We use cookies to distinguish you from other users and to provide you with a better experience on our websites. Close this message to accept cookies or find out how to manage your cookie settings.
Published online by Cambridge University Press: 11 May 2010
To the public of the 1860's, federal loans and land grants to the pioneer Pacific railroads represented aid necessary to secure an economically and politically desirable technological feat; to the public of the 1870's they represented plunder. Scandal, hinted before the driving of the legendary Golden Spike in 1869, sporadically dominated the national political debates of the 1870's and the 1880's. The issue subsided by the 1900's, leaving only a residual popular impression of ample government aid to railroads. Among professional historians, the debate was revived in the mid-1940's by Robert S. Henry's revisionist article justifying the land grants. The ensuing discussion largely restored the original consensus that the land grants represented excessive subsidization.
1 Morgan, Charles S., “Problems in the Appraisal of the Railroad Land Grants,” reprinted in Carstensen, V., ed., The Public Lands (Madison, 1968), pp. 163–164.Google Scholar
2 Robert S. Henry, “The Railroad Land Grant Legend in American History Texts,” reprinted in V. Carstensen, ed., The Public Lands, pp. 121–144.
3 The comments of the main protagonists are reprinted in V. Carstensen, ed., The Public Lands, pp. 121–179.
4 Fogel, Robert W., The Union Pacific Railroad (Baltimore, 1969).Google Scholar
5 Mercer, L. J., “Rates of Return for Land Grant Railroads: The Central Pacific System,” The Jouhnal of Economic History XXX (September 1970), pp. 606–626Google Scholar. Mercer, L. J., “Land Grants to American Railroads: Social Cost or Social Benefit?,” Business History Review, XLIII (Summer 1969), 134–151.CrossRefGoogle Scholar
6 “A New Fight Erupts Over Land Grants,” Business Week, September 16, 1972, p. 35; see also the statement by Harris, Senator Fred, Congressional Record, CXVIII, No. 146, September 19, 1972.Google Scholar
7 Treatments of Robert Fogel's justification of the Union Pacific's subsidy appear elsewhere. See H. Fleisig, “Fogel's Defense of the Credit Mobilier,” forthcoming, Journal of Political Economy; Fleisig, H., “The Union Pacific Railroad and the Railroad Land Grant Controversy,” Explorations in Economic History, XI (Fall 1973), 155–172Google Scholar; Fleisig, H., “The Financial Condition of the Union Pacific Railroad, 1864–1867,” mimeo, Cornell University, Department of Economics, December 1972.Google Scholar
8 Most recently, Lloyd Mercer found that the private rate of return on the Central Pacific “system” was 12.9 percent without the land grant and 14.1 percent with the land grant. See L. J. Mercer, “Land Grants …,” p. 142. Mercer calculated the total private rate of return on private capital in the Central Pacific system, including the original Central Pacific line from Sacramento to Promontory Point, the Southern Pacific, and the Western Pacific. This paper treats the rate of return on entrepreneurial capital on the original Central Pacific line, giving a better indication of the inducement to the original promoters to build the pioneer line. If the land grants were unnecessary for these lines, there is some presumption that they were unnecessary for later lines whose technical and economic feasibility were more certain.
9 The point has been made before in this debate. See, for example, the interrogation of Collis P. Huntington by Senator John Morgan in “Government Debt of the Pacific Railroads. Notes of Hearings before the Committee on Pacific Railroads of the Senate of the United States on the Subject of the Indebtedness of the Pacific Railroads to the Government,” U.S. Senate, February 1, 1896, pp. 227–282. See also Daggett, Stuart, Chapters on the History of the Southern Pacific (N.Y., 1922), pp. 80–81Google Scholar; Goodrich, Carter, Government Promotion of American Canals and Railroads, 1800–1890 (N.Y., 1960), p. 187Google Scholar; and Engerman, Stanley L., “Some Economic Issues Relating to Railroad Subsidies and the Evaluation of Land Grants,” The Journal of Economic History, XXXII (June, 1972), 456–457.Google Scholar
10 One simple formulation of the rate of return on entrepreneurial capital is
where (Re) is the rate of return on entrepreneurial capital, (K) is the total investment, (Rk) is the overall private rate of return on capital, (L) is borrowed capital, (i) is the cost of borrowed funds, and (K – L). is entrepreneurial capital. If
both
and
are positive; moreover,
does not exist.
11 For the estimated cost of construction and its conversion to constant value paper dollars, see Table A-1, line 2C, and accompanying notes. Estimated bond safes of $43.1 million were calculated from the bond issue information shown in Table A-1, line 6.
Interest payments on the permanent debt were taken from Poor, H. V., Manual of the Railroads of the United States (New York, 1873), p. 681Google Scholar. Bond sales were estimated by capitalizing the earliest interest payments at 7 percent (the coupon rate on the Convertible bonds) until the total interest payments were large enough to account for the sale of all Convertible bonds; the remaining interest payments were capitalized at 6 percent, the coupon borne by the First Mortgage bonds. If the assumption that the Convertible bonds were sold first is untrue, then the promoters had slightly more non-equity capital during the early years than we have estimated.
The estimate also assumes that First and Second (government) Mortgage bonds were sold at par. Price quotations on Central Pacific First Mortgage bonds are not available prior to 1868. In 1868 and 1869 bond prices ranged between 97 and 103 for three observed dates. See Commercial and Financial Chronicle, September 19, 1868; October 24, 1868; July 24, 1869.
Sales at or above par also reflect the earlier experience of the Central Pacific. In a memo submitted to the Pacific Railway Commission, Stanford listed the “discounts” on each class of bond received by the Central Pacific; U.S. Pacific Railway Commission, “Report … [and] … Testimony,” Senate Executive Document No. 51, 50th Congress, 1st Session, DI: 2505–2509; DI: 2507, p. 2731. Hereafter referred to as “P.R.C., Hearings.” The table does not indicate that they are discounts of the gold price under the currency price, but the text elsewhere makes that amply clear (DI: 2507, pp. 2527, 2528). Though this mistitled table has misled some writers into thinking that the bonds sold at substantial discounts under par in currency (for example, S. Daggett, Southern Pacific, p. 25), one later investigation partly clarified the matter, noting that the government bonds (Second Mortgage bonds) sold at slightly over par (U.S. Federal Coordinator of Transportation, Public Aids to Transportation, Vol. II, p. 138; Washington: Government Printing Office, 1938). Stanford concedes the currency premium of the Second Mortgage bonds and also states that the First Mortgage bonds were sold at “like discounts” in gold, implying that they also sold at prices slightly over par (P.R.C., Hearings, DI: 2507, pp. 2465, 2428). Stanford's testimony was aimed at demonstrating the physical and financial difficulties of construction, he had every reason to mention discounts below par in terms of currency had they existed, and, indeed, does so for other classes of bonds. Among these, the city and county aid bonds are excluded from our estimate, imparting a slight bias against our conclusions.
Using Stanford's currency price of gold of approximately $1.40 for 1865–1866, his assertion that sales of $1,483,000 in Convertible bonds yielded $830,000 in gold, implies a currency bond price of approximately 80 (DI: 2507, pp. 2528, 2731). Since originally these bonds were used directly to purchase materials, however, and not marketed for cash, the discount is not strictly a market discount (Lavender, David, The Great Persuader [Garden City, N.Y., 1970], p. 130Google Scholar). The estimate in the text assumes a market price of 80.
The same procedure applied to Stanford's estimate of the amount realized from the State Aid bonds yields a currency market price of about 92. However, the listed currency prices of these bonds ranged from 153–165 in 1864 to 114–158 between January and August 1865; Bankers' Magazine (October 1865) p. 287. The estimate in the text assumes a market price of 92.
12 Each of the four promoters subscribed to 150 shares on June 28, 1861. A 10 percent installment was required on the par value of $15,000, or $1500 each. See S. Daggett, Southern Pacific …, p. 19; Carman, Harry J. and Mueller, Charles H., “The Contract and Finance Company and the Central Pacific Railroad,” The Mississippi Valley Historical Review, XIV (December 1927), 328–329Google Scholar; D. Lavender, The Great Persuader, pp. 94–95. It is unclear when the stock was completely paid for. Lavender states that the eighth 10 percent installment was called for on November 13, 1863; D. Lavender, The Great Persuader, p. 141. The stock records of the Central Pacific show that Charles Crocker was still making payments on his 150 shares in 1864. Leland Stanford did not make his final payment on the initial 150 shares until December 31, 1865; P.R.C., Hearings, DI: 2507, pp. 2554, 3502–3503. The calculation of the internal rate of return made below assumes that the initial stock purchases were completed by mid-1863, biasing the calculated internal rate of return downward in two ways. First, it pushes investment into the early years. Second, it raises the possibility that later payments on stock were made out of railroad or construction profits; such reinvestment of profits is implicitly assumed in the calculation of the internal rate of return and is not part of the initial investment on which the rate of return is being calculated. There is no evidence on the other promoters; however, they generally maintained strict parity among themselves. Each ultimately purchased 650 shares of Central Pacific stock for cash, each was an equal participant in the divided profits of the Crocker Construction Company and the Contract and Finance Company, and each received identical blocks of Central Pacific stock in exchange for their holdings of affiliated railroads (Western Pacific, California and Oregon, San Joaquin Valley, and the San Francisco, Oakland and Alameda), which they controlled through the Contract and Finance Company. P.R.C., Hearings, DI: 2505, pp. 71–72; DI: 2507, pp. 2550–2558.
13 Of 6303 shares of stock sold for cash (at a par value of $100) by 1869, 2156 had been purchased for cash by the promoters. The increase from 705 shares to 2156 took place in 1865, when the Central Pacific began collecting federal subsidy bonds. The promoters owned 390,123 shares of stock in 1869 while non-promoters owned 11,528 shares. Of the promoters' holdings, 384,710 shares were charged to construction account by the Central Pacific. In 1870, the promoters received another 99,775 shares of Central Pacific stock in exchange for their holdings in other California railroads that they were constructing at the same time as the Central Pacific; non-promoter holdings remained at 11,528 shares. These data were obtained by tabulating the list of Central Pacific stockholder transactions presented by the Pacific Railway Commission, Hearings, DI: 2507, pp. 2550–2558.
14 The Central Pacific had available to it funds in considerable excess of those needed for construction from 1862. Estimated cumulative construction cost is shown in Appendix Table A-l, line 7; estimated bond resources of the railroad are shown in Table A-l, line 6. This table excludes all other local aid, such as the bonds of San Francisco County and the stock subscriptions of Sacramento and Placer Counties, totalling $950,000. In mid-1865, eastern journalist Samuel Bowles reported that the Central Pacific still had no need to sell its land, U.S. bonds, First Mortgage bonds, or State Aid bonds; Bowles, Samuel, Across the Continent (Ann Arbor: University Microfilm, Mich., 1966Google Scholar; originally published 1865), pp. 270–271.
15 For a reconciliation between this figure and that shown in Appendix Table A-1, line 2C, see Appendix Table A-1, notes a and c.
16 U.S. Federal Coordinator ***f Transportation, Public Aids to Transportation, Vol. 2, p. 18.
17 See the annual balance sheets of the Central Pacific, P.R.C., Hearings, DI: 2509, pp. 4526–4558.
18 According to the Central Pacific stockholder transaction list, Huntington ceased acquiring stock for public sale in 1868 and other public cash purchases ceased in 1869; P.R.C., Hearings, DI: 2507, pp. 2550–2558. This decline in non-promoter sales, at a time when the railroad was clearly going to be completed and was earning substantial profits even before completion, is curious and can probably be attributed to fears of watering or attempts by the promoters to maintain control.
The proposed method of valuing the stock is discussed in Appendix Table A-1, note f. For those skeptical of this method, the rate of return assuming the stock is valued at zero is also presented.
19 We assume the investment of $60,000 takes place in three installments ($6,000 in mid-1861, $27,000 in mid-1862, and $27,000 in mid-1863) and deflate in the manner discussed in Appendix Table A-1, note a. Assuming the stock is worth nothing, the annual rate of return derived from Appendix Table 1, line 5 is 401 percent. Including the stock at its estimated market value, the rate of return is 493 percent. To check the sensitivity of the results to the estimated time pattern of investment, we assume the entire investment takes place in mid-1861 and the entire profit is not realized until mid-1869. If stocks are valued at zero, the rate of return is 90 percent; if stocks are set at their estimated market value, the rate of return is 113 percent.
All calculations assume that the original investment is lost at the end of the investment period; however, the rates of return are so high that the converse assumption produces no significant change in results (to two decimal places). Since each series contains one sign reversal, the rates of return are unique (Descartes' Rule of Signs).
20 Oakes Ames and Sidney Dillon, key promoters of the Union Pacific, regarded rates of return of 15 percent to 20 percent sufficient to make railroad projects attractive; U.S. House of Representatives, “Affairs of the Union Pacific Railroad Company,” House Report No. 78, 42nd Congress, 3d Session, DI: 1577, pp. 28, 256. McCague claims that Comstock mining shares regularly bore annual returns of 24 percent; J. McCague, Moguls and Iron Men; (New York, 1964), p. 45; while Daggett cites 2 percent per month as a common short term interest rate in California; S. Daggett, Southern Pacific, pp. 23–24. A rebuttal of this argument of the form that entrepreneurs might commonly have faced projects yielding rates of return of 90 percent to 493 percent annually over periods as long as eight years, on investments as large as $60,000, is impossible to reconcile with the realized annual average rate of growth of the economy as a whole.
21 Act of July 1, 1862, 12 Stat. U.S., 489, section 5; Amendment of July 2, 1864, 13 Stat U.S., 356, sections 7, 8, and 10.
22 How changes in the difference between the alternative cost of funds and the internal rate of return on a given project affect the rate of investment is an unsettled issue. The argument in the text implicitly assumes that a project yielding 90 percent to 493 percent annually over an eight year period would already be calling forth a maximum entrepreneurial response.
23 Robert Fogel, Union Pacific, pp. 81–84.
24 H. Fleisig, “Fogel's Defense.”
25 Central Pacific bond prices were quoted infrequently in the Commercial and Financial Chronicle during this period. For each date on which a Central Pacific quotation was available, a yield curve (yield as a function of years-to-maturity) was fitted to the yields and maturities for the seventy to eighty other railroad bonds listed for that date. The linear curves so fitted had (adjusted) values of R2ranging between .007 to .02 and in no case was the regression as a whole significant at the 5 percent level. The coefficient of the independent variable, number of years to maturity, was never significant at the 10 percent level. The absence of autocorrelation was taken to mean that experimentation with other theoretically plausible yield curves would be equally fruitless; in the absence of any theoretical justification for higher order polynomials which would have produced a better fit, no further testing was undertaken.
In the absence of a yield curve, another appropriate measure of the market's measure of the risk in Central Pacific bonds is the distance of the Central Pacific yield from the mean yield of all railroad bonds, regardless of maturity, in standard deviation units. The mean yield to maturity of all railroad bonds on September 19, 1868 was 7.11 percent (s = .128); the yield to maturity on Central Pacific First Mortgage bonds, priced at 103 bid, was 5.8 percent; the yield to maturity on Central Pacific Convertible bonds, priced at 110 bid, was 5.10 percent. The mean yield of all railroad bonds on October 24, 1868 was 7.2 percent (s = .99) while the yield on Central Pacific First Mortgage bonds, priced at 103 asked, was 5.8 percent. The mean yield of all railroad bonds on July 24, 1869 was 7.4 percent (s = .88) while the yield on Central Pacific First Mortgage bonds, priced at 97 asked, was 6.2 percent. In each case, Central Pacific yields are more than one standard deviation below the mean yield for all railroad bonds. On the same set of dates, yields on New York Central Premium Sinking Fund Bonds ranged from 6.55 percent to 6.60 percent, Pennsylvania Railroad Second Mortgage Bonds ranged from 6.35 percent to 6.60 percent, and Baltimore and Ohio Mortgage Sinking Fund Bonds ranged from 6.00 percent to 6.40 percent. Illinois Central Construction Bonds, however, ranged from 3.75 percent to 5.55 percent. In each case, where more than one bond was quoted, the one with the lowest yield was chosen. All bond yields were calculated from price, coupon, and maturity data in the Commercial and Financial Chronicle.
26 H. Fleisig, “Fogel's Defense,” pp. 14–15.
27 The Act of 1862 (12 Stat. U.S., 489) provided for a government issue of $16,000 per mile on the easier construction, to be delivered after 40 miles of track were complete (section 5). On this section the government was to retain 25 percent of the bonds until the completion of the entire railroad (section 17). For a distance 150 miles east of the “western base” of the Sierra Nevadas, the loan was $48,000 per mile (section 11), with 15 percent retained until completion of the entire road (section 17). It was commonly agreed that the “western base” of the Sierra Nevadas was approximately 31 miles east of Sacramento. However, the “western base” was ultimately fixed by Presidential executive order at a point 7 miles east of Sacramento, and the bond retention provisions were repealed in the Amendment of 1864. Edwin L. Sabin, Building the Pacific Railway (Philadelphia, 1919), p. 67; 13 U.S. Stat. 356, section 7. Thus, the Central Pacific finally received $1,696,000 in bonds for the first 40 miles out of Sacramento.
28 The promoters formally accepted the terms of the subsidy act on November 1, 1862; the bonds were issued in December; D. Lavender, The Great Persuader, p. 121; U.S. Department of the Interior, Office of the Commissioner of Railroads, Report of the Commissioner of Railroads, 1883 (Washington: Government Printing Office, 1883), pp. 680–681.Google Scholar
29 P.R.C., Hearings, DI: 2505, p. 83; DI: 2508, p. 3797.
30 Lavender reports the prospect of discounts which apparently surprised Huntington. He has no evidence, however, on the size of the discounts (D. Lavender, The Great Persuader, pp. 128–130). See note 11 for further discussion.
31 If the promoters expected prices to remain unchanged, and to realize $75,000 in 1862 and $75,000 in 1863, they would have expected to earn 683 percent per year. If they had perfect price expectations and the case is otherwise unchanged, they expected a rate of return of 790 percent.
If we shift the expected earnings to 1863 and 1864, with prices expected to remain unchanged, the expected rate of return is 111 percent. In the same case, assuming perfect price expectations, the rate of return is 71 percent. Internal rates of return are again derived from series with one sign reversal and are unique.
32 In January 1865 the promoters projected earnings of $3.8 million (gold) on the first 31 miles of road. The estimate was based on counts of wagon freight; American Railroad Journal, January 28, 1865, pp. 83–84. The substantial wagon freight from the Comstock Lode in Nevada was well-established and the earliest Central Pacific plans called for prior construction of a wagon toll road which would feed freight into the end of the line; D. Lavender, The Great Persuader, pp. 87, 94.
33 The amendment of 1864 (13 Stat, 356) dropped the provision that the government reserve fractions of the bonds (section 7), declared that coal and iron lands were not mineral lands and that they were included in the land grant (section 4), doubled the land grant (section 4), provided for the issue of 2 / 3 of the bonds due in difficult sections prior to completion of the track (section 8), subordinated the government bonds to second mortgage status, and permitted the company to issue bonds of the same type and quantity secured by a first mortgage as well as to issue these on the completion of each 20 mile section (section 10). The amendment of March 3, 1865 (13 Stat. U.S., 504) permitted the companies to issue First Mortgage bonds 100 miles in advance of construction. The amendment of July 23, 1866 (14 Stat. U.S., 79) allowed the Central Pacific to build east until it met the Union Pacific (section 2). There were also helpful executive “clarifications”; for example, see note 27.
34 In 1863 the state legislature passed a bill providing for a grant of $1.5 million to the Central Pacific while Stanford was governor. Placer County subscribed to $250,000 in stock, Sacramento County to $300,000, and San Francisco to $600,000 (S. Daggett, Southern Pacific, pp. 30–33; D. Lavender, The Great Persuader, pp. 130–132).
35 D. Lavender, The Great Persuader, p. 128.
36 Ibid., pp. 140, 149, 158–159.
37 Ibid., p. 155.
38 Ibid., p. 148; Samuel Bowles, Across the Continent, pp. 270–271.
39 Under various assumptions, Mercer found the annual social rate of return from the Central Pacific system ranging between 12.7 percent and 35.2 percent; his choice among these estimates is 24.1 percent; L. Mercer, “Rates of Return,” pp. 624, 626.
40 This is also Engerman's reading of the debate; S. L. Engerman. “Railroad Subsidies,” p. 459.
41 Federal Coordinator of Transportation, Public Aids to Transportation, p. 111. The figure of $39.2 million represents cumulative land grant earnings net of land sale expenses to 1927; gross land grant earnings were $49.2 million.
42 The point was raised in an interesting exchange among Stanley Engerman, Peter McClelland, and Peter Passell at the Columbia University Seminar in Economic History, October 18, 1971.
