Turning points in the U.S. Civil War: Views from the grayback market.

• Author: Weidenmier, Marc D.

1. Introduction

   Recent studies of Civil War financial markets have largely focused on the role of war news in asset price determination. The news literature on Civil War asset prices centers on price movements in an inconvertible currency issued by the U.S. government called the greenback. In his famous book, A History of the Greenbacks, Wesley Mitchell (1903) argued that the greenback price of gold fluctuated in response to news that changed expectations regarding a return to specie convertibility. War, fiscal, and political news moved the price of the greenback rather than changes in money supply and money demand.

   Willard, Guinnane, and Rosen, hereafter WGR (1996), identify turning points in the time path of Northern greenback prices using structural break analysis. (1) Their analysis indicates that some events seen as important to modern day historians of the Civil War, such as Gettysburg, were important turning points. They also found that some events given less attention in today's history books, like General Early's advance on Washington in July 1864, were also important turning points. WGR interpret their findings as evidence that the methodologies traditionally used by historians (i.e., letters, diaries, etc.) do not always lead to an accurate assessment of how contemporaries viewed the past.

   One weakness of the literature on Civil War financial markets has been the poor quality of data from Confederate money markets. The absence of a high-quality price series for the Confederacy has made it difficult to compare how investors in the North and South reacted to events of the Civil War. This article introduces a new time series of Confederate money prices, graybacks, taken from the newspapers of Richmond and leading cities in the Eastern Confederacy. An exhaustive search of more than a dozen primary newspapers yields a money price series of nearly 500 observations during the course of the Civil War. A new series of weekly grayback money price is tested for turning points. The empirical analysis suggests that turning points in the grayback market were different from those identified in the Union greenback market (WGR 1996). Investors in the North and South reacted differently to events of the Civil War.

   The article begins with a review of recent studies on asset price determination in the Southern Confederacy. The new data set is described and compared with existing series on grayback prices. This is followed by a turning point analysis of the new money series. Breaks in grayback prices are then compared with those in the greenback market. The article concludes with a discussion of the implications of the findings for future studies of the American Civil War and for war finance in general.

2. Previous Studies of Confederate Asset Prices

   Mitchell (1903) hypothesized that greenback prices during the Civil War fluctuated in response to news that changed expectations regarding a return to specie convertibility. In a similar fashion, Schwab (1901) examined movements in grayback prices. He noted that Confederate money prices also seemed to mirror Confederate fortunes on the battlefield. Schwab chronicled how grayback prices rose and fell with Confederate military triumphs and defeats. The view that grayback price movements largely reflected Southern perceptions of the war is supported by contemporaries of the period. E. A. Pollard, editor of the Richmond Examiner, characterized graybacks as the "sinews of war" (Pollard 1994, p. 415).

   Recent studies of grayback prices have examined the role of news in asset price determination using time series econometrics. Davis and Pecquet (1990) were unable to reject the hypothesis that Confederate money prices contain a unit root, providing empirical evidence that graybacks followed a random walk. The presence of a stochastic trend in grayback prices supports the view that Confederate money prices fluctuated in response to news that changed expectations regarding a return to specie convertibility. However, Davis and Pecquet's study was limited because grayback prices were not available for the first and last years of the war.

   Burdekin and Langdana (1993) looked at the relationship between war news and Lern's Price Index, a monthly measure of prices based on 20 commodities taken from newspapers of major cities in the Eastern Confederacy. They concluded that war news was a proxy for the future values of money and debt issues by the Confederate government. War news was a likely instrument for fiscal news and therefore an important determinant of the Confederate price level.

   McCandless (1996) employed a semimonthly series of grayback prices to examine the impact of war shocks on short-run changes in Union and Confederate money prices. His analysis suggests that war news was the single most important determinant of currency prices during the Civil War. Battles proxied for future money financing by the Northern and Southern governments. Military engagements are assumed to emit a common symmetric signal such that a Northern victory (defeat) causes an appreciation (depreciation) in greenback prices and an equal depreciation (appreciation) in grayback prices.

   More recent studies of Confederate asset prices during the American Civil War have focused on the impact of war news on Confederate cotton bond prices trading in London. Brown and Burdekin (2000) estimate turning points in cotton bond prices. Although their analysis is limited by a relatively small sample size (109 observations), their results indicate that Atlanta was a turning point not found in WGR's (1996) analysis of Union greenbacks. They argue that the more than doubling of Confederate cotton bond prices between December 1863 and September 1864 can be largely attributed to two factors: the slow pace of Union advances during the first nine months of 1864 and the possible election of McClellan as President of the United States on a peace party ticket. They conclude that war did not always have symmetric effects on Union and Confederate asset prices. Weidenmier (2000) analyzes movements in the cotton bonds with a vector autoregression. These results suggest that the large rise in Confederate cotton bond p rices during 1864 can be explained by a rise in the price of cotton. Cotton, not war news, was apparently the most important determinant of bond prices.

3. Graybacks and the Richmond Gold Market

   During the American Civil War, both the United States and Confederate governments printed large amounts of fiat money to finance the war. This required that both countries leave the gold standard and suspend specie convertibility. The North suspended gold payments in January 1862, while the future Confederate states adopted floating exchange rates in the fall of 1860, when the election of Lincoln led to a run on Southern banks. Suspension led to the formation of a grayback/gold exchange and a greenback/gold market where traders could speculate on the fiat money price of a gold dollar. Like the North, paper money issued by the Confederate government could not be used to pay customs duties. As shown in Table 1, the South received some revenue from tariff duties that created a demand for gold. These figures place a lower bound on gold demand given that there was widespread tax evasion and the Confederate government lacked the infrastructure to collect taxes (Ball 1991).

   More important, the drying up of foreign sterling and franc exchange in the early years of the war created a demand for gold in Richmond and port cities such as Charleston, South Carolina, Mobile, Alabama, and Galveston, Texas. An active gold market even arose in Wilmington, North Carolina, a port city connected to Richmond by direct telegraph and railroad links. Blockade runners and the Confederate government demanded hard currency to carry out business transactions since there was not a market for grayback money in Europe. As shown in Table 2, the number of blockade runs to and from the Confederacy averaged around 1000 per year for the period 1862-1864 (Lebergott 1981). With the capture of Wilmington, North Carolina, in early 1865, the demand for hard currency fell and Richmond newspapers stopped quoting grayback prices (see Appendix).

   An extensive search of more than a dozen newspapers yielded a series of 486 grayback prices from public auctions and private brokerage trades in Richmond between May 27, 1861, and February 15, 1865. The weekly...