¹ See, for example, Huntington, Samuel P., Political Order in Changing Societies (New Haven: Yale University Press, 1968), pp. 32–92Google Scholar, and Ted Robert Gurr, , Why Men Rebel (Princeton: Princeton University Press, 1970)Google Scholar.

² Increased support for extremist parties is often acknowledged as a sign of discontent and potential erosion of legitimacy. Other than the Democratic Alliance (a coalition of peasant groups in central Luzon, urban middle-class elements and some Communists, whose representatives were barred from holding office after the 1946 election), there have been no extremist parties participating in Philippine elections.

³ There is a rapidly growing body of literature dealing with patron-client relationships. The literature most applicable to our analysis includes: Landé, Carl, “Networks and Groups in Southeast Asia,” American Political Science Review, 67 (March, 1973), 103–127CrossRefGoogle Scholar; Scott, James C., “Patron-Client Politics and Political Change in Southeast Asia,” American Political Science Review, 66 (March, 1972), 91–113CrossRefGoogle Scholar; and by the same author, “Corruption, Machine Politics, and Political Change,” American Political Science Review, 63 (December, 1969), 1142–1158CrossRefGoogle Scholar.

⁴ Benda, Harry, “Political Elites in Colonial Southeast Asia: An Historical analysis,” Comparative Studies in Society and History, 7 (April, 1965), 233–251CrossRefGoogle Scholar. See also Reed, Robert, “Hispanic Urbanization in the Philippines,” Journal of East Asiatic Studies (March, 1967), 1–213Google Scholar.

⁵ See Hollnsteiner, Mary, “Reciprocity in the Lowland Philippines,” in Four Readings on Philippine Values, ed. Lynch, Frank and De Guzman, Alfonso (Quezon City: Ateneo de Manila Press, 1970), p. 85Google Scholar.

⁶ See Landé, Carl H., Leaders, Factions, and Parties: The Structure of Philippine Politics, Yale University Southeast Asia Studies Monograph Series No. 6 (New Haven: Yale University, Southeast Asia Studies, 1966)Google Scholar.

⁷ See Lemarchand, Rene, “Political Clientelism and Ethnicity in Tropical Africa: Competing Solidarities in Nation Building,” American Political Science Review, 66 (March, 1972), 86CrossRefGoogle Scholar.

⁸ Moore, Barrington Jr.,, Social Origins of Dictatorship and Democracy (Boston: Beacon, 1966), chapter 3Google Scholar.

⁹ John Carroll's data on the socioeconomic origins of Filipino entrepreneurs leads him to conclude that “the old elite did not precisely “withdraw …’ it remains, in proportion to its size, the most fruitful single source of manufacturing entrepreneurs.” See Carroll, John J., The Filipino Manufacturing Enter-preneur (Ithaca: Cornell University Press, 1965), p. 100Google Scholar. Stifle, Laurence Davis, The Textile Industry — A Case Study of Industrial Development in the Philippines, Data Paper No. 49 (Ithaca: Cornell Southeast Asia Program, 1963), p. 93Google Scholar, notes that for textile entrepreneurs “… entrepreneurship has been a means of preserving or consolidating a realized upper stratum position.” Simbulan, Dante C., A Study of the Socio-Economic Elite in Philippine Politics and Government 1946–1963, (Ph.D. dissertation, Australia National University, 1965)Google Scholar notes the fused nature of the Philippine social, political and economic elite.

¹⁰ See Simbulan, pp. 107–110.

¹¹ See Golay, Frank, “The Philippines,” in Under-development and Economic Nationalism in Southeast Asia, ed. Golay, Frank H., Anspach, Ralph, Pfanner, M. Ruth, and Ayal, Eliezer B. (Ithaca: Cornell University Press, 1969), p. 103Google Scholar.

¹² The concept “backward linkages” is explicated in Hirschman, Albert O., The Strategy of Economic Development (New Haven: Yale University Press, 1958)Google Scholar.

¹³ Hicks, George L. and McNicoll, Geoffrey, Trade and Growth in the Philippines (Ithaca: Cornell University Press, 1971), p. 105Google Scholar.

¹⁴ The real wage of wage earners in selected non-agricultural industries in the Philippines dropped 1.1 per cent from 1957 to 1969. Calculated from Central Bank of the Philippines, The Statistical Bulletin, 21 (December, 1969Google Scholar), Tables 24 and 127. Between 1956 and 1971 the income concentration ratio in the Philip-pines increased from .452 to .455. Calculated from Bureau of Census and Statistics, Survey of Households, Family Income and Expenditures: 1957 and Bureau of Census and Statistics, Survey of Households, Family Income and Expenditures: 1971. The concentration ratio is obtained by dividing the area bounded by the line of equal distribution and the Lorenz curve of income distribution by the total area under the line of equal distribution. The ratio therefore varies from 0 to 1, and the higher the value the greater will be the inequality in the distribution of in-come.

¹⁵ See United Nations, Economic Commission for Asia and The Far East, Economic Survey of Asia and the Far East 1961 (Bangkok: United Nations, 1962), p. 77Google Scholar and Economic Survey of Asia and The Far East 1966, p. 24.

¹⁶ Estanislao, Jesus P., Economics of a Nation (Manila: Center for Research and Communication, 1970), p. 122 and 151Google Scholar.

¹⁷ Calculated from the Philippine Department of Public Works and Communications, Annual Report, 1969–1970, p. 38. Included in the early years (1954–1958) are Public Works Acts Nos. 1200, 1411, 1613 and 1900. Included in the later years (1959–1965) are Public Works Act Nos. 2093, 2301, 2701 and 3101. To include more recent acts would artificially accentuate these differences since releases for such acts are still occurring.

¹⁸ The rate of population growth between 1960 and 1970, and between 1948 and 1960, is computed from the Republic of the Philippines Bureau of Census and Statistics, Census of the Philippines, Population and Housing for 1948, 1960 and 1970. The rate of growth in total government expenditures per capita between 1961 and 1969 is computed from statements on expenditures at the General Auditing Office of the Philippines.

¹⁹ Averch, Harvey A., Denton, Frank, and Koehler, John, The Matrix of Policy in the Philippines (Princeton: Princeton University Press, 1971), p. 101CrossRefGoogle Scholar.

²⁰ Ibid., p. 97.

²¹ See Gideon Sjoberg, “Cities in Developing and in Industrial Societies: A Cross Cultural Analysis” and Keyfitz, Nathan, “Political-Economic Aspects of Urbanization in South and Southeast Asia,” in The Study of Urbanization, ed. Hauser, Philip Moris and Schnore, Leo F. (New York: Wiley, 1965)Google Scholar. See also Kuznets, Simon, Economic Growth of Nations (Cambridge: Harvard University Press, 1971), pp. 80 ff.CrossRefGoogle Scholar

²² Sjoberg. See also McGee, T. G., The Southeast Asian City (New York: Praeger, 1967)Google Scholar.

²³ Absente landlords in the Philippines invariably bank in major urban areas and often pay the majority of their taxes there. Sugar centrals and other large corporations in the more rural cities have offices in the dominant cities and often pay their corporate income taxes through these main offices. Since chartered cities which collect greater amounts of national revenues than they did in the past are apportioned a certain amount of the excess, the location in which corporations and individuals pay their taxes is important.

²⁴ When the Philippines did not receive its anticipated share of the U.S. sugar quota in 1970, six new sugar centrals were being constructed and eight more had been built since 1965. Information compiled from Atlantic Gulf and Pacific Progress, Second Quarter Report, Vol. 11, No. 2 (Manila, 1970), p. 5Google Scholar.

²⁵ In her study of the municipality of Hulo, Holln-steiner notes that the increasing incorporation of Hulo into the national-metropolitan arena has led to a proliferation of and differentiation among the local elite. Whereas previously the traditional land- and fish pond-owning elite had been strong enough to carry completely the dominant political faction and its auxiliaries, at present “non-elite professionals were trying to gain mastery over the faction but though fairly rich, they did not have the seemingly inexhaustible funds which their elite predecessors had had.” Hollnsteiner, Mary R., The Dynamics of Power in a Philippine Municipality (Quezon City: Community Development Research Council, University of the Philippines, 1963), p. 51Google Scholar.

²⁶ From 1958 to 1961 the real paid-up mean capital investment per newly registered business was 214,900 pesos, and by 1961 to 1964 it had increased to 233,800 pesos. Calculated from Bureau of Census and Statistics, Yearbook of Philippine Statistics, 1966 (Manila, 1966), Table 18, p. 117Google Scholar. The wholesale price index used as deflater appears on page 99 of the same source.

²⁷ Makil, Perla, IPC/PAASCU Study of Schools and Influentials (Manila: Institute of Philippine Culture, 1971), Table A-3Google Scholar.

²⁸ For examples of this phenomenon in the Philippines, see Nowak, Thomas and Snyder, Kay, “Urbanization and Clientelist Systems in the Philippines,” Philip-pine Journal of Public Administration, 14 (July, 1970), 267Google Scholar.

²⁹ The mean per cent of individuals with an elementary education or less in Philippine chartered cities in 1960 was 71 per cent. Calculated from the Bureau of Census and Statistics, Census of the Philippines, 1960, Population and Housing. In Chicago first generation immigrants comprised 52 per cent of the population by 1850. See Bradley, Donald S. and Zald, Mayer N., “From Commercial Elite to Political Administrator: The Recruitment of the Mayors of Chicago,” The American Journal of Sociology, 71 (September, 1965), 153–167CrossRefGoogle Scholar.

³⁰ Between 1849 and 1868 municipal expenditures in Chicago increased from approximately $45,000 to more than $6,000,000. See Bradley and Zald, pp. 156–159.

³¹ See the discussion by Mangin, William, “Introduction,” in Peasants in Cities, ed. Mangin, William (Boston: Houghton Mifflin, 1970), pp. xxvii ff.Google Scholar

³² Hollnsteiner, Mary, “Reciprocity in the Lowland Philippines,” p. 85Google Scholar.

³³ Scott, James, “Patron-Client Politics and Political Change in Southeast Asia,” pp. 91–113Google Scholar, does an extensive analysis of patronclient structures in which he suggests a number of dimensions along which such structures can be contrasted.

³⁴ This is both cogently argued and demonstrated by Norris, Steven Dodge, Political Behavior and Social Change: Causes of the Growth of the Indian Electorate in the Last Half Century (Ph.D. dissertation, Cornell University, 1971)Google Scholar. What mobilization theory does alert us to is that in systems such as the Philippines which have literacy requirements for voting (albeit lax), social mobilization increases the proportion of adults registered to vote. Our data show that the level of social mobilization correlates +.46 with the level of voter registration. The concept of social mobilization is described in Deutsch, Karl W., “Social Mobilization and Political Development,” American Political Science Review, 55 (September, 1961), 498–499CrossRefGoogle Scholar.

³⁵ Lenski, for example, argues that in the simplest societies, or those which are technologically most primitive, the goods and services available will be largely distributed on the basis of need. See Lenski, Gerhard E., Power and Privilege (New York: McGraw-Hill, 1966)Google Scholar.

³⁶ Nie, Norman H., Powell, G. B. Jr.,, and Prewitt, Kenneth, “Social Structure and Political Participation: Developmental Relationships, I,” American Political Science Review, 63 (June, 1969), 366CrossRefGoogle Scholar.

³⁷ Chambers, William Nisbet, “Party Development and the American Mainstream,” in The American Party Systems, ed. Chambers, William Nisbet and Burnham, Walter Dean (New York: Oxford University Press, 1967), p. 14Google Scholar. See also Burnham, Walter Dean, Critical Elections and the Mainstream of American Politics (New York: W. W. Norton, 1970), pp. 21 and 71–90Google Scholar. In Mexico, the access of the middle and upper classes to resources and policy makers through nonelectoral channels helps explain lower turnout among these sectors. See Reyna, Jose Luis, An Empirical Analysis of Political Mobilization: The Case of Mexico, Latin American Studies Program Dissertation Series No. 26 (Ithaca: Cornell University Press, 1971)Google Scholar, chapters 5 and 6.

³⁸ Gordon, Daniel N., “Immigrants and Municipal Voting Turnout: Implications for the Changing Ethnic Impact on Urban Politics,” American Sociological Review, 35 (August, 1970), 665–681CrossRefGoogle Scholar.

³⁹ Powell, John Duncan, “Peasant Society and Clientelist Politics,” American Political Science Review, 64 (June, 1970), 420CrossRefGoogle Scholar.

⁴⁰ Our unit of analysis is the chartered city. Chartered cities are legal entities which may or may not be true cities, since they display considerable variance in density, industrialization and the level of social mobilization. Such cities either alone or in clusters, as in the Greater Manila complex, form labor market areas, which makes them a useful unit in analyzing relationships between socioeconomic and political structure.

⁴¹ The information on stockholders and their stockholdings was obtained from records of the Philippine Securities and Exchange Commission. Only 16 per cent of the corporations were controlled by other corporations rather than individual families (N = 1511).

⁴² Since the total corporate assets of a city accounted for under the 70 per cent rule varied from 77 per cent to 92 per cent, each family's control over corporate assets is further standardized by dividing it by the total assets accounted for by all the families enumerated under the 70 per cent rule for a city.

⁴³ See Alker, Hayward R. Jr.,, Mathematics and Politics (New York: Macmillan, 1965), pp. 30–42Google Scholar. A Lorenz Curve is a plot of values held by cumulative proportions of a population.

⁴⁴ Hofferbert, Richard I., “Socioeconomic Dimensions of the American States: 1890–1960,” Midwest Journal of Political Science, 12 (August, 1968), 401–418CrossRefGoogle Scholar, finds measures of urbanization and industrialization loading on the same factor which he calls “industrialization.”

⁴⁵ In order to derive factors uncorrelated with each other, we used orthogonal rotation. While oblique factor solutions are often a closer approximation to reality in producing intercorrelated factor solutions, use of factor scores in regression equations influenced our decision to derive (when necessary) uncorrelated ecological variables. Entering highly intercorrelated variables into a regression equation may produce unreliable results. For a discussion of the assumptions behind different forms of rotation see Nie, Norman, Bent, Dale H., and Hull, C. Hadlai, SPSS: Statistical Package for the Social Sciences (New York: McGraw-Hill, 1970), chapter 17Google Scholar.

⁴⁶ Hofferbert.

⁴⁷ Economic census figures (while probably under-stating actual Chinese control) indicate, however, that alien control is limited in the smallest firms. Among establishments employing 10 or fewer individuals, Filipinos owned 90.0 per cent of those in commerce, 92.1 per cent of those in manufacturing, and 92.2 per cent of those in services. Bureau of Census and Statistics, Economic Census of the Philippines: 1961 Vols. 3, 6, 8 (Manila, 1965)Google Scholar.

⁴⁸ Deutsch, p. 494.

⁴⁹ Ibid., pp. 501–503.

⁵⁰ Since oblique rotation does not necessarily produce factors uncorrelated with each other, its factors are often a closer approximation of reality than orthogonal solutions. In this case, however, the factors derived from oblique rotation were virtually identical to those derived from the orthogonal solution.

⁵¹ Landé, Carl, Leaders, Factions and Parties: The Structure of Philippine Politics, p. 4Google Scholar.

⁵² While some cities had elections for certain local offices before 1959, all mayoral and city councilor positions became elective for all cities with the passage of the Local Autonomy Act in 1959 (R.A. 2264). City residents participated in national elections, how-ever, even before Philippine independence. The earliest national election data available to us for all chartered cities was for the 1957 election.

⁵³ Mean vote for winning mayor is strongly correlated with the mean percentage of city councilors belonging to the same party as the mayor ( + .62) and with turnover among all elected city officials over the three terms of office for which election data are inc included (−.59). With an N of 46 both correlation coefficients are significant beyond the .001 level.

⁵⁴ The mean begins with the 1957 election and includes elections for 2 year intervals through the 1969 election. Turnout is the proportion of registrants voting.

⁵⁵ Mean voting turnout in these national elections fluctuates between 75 per cent and 77 per cent for the chartered cities as a group, and in local elections it fluctuates between 76 per cent and 79 per cent.

⁵⁶ See Bohrnstedt, George W., “Observations of The Measurement of Change,” in Sociological Methodology, ed. Borgatta, Edgar F. and Bohrnstedt, George W. (San Francisco: Jossey-Bass, 1969)Google Scholar. See also Norris, , Political Behavior, pp. 22–24Google Scholar. For our data the use of residuals produces a solution similar to that obtained by subtracting turnout in the early years from turnout in the late years. The intercorrelations between the two alternative measures of change in local and national turnout are all over +.98.

⁵⁷ “The Top Two Hundred Foreign Firms,” Manila Chronicle Magazine Supplement, July 7, 1971, p. 1Google Scholar.

⁵⁸ Computed from Bureau of Census and Statistics, Economic Census of the Philippines, 1961, Vol. III, Table 25.

⁵⁹ See “The Top Two Hundred Foreign Firms.”

⁶⁰ See Hicks, and McNicoll, , Trade and Growth in the Philippines, p. 72Google Scholar. While the commercial, processing and assembly nature of much American investment in the Philippines helps perpetuate dependence on capital imports, equally significant for the pattern of economic growth in the Philippines is the amount of capital outflow. For the six years following 1957, earnings accruing to American foreign investors were more than twice the amount of direct foreign investment in the Philippines. See Golay, Frank H., “Economic Collaboration: The Role of American Investment,” in The United States and the Philippines, ed. Golay, Frank H. (Englewood Cliffs: Prentice-Hall, 1966), p. 110Google Scholar. From 1965 to 1968, new American investment amounted to $30.2 million while withdrawals were more than $59 million. Santos, A. B., “The Philippines—Country with an Image Problem,” Manila Chronicle, January 6, 1971, p. 9Google Scholar. These figures overstate the amount of net capital inflow since many foreign firms raise considerable amounts of their capital requirements from Philippine banks. In 1968 the National Economic Council found that 108 American firms over the period from 1956 to 1965 raised 84 per cent of their capital requirements from Philippine Banks, and only 16 per cent from foreign sources. Cited in the Manila Chronicle, May 4, 1971, p. 9Google Scholar.

⁶¹ See McGee, The Southeast Asian City, chapter 3.

⁶² Ibid., p. 133.

⁶³ Ibid., p. 126.

⁶⁴ See for example Duñgo, Nanette Garcia, A Southern Industrial Complex (Quezon City: Community Development Research Council, University of the Philippines, 1969)Google Scholar.

⁶⁵ Machado, Kit G., “Changing Patterns of Leadership Recruitment and the Emergence of the Professional Politician in Philippine Local Politics,” Philippine Journal of Public Administration, 16 (April, 1972), 147–169Google Scholar.

⁶⁶ In his study of local leaders in Camarines Norte, Benson notes that “jobs, projects and money were shown to be the most important considerations for a lider to support a candidate for governor or congressman.” See Benson, Lewis, Political Leadership Through Political Liders: A New Approach for the Analysis of Philippine Provincial Leadership Positions (Ph.D. dissertation, University of Hawaii, 1970), pp. V–VIGoogle Scholar.

⁶⁷ High levels of machine strength are measured by electoral pluralities of 52.1 per cent or higher, and low levels of machine strength by a mean plurality of 52.0 per cent or lower.

⁶⁸ See the discussion by Nie, et al. , “Social Structure and Political Participation,” pp. 366–368Google Scholar.

⁶⁹ Basilan and Marawi are the only-two of our cities with Moslem majorities (73 per cent and 86 per cent respectively). In the former the cleavages between the hill dwelling, nominally Moslem Yakan peoples and the coastal Tausog peoples are probably as intensive as those between Christian settlers and Tausogs. Such cleavages, combined with the high illiteracy among both Tausogs and Yakans, have made possible the election of a Christian mayor.

⁷⁰ Anderson, James N., “Land and Society in a Pangasinan Community,” in Social Foundations of Community Development, ed. Espiritu, Socorro C. and Hunt, Chester L. (Manila: R. M. Garcia, 1964), p. 179Google Scholar.

⁷¹ See the discussion by Matza, David, “The Disreputable Poor,” in Class, Status and Power, 2nd ed., ed. Lipset, Seymour Martin and Bendix, Reinhard (New York: The Free Press, 1966), pp. 289–302Google Scholar.

⁷² For change in local turnout, substituting the percentage of the population that is either high-school or college-educated for social mobilization raised the absolute value of the Beta slightly to — .29. When only the percentage of the population college-educated alone is substituted, the absolute value of the Beta is raised even further to —.33.

⁷³ For change in national turnout, substituting the percentage of the population either high-school or college-educated for social mobilization decreased the absolute value of the Beta to —.21. When the percentage of the population college-educated alone is substituted, the absolute value of the Beta is decreased even further to —.18.

⁷⁴ Figures on economic concentration are from the Bureau of Census and Statistics, Survey of Households, Family Income and Expenditures: 1957 and Bureau of Census and Statistics, Survey of Households, Family Income and Expenditures: 1971.

⁷⁵ Clarence Abercrombie III (counting front page stories, letters to the editors and political cartoons dealing with the subject of revolution in a sample of Manila Times in 1951, 1960, and 1970) found a decline in the mean items per day dealing with revolution from 1951 to 1960 (from 2.62 to .31). Between 1960 and 1970 the mean rose to 7.03. Clarence Abercrombie III, “Vietnam and the Philippines: A Comparative Study of Revolution in Southeast Asia" (unpublished manuscript, Yale University, 1970).

⁷⁶ Sterba, James P., “Filipino Rebels Urge Ouster of Chiefs,” New York Times, March 24, 1973, p. 3Google Scholar.

⁷⁷ Under martial law declared on September 23, 1972, the military and executive sectors are given extraordinary power.