Payment for Services Through Local Taxes also Enhances Accountability and Performance
In a similar vein, several studies found that local governments that obtain a higher proportion of their revenue from local taxes—as opposed to central transfers, royalties or aid flows—seem to perform significantly better in terms of quality. This was a central conclusion in the study on local provision of water and education in colombia, but similar results also appear in the peruvian, philippines and indonesian cases, though this was not the focus of their research. In short, local authorities seem more effective when spending resources collected at the local level, than when using transfers from the central budget or royalties that do not require taxing local citizens. These findings, combined with the fact that in most countries decentralization has been based more on central transfers than on decentralizing authority over taxes, help explain why aggregate effects of decentralization on outcomes have been ambiguous.
Several factors could be at play here. Larger local tax collections may be an indicator of higher capabilities in general: some municipal governments may be more able to both collect taxes and deliver services. In fact, the recent literature on state capacity uses taxation as a good proxy for state capacity.42 Second, they can lead to more autonomy (with respect to national transfers which are normally tied), which, in turn, facilitates higher efficiency and responsiveness to local needs. Third, they can lead to more accountability, as citizens care more about what is done with their own payments.
The econometric estimates in the colombian study showed a positive and significant effect of the ratio of local taxes to total revenues (duly instrumented to overcome endogeneity problems) on education enrollment, education quality (measured by average student tests results), water coverage and water quality, controlling by other determinants of such outcomes. Further, the impact of local taxes on these outcomes was found to be statistically more robust and quantitatively more important than those of central government’s transfers and royalties. This is, indeed, a surprising result for conventional economics: it turns out that a dollar is not always a dollar, but that a dollar’s effect on outcomes depends on how it was obtained!
The fact that in the case of Colombia local taxes were found to have a much more signihcant impact on service outcomes in comparison not just with transfers, but also with royalties (which are not tied), seems to indicate that it is not just a question of having more resources, or more autonomy. Further, the authors built an index of responsiveness of expenditure allocations to local needs (represented by gaps in service outcomes with respect to other municipalities) and found a statistically positive effect of the instrumented ratio of own taxes to total revenues on such indexes.43
As the authors claim, these results strongly suggest that higher local taxes, as a share of total municipal revenues, bring about higher citizenry awareness on the allocation of the local budget and on service outcomes, encouraging more efficiency and quality of local public investments. Indeed, based on their qualitative work, the authors concluded that “when citizens pay more local taxes they have more incentives to monitor local politicians, and electoral accountability improves as public awareness of public policy performance increases.” Some previous studies in decentralization had obtained similar results and interpreted them in a similar fashion.44
Although the quantitative results of all the case studies mentioned in this section do not permit to fully decompose the channels through which payment for services and local taxes affect quality, both the quantitative and the qualitative analyses in the colombian case study (and the qualitative analysis in other GDN case studies) strongly suggest that payment and taxation are not just associated with better provider or state capacity and autonomy, but that they also promote more effective accountability.
Most citizens are not even likely to be informed about the amount of transfers, royalties or aid flows received by their local governments and/ or their opportunity cost. In contrast, they are painfully aware of their tax contributions and their private opportunity costs, just as they are when they pay directly for services. And they appear to demand that this personal effort is compensated with better services. Thus, accountability seems to follow own financing, whether through tariffs, fees, tolls or taxes. That is, effective accountability systems are more likely to develop when citizens and users pay, at least partially, for their services directly or through local taxes. It seems that when there is no pain, there is no motivation to demand.
In addition, the GDN case study of water supply in Romania, Hungary and Armenia found that non-payment can lead to inefficient overinvestment.45