1849) | 0.1916 | (1850–1860) | 9.591 | (1835–1849) | 0.844 |
| (1850–1860) | 0.2270 | (1879–1914) | 10.020 | (1919–1932) | 1.685 |
| (1879–1914) | 0.2769 | (1835–1849) | 11.398 | (1850–1860) | 2.233 |
| (1919–1932) | 0.4568 | (1959–1989) | 11.480 | (1879–1914) | 2.709 |
| (1959–1989) | 0.5468 | (1919–1932) | 21.167 | (1959–1989) | 6.683 |
Table 17
Wholesale Price Index, 1835–1932 (1860 = 100)
| Year | Index |
| 1835 | 107.5 |
| 1836 | 122.6 |
| 1837 | 123.7 |
| 1838 | 118.3 |
| 1839 | 120.4 |
| 1840 | 102.2 |
| 1841 | 98.9 |
| 1842 | 88.2 |
| 1843 | 80.6 |
| 1844 | 82.8 |
| 1845 | 89.2 |
| 1846 | 89.2 |
| 1847 | 96.8 |
| 1848 | 88.2 |
| 1849 | 88.2 |
| 1850 | 90.3 |
| 1851 | 89.2 |
| 1852 | 94.6 |
| 1853 | 104.3 |
| 1854 | 116.1 |
| 1855 | 118.3 |
| 1856 | 112.9 |
| 1857 | 119.4 |
| 1858 | 100 |
| 1859 | 102.2 |
| 1860 | 100 |
| 1866 | 187.1 |
| 1867 | 174.2 |
| 1868 | 169.9 |
| 1869 | 162.4 |
| 1870 | 145.2 |
| 1871 | 139.8 |
| 1872 | 146.2 |
| 1873 | 143 |
| 1874 | 135.5 |
| 1875 | 126.9 |
| 1876 | 118.3 |
| 1877 | 114 |
| 1878 | 97.8 |
| 1879 | 96.8 |
| 1880 | 107.5 |
| 1881 | 110.8 |
| 1882 | 116.1 |
| 1883 | 108.6 |
| 1884 | 100 |
| 1885 | 91.4 |
| 1886 | 88.2 |
| 1887 | 91.4 |
| 1888 | 92.5 |
| 1889 | 87.1 |
| 1890 | 88.2 |
| 1891 | 87.6 |
| 1892 | 81.9 |
| 1893 | 83.9 |
| 1894 | 75.3 |
| 1895 | 76.7 |
| 1896 | 73 |
| 1897 | 73.2 |
| 1898 | 76.1 |
| 1899 | 81.9 |
| 1900 | 88.1 |
| 1901 | 86.9 |
| 1902 | 92.5 |
| 1903 | 93.7 |
| 1904 | 93.8 |
| 1905 | 94.4 |
| 1906 | 97.1 |
| 1907 | 102.4 |
| 1908 | 98.8 |
| 1909 | 106.1 |
| 1910 | 110.5 |
| 1911 | 101.9 |
| 1912 | 108.5 |
| 1913 | 109.7 |
| 1914 | 107 |
| 1919 | 217.7 |
| 1920 | 242.5 |
| 1921 | 153.3 |
| 1922 | 151.9 |
| 1923 | 158 |
| 1924 | 154.1 |
| 1925 | 162.6 |
| 1926 | 157.1 |
| 1927 | 149.9 |
| 1928 | 151.9 |
| 1929 | 149.7 |
| 1930 | 135.7 |
| 1931 | 114.6 |
| 1932 | 101.8 |
Source: Historical Statistics of the United States, 1975, Series E 40–63.
Notes: The war years 1861–1865 and 1915–1918 are excluded.
| CV | (1835–1849) = 0.1468 | MNPC | (1835–1849) = 1.095 | |
| (1850–1860) = 0.1000 | (1850–1860) = 1.425 | |||
| (1866–1878) = 0.1718 | (1866–1878) = 5.159 | |||
| (1879–1914) = 0.1245 | (1879–1914) = 0.387 | |||
| (1919–1932) = 0.2184 | (1919–1932) = 4.876 | |||
| MAPC | (1835–1849) = 6.195 | |||
| (1850–1860) = 5.822 | ||||
| (1866–1878) = 5.922 | ||||
| (1879–1914) = 4.316 | ||||
| (1919–1932) = 7.916 |
Table 18
Wholesale Price Index, 1959–1989 (1967 = 100)
| Year | Index |
| 1959 | 95.4 |
| 1960 | 95.5 |
| 1961 | 95.1 |
| 1962 | 94.8 |
| 1963 | 94.5 |
| 1964 | 94.7 |
| 1965 | 96.6 |
| 1966 | 99.8 |
| 1967 | 100 |
| 1968 | 102.5 |
| 1969 | 106.5 |
| 1970 | 110.4 |
| 1971 | 113.9 |
| 1972 | 119.1 |
| 1973 | 135.2 |
| 1974 | 160.1 |
| 1975 | 174.9 |
| 1976 | 182.9 |
| 1977 | 194.2 |
| 1978 | 209.3 |
| 1979 | 235.5 |
| 1980 | 268.8 |
| 1981 | 293.4 |
| 1982 | 299.3 |
| 1983 | 303.1 |
| 1984 | 310.3 |
| 1985 | 308.8 |
| 1986 | 299.8 |
| 1987 | 307.7 |
| 1988 | 320.1 |
| 1989 | 335.9 |
Source: Various issues of the Survey of Current Business.
Notes: CV = 0.4847
MAPC = 4.695
MNPC = 4.399
Here the Greenback era totally dominates. Free banking does fairly well, better, for example, than the classical gold standard by all three measures. The modern Fed brings up the rear, with the early Fed doing slightly better.
In addition to the issue of relative stability, one might want to compare rates of economic growth across the various monetary regimes. Tables 29 and 30 show real national income per capita for selected years. For comparative purposes, one may consider the following figures for the average annual percentage change:
(1869–1879)= + 3.49
(1959–1989)= + 2.57
(1879–1907/1911)= + 2.10
(1839–1849)= + 1.61
(1849–1859)= + 1.37
(1919–1932)= + 0.68
Table 19
Consumer Price Index, 1835–1932 (1860 = 100)
| Year | Index |
| 1835 | 114.8 |
| 1836 | 122.2 |
| 1837 | 125.9 |
| 1838 | 118.5 |
| 1839 | 118.5 |
| 1840 | 111.1 |
| 1841 | 114.8 |
| 1842 | 107.4 |
| 1843 | 103.7 |
| 1844 | 103.7 |
| 1845 | 103.7 |
| 1846 | 100 |
| 1847 | 103.7 |
| 1848 | 96.3 |
| 1849 | 92.6 |
| 1850 | 92.6 |
| 1851 | 92.6 |
| 1852 | 92.6 |
| 1853 | 92.6 |
| 1854 | 100 |
| 1855 | 103.7 |
| 1856 | 100 |
| 1857 | 103.7 |
| 1858 | 96.3 |
| 1859 | 100 |
| 1860 | 100 |
| 1866 | 163 |
| 1867 | 155.6 |
| 1868 | 148.1 |
| 1869 | 148.1 |
| 1870 | 140.7 |
| 1871 | 133.3 |
| 1872 | 133.3 |
| 1873 | 133.3 |
| 1874 | 125.9 |
| 1875 | 122.2 |
| 1876 | 118.5 |
| 1877 | 118.5 |
| 1878 | 107.4 |
| 1879 | 103.7 |
| 1880 | 107.4 |
| 1881 | 107.4 |
| 1882 | 107.4 |
| 1883 | 103.7 |
| 1884 | 100 |
| 1885 | 100 |
| 1886 | 100 |
| 1887 | 100 |
| 1888 | 100 |
| 1889 | 100 |
| 1890 | 100 |
| 1891 | 100 |
| 1892 | 100 |
| 1893 | 100 |
| 1894 | 96.3 |
| 1895 | 92.6 |
| 1896 | 92.6 |
| 1897 | 92.6 |
| 1898 | 92.6 |
| 1899 | 92.6 |
| 1900 | 92.6 |
| 1901 | 92.6 |
| 1902 | 96.3 |
| 1903 | 100 |
| 1904 | 100 |
| 1905 | 100 |
| 1906 | 100 |
| 1907 | 103.7 |
| 1908 | 100 |
| 1909 | 100 |
| 1910 | 103.7 |
| 1911 | 103.7 |
| 1912 | 107.4 |
| 1913 | 110 |
| 1914 | 111.5 |
| 1919 | 191.9 |
| 1920 | 222.2 |
| 1921 | 198.5 |
| 1922 | 185.9 |
| 1923 | 189.3 |
| 1924 | 189.6 |
| 1925 | 194.4 |
| 1926 | 196.3 |
| 1927 | 192.6 |
| 1928 | 190 |
| 1929 | 190 |
| 1930 | 185.2 |
| 1931 | 168.9 |
| 1932 | 151.5 |
Source: Historical Statistics of the United States, 1975, Series E 135–66.
Notes: The war years 1861–1865 and 1915–1918 are excluded.
| CV | (1835–1849) = 0.0865 | MNPC | (1835–1849) = 1.433 | |
| (1850–1860) = 0.0436 | (1850–1860) = 0.775 | |||
| (1866–1878) = 0.1156 | (1866–1878) = 3.375 | |||
| (1879–1914) = 0.0506 | (1879–1914) = 0.128 | |||
| (1919–1932) = 0.0791 | (1919–1932) = 1.584 | |||
| MAPC | (1835–1849) = 3.790 | |||
| (1850–1860)= 2.722 | ||||
| (1866–1878) = 3.375 | ||||
| (1879–1914) = 1.326 | ||||
| (1919–1932) = 4.859 |
Table 20
Consumer Price Index, 1959–1989 (1967 = 100)
| Year | Index |
| 1959 | 87.3 |
| 1960 | 88.6 |
| 1961 | 89. 6 |
| 1962 | 90. 6 |
| 1963 | 91.8 |
| 1964 | 93.0 |
| 1965 | 94.5 |
| 1966 | 97.3 |
| 1967 | 100 |
| 1968 | 104.2 |
| 1969 | 109.8 |
| 1970 | 116. 3 |
| 1971 | 121. 3 |
| 1972 | 125. 3 |
| 1973 | 133.1 |
| 1974 | 147.7 |
| 1975 | 161.2 |
| 1976 | 170.5 |
| 1977 | 181.5 |
| 1978 | 195.4 |
| 1979 | 217.4 |
| 1980 | 246.7 |
| 1981 | 272.4 |
| 1982 | 289.1 |
| 1983 | 298.4 |
| 1984 | 311.1 |
| 1985 | 322.2 |
| 1986 | 328.4 |
| 1987 | 340.4 |
| 1988 | 354.2 |
| 1989 | 371.3 |
Source: Various issues of the Survey of Current Business.
Notes: CV = 0.5174
MAPC = 4.994
MNPC = 4.994
Free banking seems to do rather feebly in relative terms. However, one should note that (1) the Greenback era experienced its rapid growth at least in part because it was a period of recovery from a devastating war and (2) the massive expansion of the public sector in recent decades has surely introduced an upward bias into such measures of growth for the 1959–1989 period. A direct measure of production might be preferable. Tables 31 and 32 display indexes of industrial production per capita for selected years. The summary of the average annual percentage changes appears as follows:
(1829/1839–1839/1849)= +4.73
(1839/1849–1849/1859)= +4.31
(1879–1914)= +2.81
(1959–1989)= +2.45
(1866–1878)= +1.87
(1919–1932)= -2.24
Table 21
Commercial Paper Rate of Interest at Boston and New York, 1836–1860 (an average of observed rates, by quarter)
| Quarter | Rate |
| 1836Q1 | 10.7 |
| 1836Q2 | 14.5 |
| 1836Q3 | 20.5 |
| 1836Q4 | 28.0 |
| 1837Q1 | 18.7 |
| 1837Q2 | 22.7 |
| 1837Q3 | 7.3 |
| 1837Q4 | 8 |
| 1838Q1 | 12.7 |
| 1838Q2 | 10.4 |
| 1838Q3 | 6.3 |
| 1838Q4 | 7.1 |
| 1839Q1 | 7.5 |
| 1839Q2 | 8 |
| 1839Q3 | 14.3 |
| 1839Q4 | 23.1 |
| 1840Q1 | 10.0 |
| 1840Q2 | 7.8 |
| 1840Q3 | 6.4 |
| 1840Q4 | 6.5 |
| 1841Q1 | 6.5 |
| 1841Q2 | 6.2 |
| 1841Q3 | 6.2 |
| 1841Q4 | 8.3 |
| 1842Q1 | 10.5 |
| 1842Q2 | 8 |
| 1842Q3 | 7.4 |
| 1842Q4 | 6.7 |
| 1843Q1 | 5.7 |
| 1843Q2 | 4.7 |
| 1843Q3 | 3.8 |
| 1843Q4 | 3.8 |
| 1844Q1 | 4.2 |
| 1844Q2 | 5 |
| 1844Q3 | 5.1 |
| 1844Q4 | 5.1 |
| 1845Q1 | 5.7 |
| 1845Q2 | 5.8 |
| 1845Q3 | 6 |
| 1845Q4 | 6.8 |
| 1846Q1 | 8 |
| 1846Q2 | 10.2 |
| 1846Q3 | 5.3 |
| 1846Q4 | 6.7 |
| 1847Q1 | 9.7 |
| 1847Q2 | 7.7 |
| 1847Q3 | 7.8 |
| 1847Q4 | 13.0 |
| 1848Q1 | 15.5 |
| 1848Q2 | 15.0 |
| 1848Q3 | 14.5 |
| 1848Q4 | 15.5 |
| 1849Q1 | 12.0 |
| 1849Q2 | 10.5 |
| 1849Q3 | 8.3 |
| 1849Q4 | 9.5 |
| 1850Q1 | 8.9 |
| 1850Q2 | 8.2 |
| 1850Q3 | 7.5 |
| 1850Q4 | 7.5 |
| 1851Q1 | 7.4 |
| 1851Q2 | 7.7 |
| 1851Q3 | 11.5 |
| 1851Q4 | 12.2 |
| 1852Q1 | 7.8 |
| 1852Q2 | 5.9 |
| 1852Q3 | 6 |
| 1852Q4 | 6 |
| 1853Q1 | 9.2 |
| 1853Q2 | 9.1 |
| 1853Q3 | 9.8 |
| 1853Q4 | 12.9 |
| 1854Q1 | 8.5 |
| 1854Q2 | 10.7 |
| 1854Q3 | 10.0 |
| 1854Q4 | 12.3 |
| 1855Q1 | 10.2 |
| 1855Q2 | 7.6 |
| 1855Q3 | 7.2 |
| 1855Q4 | 10.9 |
| 1856Q1 | 10.2 |
| 1856Q2 | 7.7 |
| 1856Q3 | 7.8 |
| 1856Q4 | 9.8 |
| 1857Q1 | 9.3 |
| 1857Q2 | 7.8 |
| 1857Q3 | 14.3 |
| 1857Q4 | 13.3 |
| 1858Q1 | 6.4 |
| 1858Q2 | 4.6 |
| 1858Q3 | 4.2 |
| 1858Q4 | 4.7 |
| 1859Q1 | 5.6 |
| 1859Q2 | 6.4 |
| 1859Q3 | 7 |
| 1859Q4 | 7.4 |
| 1860Q1 | 7.1 |
| 1860Q2 | 4.8 |
| 1860Q3 | 5.8 |
| 1860Q4 | 9.6 |
Source: Walter B. Smith and Arthur H. Cole, Fluctuations in American Business, 1790–1860 (Cambridge, Mass.: Harvard University Press, 1935) 192–94.
Notes: CV (1836Q1–1849Q4) = 0.5507
CV (1850Q1–1860Q4) = 0.2945
MAPC (1836Q1–1849Q4) = 2.520
MAPC (1850Q1–1860Q4) = 1.666
MNPC (1836Q1–1849Q4) = 0.022
MNPC (1850Q1–1860Q4) = 0.002
Table 22
Commercial Paper Rate of Interest at New York City, 1866–1914 (an average of observed rates, by quarter)
| Quarter | Rate |
| 1866Q1 | 7.32 |
| 1866Q2 | 6.1 |
| 1866Q3 | 5.68 |
| 1866Q4 | 6.34 |
| 1867Q1 | 7.22 |
| 1867Q2 | 7.07 |
| 1867Q3 | 6.72 |
| 1867Q4 | 8.3 |
| 1868Q1 | 7.08 |
| 1868Q2 | 6.78 |
| 1868Q3 | 6.62 |
| 1868Q4 | 8.56 |
| 1869Q1 | 8.47 |
| 1869Q2 | 9.11 |
| 1869Q3 | 10.25 |
| 1869Q4 | 10.83 |
| 1870Q1 | 7.96 |
| 1870Q2 | 6.34 |
| 1870Q3 | 6.94 |
| 1870Q4 | 7.76 |
| 1871Q1 | 6.7 |
| 1871Q2 | 5.78 |
| 1871Q3 | 5.65 |
| 1871Q4 | 9.75 |
| 1872Q1 | 8.16 |
| 1872Q2 | 7.29 |
| 1872Q3 | 8 |
| 1872Q4 | 11.08 |
| 1873Q1 | 9.59 |
| 1873Q2 | 8.78 |
| 1873Q3 | 9.26 |
| 1873Q4 | 13.60 |
| 1874Q1 | 6.53 |
| 1874Q2 | 5.82 |
| 1874Q3 | 5.77 |
| 1874Q4 | 5.81 |
| 1875Q1 | 5.45 |
| 1875Q2 | 4.8 |
| 1875Q3 | 5.05 |
| 1875Q4 | 6.44 |
| 1876Q1 | 5.72 |
| 1876Q2 | 5.1 |
| 1876Q3 | 4.05 |
| 1876Q4 | 5.66 |
| 1877Q1 | 4.83 |
| 1877Q2 | 4.15 |
| 1877Q3 | 5.39 |
| 1877Q4 | 6.35 |
| 1878Q1 | 5.43 |
| 1878Q2 | 4.57 |
| 1878Q3 | 4.01 |
| 1878Q4 | 5.21 |
| 1879Q1 | 4.4 |
| 1879Q2 | 4.71 |
| 1879Q3 | 5.08 |
| 1879Q4 | 6.05 |
| 1880Q1 | 5.4 |
| 1880Q2 | 5.08 |
| 1880Q3 | 4.91 |
| 1880Q4 | 5.52 |
| 1881Q1 | 5.39 |
| 1881Q2 | 4.25 |
| 1881Q3 | 4.88 |
| 1881Q4 | 6.27 |
| 1882Q1 | 5.54 |
| 1882Q2 | 5.01 |
| 1882Q3 | 5.67 |
| 1882Q4 | 6.35 |
| 1883Q1 | 5.75 |
| 1883Q2 | 5.55 |
| 1883Q3 | 5.46 |
| 1883Q4 | 5.73 |
| 1884Q1 | 4.75 |
| 1884Q2 | 5.18 |
| 1884Q3 | 5.65 |
| 1884Q4 | 5.23 |
| 1885Q1 | 4.55 |
| 1885Q2 | 3.73 |
| 1885Q3 | 3.64 |
| 1885Q4 | 4.31 |
| 1886Q1 | 4.03 |
| 1886Q2 | 4.05 |
| 1886Q3 | 4.98 |
| 1886Q4 | 5.98 |
| 1887Q1 | 5.22 |
| 1887Q2 | 5.24 |
| 1887Q3 | 6.49 |
| 1887Q4 | 6.04 |
| 1888Q1 | 5.22 |
| 1888Q2 | 4.82 |
| 1888Q3 | 4.57 |
| 1888Q4 | 4.93 |
| 1889Q1 | 4.47 |
| 1889Q2 | 4.01 |
| 1889Q3 | 4.95 |
| 1889Q4 | 6.03 |
| 1890Q1 | 5.29 |
| 1890Q2 | 5.06 |
| 1890Q3 | 5.46 |
| 1890Q4 | 6.66 |
| 1891Q1 | 5.36 |
| 1891Q2 | 5.32 |
| 1891Q3 | 5.71 |
| 1891Q4 | 5.16 |
| 1892Q1 | 3.94 |
| 1892Q2 | 3.19 |
| 1892Q3 | 4.05 |
| 1892Q4 | 5.27 |
| 1893Q1 | 5.62 |
| 1893Q2 | 6.97 |
| 1893Q3 | 9.92 |
| 1893Q4 | 4.65 |
| 1894Q1 | 3.27 |
| 1894Q2 | 2.98 |
| 1894Q3 | 3.12 |
| 1894Q4 | 2.81 |
| 1895Q1 | 3.54 |
| 1895Q2 | 3.12 |
| 1895Q3 | 3.48 |
| 1895Q4 | 4.55 |
| 1896Q1 | 5.68 |
| 1896Q2 | 4.68 |
| 1896Q3 | 7.05 |
| 1896Q4 | 5.81 |
| 1897Q1 | 3.21 |
| 1897Q2 | 3.39 |
| 1897Q3 | 3.74 |
| 1897Q4 | 3.66 |
| 1898Q1 | 3.67 |
| 1898Q2 | 4.55 |
| 1898Q3 | 3.8 |
| 1898Q4 | 3.25 |
| 1899Q1 | 3.27 |
| 1899Q2 | 3.53 |
| 1899Q3 | 4.32 |
| 1899Q4 | 5.5 |
| 1900Q1 | 4.74 |
| 1900Q2 | 3.88 |
| 1900Q3 | 4.19 |
| 1900Q4 | 4.73 |
| 1901Q1 | 3.82 |
| 1901Q2 | 3.96 |
| 1901Q3 | 4.58 |
| 1901Q4 | 4.17 |
| 1902Q1 | 4.28 |
| 1902Q2 | 4.48 |
| 1902Q3 | 5.02 |
| 1902Q4 | 5.9 |
| 1903Q1 | 5.07 |
| 1903Q2 | 5.02 |
| 1903Q3 | 5.79 |
| 1903Q4 | 5.89 |
| 1904Q1 | 4.78 |
| 1904Q2 | 3.86 |
| 1904Q3 | 3.91 |
| 1904Q4 | 4.27 |
| 1905Q1 | 3.91 |
| 1905Q2 | 3.9 |
| 1905Q3 | 4.35 |
| 1905Q4 | 5.45 |
| 1906Q1 | 5.12 |
| 1906Q2 | 5.34 |
| 1906Q3 | 5.99 |
| 1906Q4 | 6.27 |
| 1907Q1 | 6.09 |
| 1907Q2 | 5.62 |
| 1907Q3 | 6.27 |
| 1907Q4 | 7.39 |
| 1908Q1 | 5.74 |
| 1908Q2 | 4.01 |
| 1908Q3 | 3.76 |
| 1908Q4 | 3.98 |
| 1909Q1 | 3.58 |
| 1909Q2 | 3.4 |
| 1909Q3 | 3.87 |
| 1909Q4 | 5.06 |
| 1910Q1 | 4.56 |
| 1910Q2 | 4.81 |
| 1910Q3 | 5.45 |
| 1910Q4 | 5.24 |
| 1911Q1 | 3.98 |
| 1911Q2 | 3.65 |
| 1911Q3 | 4.17 |
| 1911Q4 | 4.31 |
| 1912Q1 | 3.95 |
| 1912Q2 | 4.11 |
| 1912Q3 | 5.03 |
| 1912Q4 | 5.88 |
| 1913Q1 | 5.16 |
| 1913Q2 | 5.58 |
| 1913Q3 | 5.95 |
| 1913Q4 | 5.64 |
| 1914Q1 | 4.07 |
| 1914Q2 | 3.83 |
| 1914Q3 | 5.83 |
| 1914Q4 | 5.43 |
Source: Frederick R. Macaulay, Some Theoretical Problems Suggested by the Movements of Interest Rates. Bond Yields, and Stock Prices in the United States since 1856 (New York: National Bureau of Economic Research, 1938) A 144–56.
Notes: CV (1866–1878) = 0.2804
CV (1879–1914) = 0.2138
MAPC (1866–1878) = 1.145
MAPC (1879–1914) = 0.673
MNPC (1866–1878) = 0.041
MNPC (1879–1914) = 0.017
Table 23
Commercial Paper Rate of Interest at New York City, 1919–1932 (an average of observed rates, by quarter)
| Quarter | Rate |
| 1919Q1 | 5.29 |
| 1919Q2 | 5.34 |
| 1919Q3 | 5.38 |
| 1919Q4 | 5.46 |
| 1920Q1 | 6.42 |
| 1920Q2 | 7.8 |
| 1920Q3 | 8.13 |
| 1920Q4 | 8.09 |
| 1921Q1 | 7.71 |
| 1921Q2 | 7.09 |
| 1921Q3 | 6.17 |
| 1921Q4 | 5.5 |
| 1922Q1 | 4.88 |
| 1922Q2 | 4.42 |
| 1922Q3 | 4.13 |
| 1922Q4 | 4.67 |
| 1923Q1 | 4.75 |
| 1923Q2 | 5.13 |
| 1923Q3 | 5.21 |
| 1923Q4 | 5.17 |
| 1924Q1 | 4.88 |
| 1924Q2 | 4.42 |
| 1924Q3 | 3.29 |
| 1924Q4 | 3.34 |
| 1925Q1 | 3.75 |
| 1925Q2 | 3.92 |
| 1925Q3 | 4.04 |
| 1925Q4 | 4.38 |
| 1926Q1 | 4.34 |
| 1926Q2 | 4.13 |
| 1926Q3 | 4.34 |
| 1926Q4 | 4.54 |
| 1927Q1 | 4.17 |
| 1927Q2 | 4.17 |
| 1927Q3 | 4.08 |
| 1927Q4 | 4 |
| 1928Q1 | 4.04 |
| 1928Q2 | 4.54 |
| 1928Q3 | 5.38 |
| 1928Q4 | 5.42 |
| 1929Q1 | 5.59 |
| 1929Q2 | 6 |
| 1929Q3 | 6.13 |
| 1929Q4 | 5.67 |
| 1930Q1 | 4.63 |
| 1930Q2 | 3.71 |
| 1930Q3 | 3.08 |
| 1930Q4 | 2.92 |
| 1931Q1 | 2.67 |
| 1931Q2 | 2.21 |
| 1931Q3 | 2 |
| 1931Q4 | 3.67 |
| 1932Q1 | 3.8 |
| 1932Q2 | 3.13 |
| 1932Q3 | 2.29 |
| 1932Q4 | 1.71 |
Source: Board of Governors of the Federal Reserve System, Banking and Monetary Statistics, 1914–1941 (Washington, D.C.: Government Printing Office, 1943) 449–51.
Notes: CV = 0.3071
MAPC = 0.405
MNPC = 0.065
Table 24
Commercial Paper Rate of Interest at New York City, 1959–1989 (an average of observed rates, by quarter)
| Quarter | Rate |
| 1959Q1 | 3.3 |
| 1959Q2 | 3.6 |
| 1959Q3 | 4.19 |
| 1959Q4 | 4.76 |
| 1960Q1 | 4.69 |
| 1960Q2 | 4.07 |
| 1960Q3 | 3.37 |
| 1960Q4 | 3.27 |
| 1961Q1 | 3.01 |
| 1961Q2 | 2.86 |
| 1961Q3 | 2.9 |
| 1961Q4 | 3.06 |
| 1962Q1 | 3.24 |
| 1962Q2 | 3.2 |
| 1962Q3 | 3.33 |
| 1962Q4 | 3.26 |
| 1963Q1 | 3.31 |
| 1963Q2 | 3.32 |
| 1963Q3 | 3.7 |
| 1963Q4 | 3.91 |
| 1964Q1 | 3.95 |
| 1964Q2 | 3.93 |
| 1964Q3 | 3.91 |
| 1964Q4 | 4.06 |
| 1965Q1 | 4.3 |
| 1965Q2 | 4.38 |
| 1965Q3 | 4.38 |
| 1965Q4 | 4.47 |
| 1966Q1 | 4.97 |
| 1966Q2 | 5.43 |
| 1966Q3 | 5.79 |
| 1966Q4 | 6 |
| 1967Q1 | 5.45 |
| 1967Q2 | 4.72 |
| 1967Q3 | 4.97 |
| 1967Q4 | 5.3 |
| 1968Q1 | 5.58 |
| 1968Q2 | 6.08 |
| 1968Q3 | 5.96 |
| 1968Q4 | 5.96 |
| 1969Q1 | 6.66 |
| 1969Q2 | 7.54 |
| 1969Q3 | 8.49 |
| 1969Q4 | 8.62 |
| 1970Q1 | 8.55 |
| 1970Q2 | 8.17 |
| 1970Q3 | 7.84 |
| 1970Q4 | 6.29 |
| 1971Q1 | 4.59 |
| 1971Q2 | 5.04 |
| 1971Q3 | 5.74 |
| 1971Q4 | 5.07 |
| 1972Q1 | 4.06 |
| 1972Q2 | 4.58 |
| 1972Q3 | 4.93 |
| 1972Q4 | 5.33 |
| 1973Q1 | 6.3 |
| 1973Q2 | 7.47 |
| 1973Q3 | 9.87 |
| 1973Q4 | 8.98 |
| 1974Q1 | 8.3 |
| 1974Q2 | 10.46 |
| 1974Q3 | 11.53 |
| 1974Q4 | 9.05 |
| 1975Q1 | 6.56 |
| 1975Q2 | 5.92 |
| 1975Q3 | 6.67 |
| 1975Q4 | 6.12 |
| 1976Q1 | 5.29 |
| 1976Q2 | 5.57 |
| 1976Q3 | 5.53 |
| 1976Q4 | 4.99 |
| 1977Q1 | 4.81 |
| 1977Q2 | 5.24 |
| 1977Q3 | 5.81 |
| 1977Q4 | 6.59 |
| 1978Q1 | 6.8 |
| 1978Q2 | 7.2 |
| 1978Q3 | 8.08 |
| 1978Q4 | 9.9 |
| 1979Q1 | 10.10 |
| 1979Q2 | 9.85 |
| 1979Q3 | 10.60 |
| 1979Q4 | 13.10 |
| 1980Q1 | 14.25 |
| 1980Q2 | 10.75 |
| 1980Q3 | 9.65 |
| 1980Q4 | 14.51 |
| 1981Q1 | 14.52 |
| 1981Q2 | 15.35 |
| 1981Q3 | 16.21 |
| 1981Q4 | 12.94 |
| 1982Q1 | 13.70 |
| 1982Q2 | 13.48 |
| 1982Q3 | 11.55 |
| 1982Q4 | 8.81 |
| 1983Q1 | 8.34 |
| 1983Q2 | 8.61 |
| 1983Q3 | 9.44 |
| 1983Q4 | 9.19 |
| 1984Q1 | 9.45 |
| 1984Q2 | 10.77 |
| 1984Q3 | 11.15 |
| 1984Q4 | 9.26 |
| 1985Q1 | 8.69 |
| 1985Q2 | 7.91 |
| 1985Q3 | 7.72 |
| 1985Q4 | 7.7 |
| 1986Q1 | 7.41 |
| 1986Q2 | 6.54 |
| 1986Q3 | 5.89 |
| 1986Q4, | 5.73 |
| 1987Q1 | 5.95 |
| 1987Q2 | 6.85 |
| 1987Q3 | 7.03 |
| 1987Q4 | 7.54 |
| 1988Q1 | 6.71 |
| 1988Q2 | 7.25 |
| 1988Q3 | 8.16 |
| 1988Q4 | 8.59 |
| 1989Q1 | 9.45 |
| 1989Q2 | 9.29 |
| 1989Q3 | 8.39 |
| 1989Q4 | 8.06 |
Source: Various issues of the Survey of Current Business.
Notes: CV = 0.4264
MAPC = 0.686
MNPC = 0.019
Table 25
Index of Common Stock Prices, 1835–1860 (1853 = 100)
| Year | Index |
| 1835 | 109.7 |
| 1836 | 97.7 |
| 1837 | 80.5 |
| 1838 | 74.4 |
| 1839 | 73.5 |
| 1840 | 67.4 |
| 1841 | 64.3 |
| 1842 | 49 |
| 1843 | 55.6 |
| 1844 | 84.2 |
| 1845 | 87.9 |
| 1846 | 86.1 |
| 1847 | 89.4 |
| 1848 | 80.8 |
| 1849 | 81.4 |
| 1850 | 84.9 |
| 1851 | 92.4 |
| 1852 | 98.7 |
| 1853 | 100.0 |
| 1854 | 85.2 |
| 1855 | 74.3 |
| 1856 | 69.9 |
| 1857 | 59.4 |
| 1858 | 55 |
| 1859 | 50 |
| 1860 | 59.6 |
Source: Walter B. Smith and Arthur H. Cole, Fluctuations in American Business, 1790–1860, 1935, 183–84.
Notes: Composed entirely of railroad stocks, which were the most important stocks of the time.
CV (1835–1849) = 0.1916
CV (1850–1860) = 0.2270
MAPC (1835–1849) = 11.398
MAPC (1850–1860) = 9.591
MNPC (1835–1849) = 0.844
MNPC (1850–1860) = 2.233
In terms of industrial production, free banking exhibits strong growth, being exceeded only by the era of chartered banking, which preceded it. Of course, the early years of the Fed—including as they do the onset of the Great Depression34—show a net decline in industrial production.
A Tentative Judgment
Overall, what picture of free banking is painted by the data presented in this section? One sees that money supply fluctuations were often large in the short run but showed little secular trend over the long run. This pattern was repeated for interest rates. Regarding both consumer prices and wholesale prices, free banking was among the top three periods by every measure. Stock prices during the heyday of free banking were only modestly stable, however. In relative terms, free banking was accompanied by little growth in real national income but substantial growth in industrial production.
Table 26
Index of Common Stock Prices, 1866–1878 (1900 = 100)
| Year | Index |
| 1866 | 58.28 |
| 1867 | 59.2 |
| 1868 | 66.63 |
| 1869 | 72.48 |
| 1870 | 73.5 |
| 1871 | 76.57 |
| 1872 | 80.66 |
| 1873 | 77 |
| 1874 | 72.69 |
| 1875 | 70.67 |
| 1876 | 64.44 |
| 1877 | 49.55 |
| 1878 | 53.68 |
Source: Frederick R. Macaulay, Some Theoretical Problems Suggested by the Movements of Interest Rates, Bond Yields, and Stock Prices in the united States since 1856 (New York: National Bureau of Economic Research, 1938) A 144–47.
Notes: Composed entirely of railroad stocks, which were the most important stocks of the time.
CV = 0.1379
MAPC = 7.251
MNPC = 0.222
One might think about two benchmarks: the classical gold standard (1879–1914) and the modern Fed (1959–1989). The former is often lauded as a period of considerable stability and prosperity. The latter should—if the apologists for central banking are correct—possess all the virtues that have so often been alleged to have been absent from free banking. Free banking exhibited greater long-run stability of the money supply, interest rates, and stock prices than either the classical gold standard or the modern Fed. The variability of (1) consumer prices (both short- and long-run) and (2) wholesale prices (long-run) was less under free banking than during the years 1959–1989. Also, the growth rate of industrial production exceeded that of either the classical gold standard or the modern Fed.
One must be cautious in drawing conclusions from the data presented here. First of all, much of those data have been resurrected from the rather distant past and, as such, their reliability is open to question. However, as a partial counterpoint, it may be noted that the results for the money supply under free banking remain robust for three different estimates of that series. Second, the American experiment with free banking was not a pure test of the application of laissez-faire principles to banking. Be that as it may, and assuming the data are reasonably accurate, it appears that American free banking was indeed consistent with a significant degree of (particularly long-run) macroeconomic stability, contrary to popular belief.
Table 27
Standard & Poor’s Index of Common Stock Prices, 1879–1932 (1941–1943 = 100)
| Year | Index |
| 1879 | 41.2 |
| 1880 | 52.1 |
| 1881 | 62.5 |
| 1882 | 59 |
| 1883 | 56.3 |
| 1884 | 47.4 |
| 1885 | 46 |
| 1886 | 53.6 |
| 1887 | 55.3 |
| 1888 | 52 |
| 1889 | 53.2 |
| 1890 | 52.7 |
| 1891 | 50.3 |
| 1892 | 55.5 |
| 1893 | 47.8 |
| 1894 | 43.9 |
| 1895 | 45.3 |
| 1896 | 42.3 |
| 1897 | 44.5 |
| 1898 | 50.5 |
| 1899 | 62.9 |
| 1900 | 61.5 |
| 1901 | 78.4 |
| 1902 | 84.2 |
| 1903 | 72.1 |
| 1904 | 70.5 |
| 1905 | 89.9 |
| 1906 | 96.4 |
| 1907 | 78.4 |
| 1908 | 77.8 |
| 1909 | 97.1 |
| 1910 | 93.5 |
| 1911 | 92.4 |
| 1912 | 95.3 |
| 1913 | 85.1 |
| 1914 | 80.8 |
| 1919 | 87.8 |
| 1920 | 79.8 |
| 1921 | 68.6 |
| 1922 | 84.1 |
| 1923 | 85.7 |
| 1924 | 90.5 |
| 1925 | 111.5 |
| 1926 | 125.9 |
| 1927 | 153.4 |
| 1928 | 199.5 |
| 1929 | 260.2 |
| 1930 | 210.3 |
| 1931 | 136.6 |
| 1932 | 69.3 |
Source: Historical Statistics of the United States, 1975, Series X 492–98.
Notes: The war years 1915–1918 are excluded.
CV (1879–1914) = 0.2769
CV (1919–1932) = 0.4568
MAPC (1879–1914) = 10.020
MAPC (1919–1932) = 21.167
MNPC (1879–1914) = 2.709
MNPC (1919–1932) = 1.685
The impression has been assiduously cultivated by the opponents of a banking currency that the early American banks issued a volume of circulating notes enormously in excess of the legitimate demands of business. This impression is absolutely unfounded and the proof is afforded by the figures. (Conant 1927, 394)
CONCLUSION
Neither the Scottish nor the American episode of free banking was a case of true laissez-faire banking. They were only approximations to the model. The various shortcomings of and restrictions upon Scottish free banking were enumerated in Chapter 5. As for the United States, there were few interest rate controls, impositions of unlimited liability, prohibitions of small-denomination notes, or specially privileged banks. However, the American free banks labored under two key restrictions that were absent from the Scottish system: (1) They were required to hold state bonds as security for their note issues, and (2) they were subject to severe limitations on both intrastate and interstate branching. These restrictions had potentially adverse effects on both the assets and liabilities of banks.
Table 28
Standard & Poor’s Index of Common Stock Prices, 1959–1989 (1941–1943 = 100)
| Year | Index |
| 1959 | 573.8 |
| 1960 | 558.5 |
| 1961 | 662.7 |
| 1962 | 623.8 |
| 1963 | 698.7 |
| 1964 | 813.7 |
| 1965 | 881.7 |
| 1966 | 852.6 |
| 1967 | 919.3 |
| 1968 | 987 |
| 1969 | 978.4 |
| 1970 | 832.2 |
| 1971 | 982.9 |
| 1972 | 1092.0 |
| 1973 | 1074.3 |
| 1974 | 828.4 |
| 1975 | 861.6 |
| 1976 | 1020.1 |
| 1977 | 982.0 |
| 1978 | 960.2 |
| 1979 | 1030.1 |
| 1980 | 1187.8 |
| 1981 | 1280.4 |
| 1982 | 1197.1 |
| 1983 | 1604.1 |
| 1984 | 1604.6 |
| 1985 | 1868.4 |
| 1986 | 2363.4 |
| 1987 | 2868.3 |
| 1988 | 2657.9 |
| 1989 | 3228.4 |
Sources: Historical Statistics of the United States, 1975, Series X 492–98. Various issues of the Survey of Current Business.
Notes: CV = 0.5468
MAPC = 11.480
MNPC = 6.683
The bond security requirement forced a connection between banks’ issuance of notes (their chief liability) and the holding of state bonds. Although the “ostensible aim of the restriction was to reduce the noteholders’ exposure to losses, its actual impact may well have been the reverse” (White 1986, 892). This seemingly paradoxical result stemmed from the following: (1) Allowing free banks to buy state bonds at market value and then to issue notes based on the face value of those bonds encouraged some degree of fraud, a possibility discussed earlier in this chapter; (2) the bond security requirement forced free banks to be exposed unnecessarily to the risk of falling state bond prices; that is, it caused them to hold suboptimal asset portfolios; and (3) since free banks could only issue notes in proportion to their existing bond holdings, in the short run, they sometimes could not deal with a currency run by merely issuing more notes—they might have to pay out specie in order to satisfy consumer demand for cash (White 1986, 892–94). This last item is a particularly important violation of the model in Chapters 2 and 3. In that model, an advantage of free banking over central banking was the capacity of individual banks to issue notes on short notice so as to keep currency runs from becoming redemption runs. In practice, the bond restriction significantly reduced such a theoretical advantage.
Table 29
Real National Income per Capita, 1839–1932 (in 1860 prices)
| Year | Income | Percent Change |
| 1839–1840 | $101.45 | ------- |
| 1849–1850 | 119.02 | +17.3 |
| 1859–1860 | 136.45 | +14.6 |
| 1869 | 107.65 | -21.1 |
| 1879 | 151.72 | +40.9 |
| 1889 | 198.88 | +31.1 |
| 1897–1901 | 238.33 | +19.8 |
| 1907–1911 | 283.30 | +18.9 |
| 1919 | 308.53 | + 8.9 |
| 1929 | 476.18 | +54.3 |
| 1932 | 336.78 | -29.3 |
Sources: National Bureau of Economic Research, Trends in the American Economy in the Nineteenth Century (Princeton: Princeton University Press, 1960) 360. Historical Statistics of the United States, 1975, Series F 6–8, F 250–261.
Notes: Nominal income was adjusted using the price indexes from Table 17.
Average Annual Percentage Change:
(1839–1849) = +1.61
(1849–1859) = +1.37
(1859–1879) = +0.53
(1869–1879) = +3.49
(1879–1907/1911) = +2.10
(1919–1932) = +0.68
Table 30
Real National Income per Capita, 1959–1989 (in 1967 prices)
| Year | Income | Percent Change |
| 1959 | $2,363 | ----- |
| 1969 | 3,595 | +52.1 |
| 1979 | 3,713 | + 3.3 |
| 1989 | 5,065 | +36.4 |
Sources: Various issues of the Survey of Current Business. Statistical Abstract of the United States, 1990, 7.
Notes: Nominal income was adjusted using the price indexes from Table 18.
Average annual percentage change
(1959–1989) = +2.57
Donald Wells and Leslie Scruggs point out the very real practical benefit that accrues to banks and their customers when banks are granted the legal right to issue notes. Oddly enough, the event they cite is the liquidity crisis of the winter of 1914–1915 (1986a, 907–8). Banks were at that time permitted to issue currency under the provisions of the Aldrich-Vreeland Act of 1908 (amended in 1913 and 1914). The result was that “currency was issued in exchange for deposits when the public demanded it, and was extinguished when demand subsided. Banks were able to conserve their gold certificates and greenbacks as reserves by paying out the bank notes for use as hand-to-hand currency . . . the Aldrich-Vreeland Act worked far better than the Fed did in the early 1930s” (Wells and Scruggs 1986a, 908).
The other major distorting restriction on American free banks was the widespread prohibition on establishing branch offices. White explains that restrictions on branching not only prevent banks from holding properly diversified asset portfolios, but they also affect banks’ liabilities. This follows from the fact that “a branched bank is less susceptible to random withdrawals or even runs by depositors in any area because it has a broad source of deposits (it can rely on the law of large numbers) and can transfer reserves from surplus to deficit branches. A unit bank rests on a narrower and less stable base” (White 1986, 895–96).35 Stephen Williamson concurs: “The unit banking restriction results in a banking system in which banks are less diversified than they would be otherwise. These banks are therefore more sensitive to idiosyncratic shocks, and they experience runs and fail with higher probability” (1989, 22).
The negative impact of branching restrictions may be seen by comparing the American and Canadian systems. Canadian banks in the period 1870 to 1913 were subject to no branching restrictions (Williamson 1989, 23). American banks of the same period (the National Banking System era) could rarely establish branch offices. The numbers of failures in the two systems were dramatically different. “Canada had 23 bank liquidations while, at the same time, the United States had 3,208” (Williamson 1989, 24). Furthermore, it is well known that between 1921 and 1933, a time of continuing limitations on branching, almost 15,000 American banks failed. In stark contrast, “between 1923 and 1985, no Canadian banks failed” (emphasis added) (Williamson 1989, 24). Wells and Scruggs summarize the costs of antibranching statutes very well when they note that to limit banks’ ability to branch is to encourage the formation of banks that are small, inefficient, poorly diversified, undercapitalized, and unstable (1986a, 900).
Table 31
Index of Industrial Output per Capita, 1829–1932 (1899 = 100)
| Year(s) | Index | Percent Change |
| 1829–1839 | 11.4* | ----- |
| 1839–1849 | 18.1* | + 58.8 |
| 1849–1859 | 27.6* | + 52.5 |
| 1866 | 43.0 | + 55.8 |
| 1879 | 54.7 | + 27.2 |
| 1914 | 144.2 | +163.6 |
| 1919 | 163.2 | + 13.2 |
| 1929 | 229.7 | + 40.7 |
| 1932 | 121.6 | - 47.1 |
Sources: Historical Statistics of the United States, 1975, Series A 6–8. Edward Ames, “Trends, Cycles, and Stagnation in U.S. Manufacturing since 1860” (Oxford Economic Papers, Vol. XI, No. 3, October 1959) 272. W. W. Rostow, The Process of Economic Growth (London: Oxford University Press, 1960) 350.
Notes: *Represents the annual average for the period.
Average Annual Percentage Change:
(1829/39–1839/49) = +4.73
(1839/49–1849/59) = +4.31
(1866–1879) = +1.87
(1879–1914) = +2.81
(1919–1932) = -2.24
Table 32
Index of Industrial Output per Capita, 1959–1989 (1987 = 100)
| Year | Index | Percent Change |
| 1959 | 51.26 | ----- |
| 1969 | 76.75 | +49.73 |
| 1979 | 92.89 | +21.03 |
| 1989 | 105.99 | +14.10 |
Sources: Various issues of the Survey of Current Business. Statistical Abstract of the United States. 1990, 7.
Note: Average annual percentage change
(1959–1989) = +2.45
It is obvious from the above that American free banks were burdened with important restrictions and, thus, did not constitute a pure laissez-faire banking system. One begins to marvel at how relatively well free banking in the United States did perform in the context of such constraints.
NOTES
1. The beginning of the period dates from Michigan’s adoption of the first free-banking statute in 1837. The end of the era is usually identified with the passage of the National Currency Act of 1863.
2. Insofar as free banking is concerned, this statement must be discounted somewhat. The reason is that, although Ohio and Indiana did adopt free banking in the early 1850s, Missouri never enacted a free banking law (Rockoff 1975, 3).
3. This was the year New York first allowed free banking.
4. Hepburn suggests that this term originated because the notoriously unsound banks in Michigan often used “a dog in red color and the wild cat” as decorative devices on their notes (1903, 138).
5. These bonds usually had to be deposited with the state auditor.
6. The precise figures are 47.8 percent (339) failed and 30.7 percent (104) of those did not redeem their notes at par. That is, 14.7 percent of the total banks failed below par.
7. That is if one defines inflation in the conventional way as a rise in the price level. It is perhaps preferable to define inflation as an excess supply of money.
8. In New York, there were fifty-five free banks that lasted fifteen years or more.
9. Capital gains and losses would be shared with noteholders if banknotes were, for example, mutual fund shares rather than debt instruments.
10. Despite this apparent benefit, it should not be forgotten that the bond requirement seriously compromised the safety of U.S. free banks by imposing on them an inadequately diversified asset portfolio. The health of free banks was, to a considerable extent, tied to the degree of fiscal responsibility exhibited by the various state legislatures.
11. Of course, there are economists—even Nobel prize winners—who reject the idea that free markets are efficient. The author is not among them. This writer perceives free markets as both just and efficient, though not perhaps in the Pareto-optimal sense.
12. It is clear that Rockoff is thinking in terms of the conventional “perfect competition” model in which there are a large number of firms, there is ease of entry and exit, and the long-run expected rate of economic profit is zero.
13. See Historical Statistics of the United States, 1975, Series E 135–66.
14. See the Economic Report of the President, 1991.
15. This is not to deny that banks need to be adequately capitalized. However, given that option clauses were not used in the United States and free banks were required by law to redeem in specie on demand, it seems clear that the more crucial factor was adequate primary reserves.
16. This was Michigan, which repealed its free-banking law in 1839, but reintroduced free banking in 1857.
17. This is one reason why one might argue that “free banking” lasted from the late 1830s until the Civil War. The problem with that is that the necessity of a charter must have curtailed the flexibility of the currency supply, as well as distorting it in the direction of politically favored persons. Given that both transportation and communication were somewhat primitive and the states prohibited branching, an adequate supply of notes required a large number of banks. This obviously would not be true today.
18. One suspects that these data are based on book values rather than market values, but the source gives no indication. The difference could be large and therefore quite important.
19. See Chapter 8.
20. The effect of the antibranching statutes will be discussed later in this chapter.
21. Nothing said here is meant to disparage microeconomics. Indeed, the author would argue that a macroeconomics that is not firmly rooted in microprinciples is no economics at all.
22. The author reasons within a monetarist/Austrian context in which monetary instability has both nominal and real effects. See Chapter 3.
23. See Christina Romer (1986a).
24. This represents the interval between (1) the demise of the second Bank of the United States as the government’s fiscal agent (and embryonic central bank) and (2) the proliferation of free banking in the 1850s.
25. The United States did not, of course, enter the war formally until April of 1917, but the American economy was disrupted long before then. See Benjamin Anderson (1979, 25–43).
26. Symbolically, these are:
except for interest rates, where
except for interest rates, where
27. One could make the argument that wars should be included because they are made possible (or more likely) by changes in the institutional structure. That is, wars may be—at least in part—the product of the particular monetary regime. For example, the Civil War prompted the rejection of free banking, and World War I brought on the abandonment of the classical gold standard.
28. Very few free banks seem to have been established in Massachusetts and Pennsylvania, however.
29. See Historical Statistics of the United States, 1975, Series A 6–8, A 195–209, and Rockoff (1975, 3). Based on a linear interpolation of census figures, in 1853 13,888,700 (54.0 percent) of the 25,736,000 residents lived under free banking.
30. New York’s rivals were Boston and Philadelphia.
31. Total figures are available from the author upon request.
32. Such data extends into the eighteenth century (Smith and Cole 1935).
33. See Kevin Dowd (1992, 3), for example.
34. Some economists seem to think of the depression as some sort of “natural disaster.” Given that very strange premise, they argue that the depression years should not be included in any series on the basis of which comparative economic judgments are to be made. That is nonsense. The Federal Reserve clearly played a large role in the precipitation of the depression and should be called to task for it.
35. A “unit bank” is one that has no branch offices.
Free Banking: Theory, History, and a Laissez-Faire Model
Read the whole book online · Book details
Free to read online and to download from this archive.