Skip Navigation Planning & Markets
Subscribe Submission Requirements Editorial Board Archive Links Search Home

IV. The Dynamic Narrative Of The Local Politico-economic Process

The local planning and policy process devolved as shown in Figure 2 above when following the numbered steps. To illustrate some of the more important processes, detailed views of the complete model are shown.

Development Frustration as a result of steps 1 and 2 is defined as the discrepancy between actual and expected Development Status, and gains in strength with increasing organizational power of the villagers. Organizational Power is defined as the organization memberships per villager as determined through the field research. Rueland and Ladavalya (1993) show the importance of peoples' organizational intensity in Thailand in the local political power play, and that in a first step towards devolution of political and administrative powers, local organizations and community life have to be built and strengthened. Iddagoda, K.S., and Dale, R, (1997 p. 28) declare with respect to Sri Lanka, "Organization is, ..., closely linked to empowerment." According to the field research results, the organizational intensity of the villagers changes with changing household incomes. It first increases directly with household income, and then from a middle income decreases slightly to increase again from an above average income onwards. For each locality, organization intensity lookup functions were developed by means of a regression analysis. These changing values power the Development Frustration (steps 1 and 2).

Part of the regression output from SPSS and its translation into the SD model for Zorita is shown below as an example for the other three villages. See Table 5.

TABLE 5: Zorita SPSS Regression Output of Organization Frequency as Function of Household Income

Dependent variable ORGPART Method CUBIC
Multiple R .59453
R Square .35347
Adjusted R Square .25139
Standard Error .67064
F = 3.46253 Signif F = .0369

The OLS R2 result for questions 2, 5, and 77 of the survey ("Are you a member of a political party?," "Are you a member of any local organization, association or club?," "Please assign yourself to one of the following groups: how large is your total monthly household income approximately?") in Zorita is only 0.35347, and suggests a cubic relationship between organization membership and income. It is assumed that this relationship describes reality to some extent. The regression graph is translated into a lookup function in the system dynamic model. The horizontal axis shows income group, the vertical axis average organization membership per person.

The relationship suggests that villagers first become members of local labor unions, farmer and housewife organizations, and local political parties; and then with increasing income refrain from engaging themselves further. Beyond a medium income, they become members in regional organizations and political parties or service clubs.

Translating the cubic regression into the system dynamic lookup function, the following values are found. Membership increases first from its original 0.566 value, i.e., every person in the village is on average member in 0.566 organizations, to 0.9, then decreases to a value of 0.2 and increases again to a value of 1. The estimated values entered are as follows: household income $10'000 = 0.566 is set equal to 1, $20'000 = 0.9 = 1.6 X 0.566, $40'000 = 0.2 = 0.35 X 0.566, $60'000 = 1= 1.77 X 0.566. This lookup function drives the membership frequency variable in time as household income changes.

Depending on the local legal framework, villagers are either given no direct fiscal nor project power, meaning the locality is ruled through a complete representative democratic system; or they are given power to directly participate in project and/or fiscal policy decisions, i.e., giving the villagers total power thus activating taxing and project power simultaneously. This means that the locality is ruled by a complete direct democratic system (steps 3, 4, 5).

Figure 3 shows the education and culture project bargaining structure, which is one of the three project planning processes. The three project planning processes are: education and culture, welfare and, infrastructure. It shows how powers and needs, which are empirical parameters, drive the planning processes to eventually reach a compromise plan decision. However, villagers' needs and powers become effective in the planning process only if villagers are project empowered.

The group sums of the councilors' individual SNA power measures were added to represent the total bargaining power of the Social and Infrastructure Groups within the village council. A villagers' project pressure factor based on unfulfilled development expectations was added to this. The sum of these three factors represents the total village political power and the group's percentage in the total their relative bargaining strength within the village. These powers were then entered into the dynamic systems' planning module. The outcome is a compromise allocation in percent of the total regular Local Government (LG) income (step 6 in Figure 2).

Figure 3

FIGURE 3: Sukhapiban Huay Yai: The Education and Culture Project Bargaining Process

Note: In the following model details the names of the local Thai administrative units are used; Sukhapiban = Sanitary District, is the lowest political administrative unit with some legislative and taxing powers of its own, since completion of the research, Sukhapiban have been upgraded to" Tessaban Tambon" which is to a status of lowest Municipality.

Project plans are tallied against available finances in steps 7 through 10 of Figure 2. If planned expenditures exceed available finances, and if irregular financing is possible; then the councilors seek irregular finances from private sponsors or upper government sources. Irregular finances are mostly granted against promises to satisfy vested interests. If irregular finacing is constrained or not possible, the LG seeks credit financing if allowed by law. The total finances available to the LG are then divided according to agreed plan allocations.

Figure 4 shows the regular financing process of a locality where villagers want to push down taxes and levies and local government is forced to increase them due to decentralization of public services to the localities. Upper government is assumed to reduce contributions to the locality to a level of 25 percent of regular LG income.

The conditional equation of adjusted local tax, charges and levy rate is shown below as an example of how equations must be entered in Vensim PLE.

(1.) Adjusted local tax, charges and levy rate =IF THEN ELSE(villagers' public finance bargaining power in a decentralized system according to general Q48 ff+villagers' project bargaining power in a decentralized system according to general Q22=0,
decentralization adjusted longterm overall local tax and levy rate,
villagers' tax rate lookup(villagers' finance pressure)*decentralization adjusted longterm overall local tax and levy rate)

Equation 1 above shows that the variable "adjusted local tax, chargers and levy rate" is simply the "decentralization adjusted longterm overall local tax and levy rate" if public finance bargaining power and project bargaining power is set to 0, i.e., if villagers cannot decisively participate in the financing nor in the projects decisions. Villagers' influences become active only if they are empowered as revealed from the results of questions 22 and 48 of the field survey, and the respective powers are set to 1; then the tax rate lookup function becomes active.

Figure 4

FIGURE 4: Sukhapiban Huay Yai: The Regular Local Government Financing Process

Figure 5 shows the irregular financing process of Sukhapiban Huay Yai as an example of an irregular financing process. There are two parallel governmental setups in the Thai localities: the Central Governmental Setup with Village Headmen (Phuyai Baan), Subdistrict Head (Kamnan) and Distirict Authority (Amphur) and the local government setup, which was the case at the time of the research, the Sukhapiban (now Tessaban Tambon). The structure also shows the connections of these local entities to the upper governmental levels, the Province (Changwat) and the Provincial Parliament Representative (Sor Chor) and to the Central Government Ministries and National Parliament Representatives (Sor Sor). The link via outside "sponsors", here called "Chao Po" (in English: "Godfather"), is also shown.

On top of Figure 5 is the accountability pressure (Steps 13 through 16 in Figure 2). This shows the development of accountability pressure, which is thought to increase with increasing education status, defined as accumulated private and public education expenditures. The higher the education status in a locality, the more constraints are imposed on irregular financing and public finance leaks. Though Figure 5 shows the reality of Huay Yai only, the process is similar in the Philippines according to field information.

Figure 5

FIGURE 5: Irregular Financing Process Sukhapiban Huay Yai

Steps 11 through 14 of Figure 2 show the links between expenditures, growth and disposable household income. Growth is modeled according to the empirical cross-country growth model of Barro and Sala-I-Martin (1995) and Barro (1999), which is based on the neoclassical growth model a la Solow (1956), Swan (1965) and Ramsey (1928), enriched by government policies, human capital, and the diffusion of technology. Barro and Sala-I-Martin (1995, p. 421) define their function of a country's per capita growth rate in period t, Dy as

(2.) Dyt = F(yt-1, ht-1;...)

Where yt-1 is initial per capita GDP and ht-1 is initial human capital per person (based on measures of educational attainment and health). The omitted variables comprise a number of variables such as government policies, market distortions and others. Dyt is assumed to decrease with increasing yt-1 and ht-1. This is the standard observation that countries with higher GDP grow slower than countries with lower GDP or health status. This function specification represents the growth rate convergence hypothesis of Barro and Sala-I-Martin (1991).

For the present research, it is assumed that the Barro and Sala-I-Martin regression parameters can also be applied to the local level, and that it is admissible to superimpose the local growth rates on the national growth rate. This would explain why the development status in all countries differ form region to region or even from locality to locality. It is assumed that the local growth rate is enhanced by those public policies the local government can control independently, and by the private investments made in the locality. Each local influence factor is multiplied by the respective regression factors of the Barro and Sala-I-Martin empirical growth function. It is assumed that these local growth factors enhance local growth in addition to the national growth factors.

The model's growth function, in Vensim PLE language, is specified as follows:


Income growth rate = converged longterm average GDP growth rate+growth effect of the rule of law+growth enhancing effect of investment expenditures+growth enhancing effect of local education expenditures+growth reducing effect of local government consumption

Thus the local income growth rate is determined by the following five main factors:

  • The first factor is a converged longterm average GDP growth rate. The national growth rates are assumed to converge to 2 percent p.a. as described in Barro (1999). This is assumed to be reached at a household income of 40,000 US$, which is the average household income 1998 in Schwende Switzerland.
  • The second factor is the growth reducing effect of local government consumption. Barro and Sala-I-Martin (BSM, 1995) estimated that government consumption expressed in percent of national income (here total village income) in per capita terms contributes -12.9 percent to the change of the national growth rate, here longterm village growth rate.
  • The third factor is the growth enhancing effect of local education expenditures. The BSM coefficient is +22 percent of total education expenditures expressed here in percent of total village income.
  • The fourth factor is the growth enhancing effect of investment expenditures. The BSM coefficient is 3 percent of public and private investment in percent of total village income.
  • The fifth factor is the growth effect of the rule of law. The BSM coefficient is 2.93 perecent. It is assumed if the rule of law improves the growth rate is enhanced by 2.93 percent. If it deteriorates the growth rate declines by 2.93 percent.

The rule of law is expressed as the sum of public finance leaks and irregular finances expressed in percent of total regular LG income, and is called rule of law index. The effect is based on the Knack and Keefer (1995) country index prepared for the International Country Risk Guide, which is the basis of the Barro and Sala-i-Martin rule of law coefficient. Knack and Keefer use a 1 to 6 scale with 6 indicating the most favorable and 1 indicating the worst state of the rule of law present in a country. BSM have adjusted this scale to a range of 0 to 1, with 0 indicating the worst maintenance of the rule of law, and 1 the best. Here the scale is transposed and divided by 10, thus 0.1 indicates the worst and 0 indicates the best maintenance of the rule of law. This was necessary because here the coefficient enters negatively into equation 3, with more irregular finance decreasing the longterm average growth rate, as Barro and Sala-i-Martin use it as a positive effect. It is assumed that the worst state is reached when the irregular public finances in a locality reach 10 percent of total local village income. Figure 6 shows the local growth process.

Figure 6

FIGURE 6: Income Growth Process

In step 17 of Figure 2, disposable household income, public expenditures reduced by the villagers valuation of the tax costs incurred and, private investment determine the perceived development status attained through the villagers' development perception process. In this process, the villagers' valuation of their infrastructure and non-infrastructure needs are expressed as percentage of all needs cited in the field survey, and are entered as utility values into valuation lookup functions that drive the villagers' valuation of the local infrastructure and non-infrastructure expenditures according to the law of diminishing marginal utility. Table 6 below reports the needs and tax cost coefficients of the four localities.

TABLE 6: Villagers' Tax Burden and Needs as Percent of all Needs Cited in Question 30 of the Field Research

Locality Infrastructure Needs Non-Infrastructire Needs Tax Burden
Huay Yai 74.8% 16.5% 0.415
Labac 45.3% 54.7% 0.2825
Zorita 12.2% 87.8% 0.425
Schwende 17.9% 53.5% 0.3025
The difference of the needs with respect to 100 percent accounts for the villagers who stated that they had no needs.

As is to be expected, there are large differences in needs between the localities. A striking result is the low value of Huay Yai, and the high value of Zorita with respect to non-infrastructure needs. These values represent the local culture quite well. The Thais are used to being independent and to helping themselves, a fact which greatly helped in softening the effects of the recent economic crash in Thailand. The Spaniards, however, clearly assign the social safety net function to the state. The tax burden values are also a clear reflection of the two countries' tax levels. Thailand and Spain each charge higher taxes in their country category, and thus the villagers also give the perceived burden a higher value.

The tax burden valuation is integrated into each SD Model as described hereafter for Huay Yai. According to Q57 of the field research in Huay Yai, respondents assessed the value received from their tax, levies and fees payments on average as 2.34 (standard deviation 0.94) on a scale of 1 = practically no value to 4 = very good value. This scale and the average value was transformed to a scale ranging from 0 - 1 to match the metrics of villagers' development perception. The difference between the average value 2.34/4 = 0.585 and 1 (1 - 0.585 = 0.415) was then entered in the development perception as (negative) cost value, symbolizing the cost incurred with the local governments' efforts. This cost value is entered in the tax burden function at the initial tax rate of 5.81 percent, and increases with increasing tax rates.

The period-by-period aggregation of these perceived development and cost values establish the perceived development status. If the villagers' development expectations, defined as the previous period's development status grown at the converged economic growth rate, are not met, perceived development status leads to "Development Frustration," and thus a new process of bargaining and planning begins.

page 33


USC Seal

Main Page | Subscribe | Submission Requirements | Editorial Board | Archive | Links

ISSN 1548-6036

Copyright 1999-2000
University of Southern California
Los Angeles, California 90089-0626