No CrossRef data available.
Article contents
On Methodology and the Use of Historical Statistics
Published online by Cambridge University Press: 24 October 2022
Abstract
Core share and HTML view are not available for this content. However, as you have access to this content, a full PDF is available via the ‘Save PDF’ action button.
In Raising Questions Concerning Methodology in My Book, The Mexican Revolution: Federal Expenditure and Social Change Since 1910, Professors Thomas E. Skidmore and Peter H. Smith have offered a reminder that statistics “do not speak for themselves.” The authors of “Notes on Quantitative History” are to be commended for undertaking a lengthy article examining some long-standing problems which historians face in attempting to understand the Latin American past.
- Type
- Forum
- Information
- Copyright
- Copyright © 1970 by the University of Texas Press
References
Notes
1. Skidmore and Smith (in notes 3 and 8) misleadingly compare my book directly to Wallich and Adler's economic study of El Salvador. If they did not have a fixed notion of how expenditure may be studied (there is, of course, no one method), Skidmore and Smith might have discussed, for example, Alan T. Peacock and Jack Wiseman, The Growth of Public Expenditure in the United Kingdom (Princeton: Princeton University Press, 1961). Though Peacock and Wiseman are concerned with economic rather than political policy, their study (like my work) concentrates on expenditure and generally excludes discussion of revenue and taxation matters.
2. In discussing expenditure and social change, Skidmore and Smith become enmeshed in a number of difficulties which may be exemplified by the following: First, they not only apparently are unaware that the political impact of central government may be every bit as important to national development as the relation of central government expenditure to GNP, but they fail to take into account the need to disaggregate GNP by sectors in order to make any meaningful statement.
Second, though they might have disagreed with my definition of “ideology” (pp. 40-41), for example, they reveal a careless reading of the book by claiming that the word is not defined at all, thus ignoring quotations of three authors who define the term (one of whom gives a dictionary definition).
Third, they do not realize that the functional accounting system developed in my Table 1—4 (which is very close to that used by several Latin American countries in recent years) is not simply an intermediary step between (a) traditional listing of expenditure by ministry and (b) analysis of capital expenditure, but (c) a necessary alternative way of gauging state policy because no single method will suffice for varying purposes of analysis. Fourth, in order to prove that my interpretation steps “far beyond” the bounds of data directly measuring social change in the Poverty Index, they quote a sentence from my conclusion to the book (p. 277) which summarizes data indirectly assessing psychological benefits gained from the right to strike (p. 183) and land reform (p. 195).
Fifth, in all of their discussion of my Poverty Index they neglect to examine its meaning (p. xxvii):
Persons included in the index may exhibit several characteristics of poverty and yet have a relatively high income. Nevertheless, collectively speaking, the integration of the Mexican nation is greatly impeded by the persistence of a high level in characteristics of poverty. Social modernization, along with economic development, is required in order to raise general standards of living. The Poverty Index seeks to measure decrease in the collective level of social deprivation in Mexico at different historical times.
Furthermore, Skidmore and Smith apparently are unaware of difficulties in attempting to use personal income analysis or taxation studies for groups which are on the fringe of monetary economy.
3. In their text and interpretation relating to note 4, Skidmore and Smith suggest that presentation of actual expenditures in my book matches precisely data given by Rosas Figueroa and Santillán López. Such a suggestion is misleading because Skidmore and Smith neglect to add that simple total federal expenditure (given as an appendix without interpretation by Rosas Figueroa and Santillán López) is useless for purposes of analyzing state policy developed in my book. Whereas I compare disaggregated projected amounts with actual figures by ministry and functional category as well as percentage terms and standard pesos per capita, the latter authors not only make none of these analyses but give aggregate totals for only a limited number of years in order to mention outlay from an economic rather than a political point of view.
4. James W. Wilkie, The Bolivian Revolution and U. S. Aid Since 1952: Financial Background and Context of Political Decisions (Los Angeles: Latin American Center, University of California, 1969).
5. Requirements that the central government be made aware of budgeted and actual expenditure by decentralized agencies in Mexico and Bolivia have meant little in practice because, as in Costa Rica, each agency conducts its own audit and reporting generally is non-standardized. In the Costa Rican case, various officers of the central government are in disagreement about how many agencies even exist; and in Mexico, José P. González Blanco, the chief investigator who compiled the authoritative but incomplete figures on projected public sector investment, has remarked to this writer that even with presidential authority he could not persuade many of Mexico's several hundred autonomous and semi-autonomous agencies to open their files on projected amounts, let alone actual expenditure.
6. In the case of a test which Skidmore and Smith offer, their interpretation correlating the regional rank order of poverty level to change in the level itself is misleading because it disingenuously restates my view as if it were their own discovery; see my summary statement (p. 235): “Historically, the regions in Table 9-10 have maintained the same relation to each other in terms of poverty level.” Furthermore, Skidmore and Smith ignore data in my Chapter 10 (especially Tables 10-1, 10-2, 10-3) which suggest that since 1926 Mexican federal expenditure has been devoted to regions with less rather than more poverty. Then, when neither expenditure nor the Poverty Index are mentioned in Table 10-10, Skidmore and Smith offer us the erroneous view that the table in question “makes an implicit effort to gauge the impact of one type of expenditure upon one component of the Poverty Index….”
You have
Access