The successful execution of public works programmes undertaken by governments often hinges on the completion of a vast array of local-level projects intended to create tangible amenities to improve the lives of citizens. However, in practice, a significant proportion of projects often remain incomplete. This article examines what drives incomplete projects under the Member of Parliament Local Area Development Scheme (MPLADS), a constituency development fund for public works projects in India.
In many parts of the world, governments implement public works programmes aimed at providing a range of goods and services to citizens. The successful execution of these programmes often hinges on the completion of a vast array of local-level projects intended to create tangible amenities to improve the lives of citizens. However, in practice, a significant proportion of projects often remain incomplete. Why do such failures in project implementation occur? Our research examines this question in the context of the Member of Parliament Local Area Development Scheme (MPLADS), a constituency development fund for public works projects that gives each Indian MP an annual allotment of Rs. 50 million per year to spend on creating basic infrastructure in their constituencies such as roads, school buildings or improved drinking water facilities. The scheme gives MPs discretion in deciding which public works projects to recommend to district officials, who are responsible for approving these recommendations and implementing the resulting projects. While this implementation is officially in the hands of the district bureaucracy, there are several informal channels through which MPs can influence the process. Yet, in many instances, projects that MPs recommend and that gain initial approval often fail to be eventually completed. These individual failures in project completion contribute to overall difficulties that the MPLADS has faced in achieving its stated goal of “meeting the felt needs of the people”.1 Our recent research (Thomas and Darsey, in progress) seeks to shed light on what drives incomplete projects under the MPLADS.
Explaining the failure to complete MPLADS projects
We argue that one important explanation for the failure to complete projects under the MPLADS can be found in the incentives for MPs that are generated by electoral cycles. Indian MPs in the Lok Sabha (lower house of parliament) face elections at least every five years and these elections are highly competitive affairs. While electoral competition is generally perceived to be a positive, we argue that the competitive pressures that MPs face in the run-up to elections could lead to incomplete projects via two channels.
The first – which we term the ‘voter demand’ channel – posits that the projects demanded by voters differ in terms of their inherent likelihood of completion. While some proposed projects are relatively straightforward, others are more complex and have a higher risk of remaining incomplete for administrative or technical reasons. Such reasons include project proposals requiring one or more exceptions to the programme’s requirements or where permissions to use land or other resources may be difficult to obtain. Ordinarily, an MP might be reluctant to recommend the latter type of projects. However, in the run-up to elections, MPs face increased pressure to satisfy voters by recommending the projects they demand regardless of the likelihood that these projects will be completed. This is in part because if an MP proposes a project in the run-up to an election, voters may only learn the final outcome of the project – its completion or lack thereof – after the upcoming election and often far away from the subsequent one. Thus, an MP seeking to signal her responsiveness to voters may be more willing to adopt more lax standards in terms of the types of projects they agree to recommend in the run-up to an election than at other times in the election cycle.
The second – which we term the ‘incumbent turnover’ channel – posits that the failure to complete projects is driven by the fact that elections frequently produce a change in the identity of the MP representing a given constituency. A newly elected MP may not be able to take credit for projects proposed by her predecessor and may therefore be reluctant to exert much effort to ensure the completion of such projects.
The two channels that we describe above suggest several testable hypotheses about MPLADS projects. First, we would expect there to be a greater proportion of funds devoted to eventually incomplete public works projects in the run-up to an election than at other times in the electoral cycle. Second, there should be a greater proportion of funds devoted to eventually incomplete public works projects in the run-up to an election when the election results in a turnover of an incumbent in the constituency than when it does not. Third, we would expect that project proposals submitted in the run-up to an election would be more likely to violate cost, administrative or technical criteria than project proposals at other times. We do not have a direct measure of the extent to which project proposals fail to meet these qualifications. However, since such violations typically give rise to back and forth discussions and negotiations between the MP and the district bureaucrat’s office, we test this third implication by examining whether project proposals submitted in the run-up to elections take a longer time to approve on average than project proposals submitted at other times.
Our analyses are based on a dataset that we assembled with publicly available project-level data from the MPLADS programme website and from the website of the Electoral Commission of India covering the period between 1993 and 2015. We assign each individual project proposal to a given constituency-month based on the date on which it was proposed by the incumbent MP in the constituency. We then record each project proposal’s eventual completion status at the time of data collection. Using this information, we are able to determine budget allocations under MPLADS in each constituency-month to projects that end up completed and those that do not.
Figure 1. Average allocation to projects by election period and completion status
To examine the first implication of our argument, Figure 1 shows how budgetary allocations to completed and not completed projects vary based on whether they occur in the 24 months prior to a national election (pre-election period) or at other times (non-election periods). The figure clearly shows that while the average amount allocated to eventually completed projects is much greater than the average amount allocated to incomplete projects in non-election periods, this difference shrinks in the pre-election period. Thus, consistent with our argument, pre-election periods witness an increase in the average amount of funds allocated to incomplete projects and a decrease in the average amount of funds allocated to completed projects. The same pattern holds even when we conduct a more rigorous statistical analysis that accounts for several alternative explanations such as the partisanship of the MP and the strain on bureaucratic capacity that may be produced by an increase in the total number and budget of projects proposed in the pre-election period.
Further analyses also show support for our second implication which is that projects proposed in the run-up to elections are less likely to be eventually completed especially in constituencies that have experienced incumbent turnover. Consistent with our third implication, we also find that even controlling for the number and budget of projects proposed in a given constituency-month, projects proposed in the run-up to an election take a longer time to approve on average than projects proposed at other times. Overall, thus, the results are consistent with our explanation for the link between electoral cycles and incomplete projects.
Since its inception in 1993, MPLADS has attracted increasing scholarly and media attention focusing on the various ways in which its politicised nature leads to the underutilisation of funds or the misallocation of funds over space and time (e.g. Keefer and Khemani 2009, Blair 2017, Thomas Bohlken 2018). Our research sheds light on a separate but related phenomenon related to the scheme – failures in the completion of individual projects – and shows how these failures are shaped by electoral cycles. Indeed, our findings shine a light on a curious pattern whereby MPs in the run-up to an election allocate significantly more funds than at other times to projects that eventually end up incomplete. We show how the competitive pressures generated by elections give rise to these patterns through voter demand and incumbent turnover.
Our findings have several policy implications for India, some of which can be addressed with changes to the MPLADS programme and others that should be considered in the design of future public works laws & regulations. Within MPLADS, the automatic ‘rollover’ of unspent MPLADS funds from one year to the next enables politicians to concentrate their project recommendations before anticipated election dates, which is associated with higher project failure rates. This effect could be reduced by eliminating this automatic rollover provision. In addition, providing information to voters on the efforts of incumbents, or lack thereof, with regard to the progress of specific public works projects could incentivize newly elected MPs to follow through on the proposals made by their predecessors. Political parties could also help by encouraging competent incumbents to stand for election again in the same constituency which could have beneficial effects on any future discretionary spending programmes. More broadly, our research suggests that policies that minimise discretion and that require more stringent and standardised criteria for the approval of project proposals could also reduce the negative effect of democratic elections on public service provision, as could demands by the public for greater accountability and transparency from lawmakers to design programmes to benefit the public instead of supporting the interests of incumbent politicians.
- Guidelines on Members of Parliament Local Area Development Scheme (MPLADS), Government of India, Ministry of Statistics and Program Implementation, June 2016: https://www.mplads.gov.in/MPLADS/UploadedFiles/MPLADSGuidelines2016English_638.pdf.
- Blair, Harry (2017), “Constituency Development Funds in India: Do they Invite a Political Business Cycle?”, Economic and Political Weekly, 5 August 2017, LII(31):99-105.
- Keefer, Philip and Stuti Khemani (2009), “When do Legislators Pass on Pork? The Role of Parties in Determining Legislator effort”, American Political Science Review, 103(1):99-112
- Thomas Bohlken, Anjali (2018), “Targeting Ordinary Voters or Political Elites? Why Pork is Distributed along Partisan Lines in India”, American Journal of Political Science, 62(4):796-812.
- Thomas, A and J Darsey, ‘How Electoral Cycles Shape the Implementation of Public Programs: Evidence from India’, Working Paper.