¹ Lerner, Eugene, “The Monetary and Fiscal Programs of the Confederate Government, 1861–1865,”Journal of Political Economy, 62 (12 1954), pp. 506–22.CrossRefGoogle Scholar

² Calomiris, Charles, “Price and Exchange Rate Determination during the Greenback Suspension,” Oxford Economic Papers, 40 (12 1988), pp. 719–50.CrossRefGoogle Scholar

³ Roll, Richard, “Interest Rates and Price Expectations During the Civil War,” this JOURNAL, 32 (06 1972), pp. 476–98.Google Scholar

⁴ See Calomiris, Charles W., “Institutional Failure, Monetary Scarcity, and the Depreciation of the Continental,” this JOURNAL, 48 (03 1988), pp. 47–68Google Scholar; Calomiris, “Price and Exchange Rate Determination”; Smith, Bruce, “Some Colonial Evidence on Two Theories of Money: Maryland and the Carolinas,” Journal of Political Economy, 93 (12 1985), pp. 1178–1211CrossRefGoogle Scholar; Smith, Bruce, “American Colonial Monetary Regimes: The Failure of the Quantity Theory of Money and Some Evidence in Favor of an Alternative view,” Canadian Journal of Economics, 18 (08 1985), pp. 531–65CrossRefGoogle Scholar; and Wicker, Elmus, “Colonial Monetary Standards Contrasted: Evidence from the Seven Years' War,” this JOURNAL, 45 (12 1985), pp. 869–84.Google Scholar

⁵ See the rational expectations literature for the theoretical underpinnings including Wallace, Neil, “A Miller-Modigliani Theorem for Open Market Operations,” American Economic Review, 71 (06 1981), pp. 267–74Google Scholar; and Barro, Robert J., “Are Government Bonds Net Wealth?” Journal of Political Economy, 82 (11 1974), pp. 1095–1117.CrossRefGoogle Scholar

⁶ Godfrey, John Munro, Monetary Expansion in the Confederacy (New York, 1978), p. 14. The absence of efficient tax collections also explains the South's reliance on the inflation tax to finance their war effort.Google Scholar

⁷ Late in the war, when military defeat became almost certain, the currency could only be supported by the prospect of near-term tax payments. For an analysis of the use of taxes to support Confederate money late in the war, see Pecquet, Gary M., “Money in the Trans-Mississippi and the Confederate Currency Reform Act of 1864,” Explorations in Economic History, 24 (04 1987), pp. 218–43.CrossRefGoogle Scholar

⁸ The demand for Confederate treasury notes initially depended upon making them receivable for state dues and taxes and forcing banks to suspend specie payments and accept the new currency. In this way the treasury notes were placed into circulation. See Bragg, Jefferson Davis, Louisiana in the Confederacy (Baton Rouge, 1941), pp. 67–72Google Scholar; Morgan, James F., Graybacks and Gold:Confederate Monetary Policy (Pensacola, 1985), pp. 23–25Google Scholar, Schweikart, Larry, “Secession and Southern Banks,” Civil War History, 31 (06 1985), pp. 111–25CrossRefGoogle Scholar; and Schweikart, Larry L., Banking in the American South: From the Age of Jackson to Reconstruction (Baton Rouge, 1987), pp. 293–95. For a listing of all Confederate securitiesGoogle Scholar, see Thian, Raphael P., compiler, Register of the Confederate Debt (Boston, 1972 [originally published in 1880]).Google Scholar

⁹ The 7.3 percent interest note accrued interest at the rate of 2 percent per day on the $100 par value. At semiannual intervals the note could be brought to the treasury and stamped “paid.” See Godfrey, Monetary Expansion in the Confederacy, p. 8. A similar class of 3.65 percent notes were also issued, but only in comparatively small quantities. The Currency Reform Act of 1864 provided that the 7.3 percent notes would no longer be tax receivable and would be payable two years after a successful peace treaty.Google Scholar See Pecquet, “Money in the Trans-Mississippi,” p. 224. It is perhaps noteworthy that the North also experimented with 3.65 percent and 7.3 percent interest-bearing notes.Google Scholar See Mitchell, W. C., A History of the Greenbacks (Chicago, 1903), p. 16.Google Scholar

¹⁰ See Todd, Richard C., Confederate Finance (Athens, 1954), pp. 245, 253Google Scholar; Godfrey, Monetary Expansion in the Confederacy, p. 9Google Scholar; and Yearns, W. B., The Confederate Congress (Athens, 1960), p. 195. The North also experimented with call certificates for a short time.Google Scholar See Mitchell, A History of the Greenbacks.Google Scholar

¹¹ Call certificates were issued and redeemed quite freely. Confederate treasury records indicate the exact totals between Oct. 13, 1862, and Jan. I, 1863. See Thian, Raphael P., compiler, Reports of the Secretary of the Confederate States of America 1861–1865, Appendix Part 3 (Washington, DC, 1878 (on microfilm at the National Archives)), pp. 136–38.Google Scholar Quarterly estimates of the number of outstanding call certificates and other monetary aggregates can be found in Godfrey, Monetary Expansion in the Confederacy, pp. 40–46.Google Scholar

¹² Todd, Confederate Finance, p. 26.Google Scholar

¹³ Ibid.

¹⁴ Ibid.

¹⁵ Ibid., pp. 74–75, 83. Of less significance for our purposes are certain special experimental classes of bonds including 6 percent cotton interest bonds, the Erlanger loan from Europe, 6 percent “income tax” bonds, and an 1864 issue of 6 percent nontaxable bonds. Ibid., pp. 25–85.

¹⁶ As late as Sept. 1864 the second Confederate treasury secretary seriously contemplated and described how the Confederate debt could be retired following a successful peace treaty. See Thian, Raphael P., compiler, Correspondence of the Treasury Department of the Confederate States of America (Washington, DC, 1879 [on microfilm at the National Archives]), pp. 754–57. Also note that the attempt to minimize the postwar interest burden implies that, in the absence of an efficient tax collection system, money creation and the attendant changes in prices bear the brunt of government appropriations.Google Scholar

¹⁷ The equation used to calculate the nominal and gold bond yields is where P^B is the price of the bond, c is the coupon payment, N is the term to maturity, F is the face value, and i is the yield to maturity. For the calculation of the nominal (gold) yield P^B, c, and F are expressed in Confederate dollars (gold).Google Scholar

¹⁸ Calls could be exchanged for Confederate currency, but in the event of a southern defeat the currency would also default.Google Scholar

¹⁹ This premium is relatively small. Its magnitude reflects the degree of substitutability between bonds and calls.Google Scholar

²⁰ According to our argument, expected inflation is the difference between the pegged nominal rate of interest and the real rate of interest. If default risk on bonds raised the real rate of interest above the nominal rate, then deflation would be implied. It is possible that this would make the real return on bonds negative. However, this would not necessarily lead to a conversion of calls into money because the increase in risk on bonds is matched by an increase in the riskiness of holding currency. The relative risks, and hence the relative demands, for calls and money are essentially unaffected.Google Scholar

²¹ The Fisher effect refers to the change in nominal rates of interest when expected inflation changes. See Fisher, Irving, Theory of Interest (New York, 1930). In the present context a Fisher-type equation can be obtained with an argument similar to the reasoning that lies behind the interest-rate parity results of international finance. Specifically, the parity analogy leads to i^G + Δ_p = i^D, where i^G(i^D) is the gold (dollar) yield and δ^p is the expected rate of change in the price of gold. There is, however, a complication with this analogy. If the Confederacy eventually paid its obligation in gold, then, ignoring coupon payments, Δ^p would equal zero. Also, since payment was uncertain there should also be a rather hefty risk premium.Google Scholar

²² See Calomiris, “Price and Exchange Rates,” for similar reasoning applied to the northern experience. We are grateful to an anonymous referee for this argument and the one that directly follows in the text.Google Scholar

²³ This implication could be tested if adequate money supply data were available. Godfrey presented quarterly data but some of his estimates were obtained by interpolation.Google Scholar

²⁴ See Roll, “Interest Rates”; and Calomiris, “Price and Exchange Rates.”Google Scholar

²⁵ Thian, compiler, Correspondence of the Treasury, p. 370.Google Scholar

²⁶ Ibid., pp. 432–33.

²⁷ In Dec. 1863 Treasury Secretary Christopher G. Memminger proposed sweeping monetary reform measures in order to curb inflation. Various currency reform measures were proposed and debated in the Confederate Congress from Dec. 1863 until Feb. 1864. Some of these would have wiped out most of the existing currency by forced conversion into bonds. See Yearns, The Confederate Congress, pp. 203–8Google Scholar; and Thian, Raphael P., compiler, Extracts from the Journals the Provisional Congress and of the First and Second Congresses of the Confederate States of America (Washington, DC, 1890 [on microfilm at the National Archives]), p. 526–678.Google Scholar

²⁸ In April and May $5 bills circulated at a 15 to 20 percent discount in anticipation of a tax change in July. See Pecquet, “Money in the Trans-Mississippi,” p. 228.Google Scholar

²⁹ Thian, compiler, Correspondence of the Treasury, pp. 588–92; and Todd, Confederate Finance, pp. 74–75. The similarities between northern and southern financial policies are interesting. A possible explanation for the similarities is that both shared the common goal of interest-rate controls and lacked a central bank to achieve their goal. This left the North and South with similar policy options.Google Scholar

³⁰ An additional impact of reform was felt in securities markets. The market for most Confederate securities became very thin between Apr. and Aug. 1864. The few bond price quotations during the summer months came from Wilmington, NC, rather than Richmond. The apparent reason for this “thinness” and the somewhat higher rates on 8 percent bonds is that people had been induced to buy so many 4 percent bonds that the total demand for bonds had virtually disappeared. Confederate treasury records indicate that many 8 percent bonds were shipped overseas during the summer of 1864. Perhaps this explains the presence of bond price quotations in Wilmington (a port city) during this period. See Thian, compiler, Correspondence of the Treasury, pp. 729–31.Google Scholar

³¹ According to Godfrey, Monetary Expansion of the Confederacy, who included all treasury notes and call certificates in his definition of the money supply, these aggregates fell from $926 million to $632 million between Feb. 17 and Apr. 1. He also estimated net bank money, which similarly fell from $91 to $17 million Confederate dollars during the same period. If we are correct that bank notes were worth approximately three Confederate dollarsGoogle Scholar (see Pecquet, “Money in the Trans-Mississippi,” p. 222)Google Scholar, then the decline in nominal money balances can be estimated by The decline in real money balances can be estimated by adjusting for the decline in the price of gold, which fell from $25 to $22 during the same period. See Todd, Confederate Finance, p. 198Google Scholar. This becomes

³² An anonymous referee suggested that an additional decline in money demand may have been caused by a flight from an uncertain currency.Google Scholar

³³ See Roll, “Interest Rates,” p. 480.Google Scholar

³⁴ This structure holds after correction for the preferential tax treatment given to state and local bonds.Google Scholar

³⁵ See the Appendix for the data sources.Google Scholar

³⁶ See Friedman, Milton, “Prices, Income, and Monetary Changes in Three Wartime Periods,” American Economic Review, 42 (05 1952), pp. 612–25Google Scholar, reprinted in The Optimum Quantity of Money and Other Essays (Chicago, 1969), pp. 157–97.Google Scholar