NEW UPDATE: Completed trials may now upload and register supplementary documents (e.g. null results reports, populated pre-analysis plans, or post-trial results reports) in the Post Trial section under Reports, Papers, & Other Materials.
Tax Farming Redux: Experimental Evidence on Incentive Pay for Tax Collectors
Last registered on October 19, 2014


Trial Information
General Information
Tax Farming Redux: Experimental Evidence on Incentive Pay for Tax Collectors
Initial registration date
October 19, 2014
Last updated
October 19, 2014 11:31 AM EDT
Primary Investigator
Other Primary Investigator(s)
PI Affiliation
PI Affiliation
Harvard Kennedy School
Additional Trial Information
Start date
End date
Secondary IDs
One of the oldest questions in the design of a civil service system is the role of incentives, yet there is little rigorous evidence on their impact. In tax, performance pay for tax collectors could increase effort and revenues, but it might come at a cost if taxpayers experience greater pressure by collectors especially if the latter resort to extortion and over-taxation. We investigate these issues by conducting a large-scale, randomized field experiment. Working with the Punjab, Pakistan provincial government, property tax inspectors were randomly allocated into one of three high-powered incentive schemes or a control group for up to two fiscal years. We investigate the impacts on tax revenues, accuracy of tax assessments, and satisfaction with interactions with the tax department.
External Link(s)
Registration Citation
Khan, Adnan, Asim Khwaja and Benjamin Olken. 2014. "Tax Farming Redux: Experimental Evidence on Incentive Pay for Tax Collectors." AEA RCT Registry. October 19. https://doi.org/10.1257/rct.252-1.0.
Former Citation
Khan, Adnan et al. 2014. "Tax Farming Redux: Experimental Evidence on Incentive Pay for Tax Collectors." AEA RCT Registry. October 19. http://www.socialscienceregistry.org/trials/252/history/2933.
Sponsors & Partners

There are documents in this trial unavailable to the public. Use the button below to request access to this information.

Request Information
Experimental Details
Three performance pay schemes were tested. The "Revenue" scheme provided incentives based solely on revenue collected above a benchmark predicted from historical data. The “Revenue Plus” scheme provided incentives exactly as in the Revenue scheme, but made adjustments (plus/minus three-fourths of baseline salary) based on whether the circle ranked in the top, middle, or bottom third of circles in terms of taxpayer satisfaction and accuracy of tax assessments, as determined by an independent survey of taxpayers. The third scheme, “Flexible Bonus,” took this a step further by both rewarding collectors for a much wider set of pre-specified criteria set by the tax department, and by allowing for subjective adjustments based on period-end overall performance.
Intervention Start Date
Intervention End Date
Primary Outcomes
Primary Outcomes (end points)
Key outcomes of interest:
* Tax Revenue
* Taxpayer satisfaction and accuracy of tax assessments
Primary Outcomes (explanation)
Secondary Outcomes
Secondary Outcomes (end points)
Secondary Outcomes (explanation)
Experimental Design
Experimental Design

Revenue: This incentives treatment group rewards tax circle staff (inspectors, constables, and clerks) on the basis of increases in revenue collection. Each circle is assigned one of three bonus rates i.e. either 40%, 30% or 20%. The differential bonus rate takes into account the differences in the size of circles i.e. to promote equity larger circles, where revenue increases are relatively larger, will be compensated at a lower rate than smaller circles where it is more difficult to raise revenue. The benchmark for each circle is generated using historical collection data for that specific circle and represents expected revenue collection. Each inspector continues to receive his or her current base salary, plus a bonus proportional to the tax revenue he or she collects above the benchmark.

Revenue Plus:

The Revenue Plus treatment is similar to the Revenue treament, but checks against over-aggressive tax collection by factoring in assessment accuracy and taxpayer satisfaction by utilizing an objective third party customer feedback assessment conducted on a randomly selected sample of 50-100 properties in each tax circle. The idea is that in additional to simple increased revenue collection, other components of the tax collection process are important.

In addition to the bonus pay based on increase in tax revenue collected, tax personnel may also be punished for poor performance in situations such as under- or overtaxing properties (the revenue based honorarium is particularly sensitive to this possibility) and rewarded in cases where they are doing an accurate job. There is ample evidence from the theoretical literature that imposing such penalties and rewards raises performance. Moreover, since an important source of inaccuracy in property assessment is due to misclassification of properties, such customer feedback mechanisms can provide reasonably reliable checks on tax inspector performance.

The customer satisfaction and accuracy adjustment is based on the results of a third party survey conducted in all circles in the scheme. Accuracy is measured as the correlation between true tax liability and official tax liability. Scores for customer satisfaction and tax assessment accuracy are calculated using survey-based questions such as satisfaction with customer service, the degree to which proper procedures were followed, and satisfaction with the outcome in dealing with the tax department. Then these scores are added together to give each circle a total score, which is used to rank circles in this scheme and divide them into three groups (top, middle, and bottom) of approximately equal size. Circle staff get paid as in revenue treatment, but the top group gets an additional bonus equal to 0.75 times their base salary, and the bottom group lose 0.75 times their base salary.

As in the revenue scheme, each inspector continues to receive his or her current base salary, but the proportional bonus is adjusted based on accuracy and customer satisfaction.

Flexible Bonus:

This treatment aims to mimic the way bonuses work the private sector for many complex jobs: managers distribute a fixed bonus pool to talented employees based on all factors they observe. Everyone gets a base salary supplement equal to 1.5 times their salary: each inspector is entitled to an honorarium of Rs. 30,000/month, and each constable and clerk is entitled to an honorarium of Rs. 23,000/month. Each quarter, 50% of this payment will be made unconditionally, so that in Q1-Q3, each inspector will receive Rs. 15,000/month and each constable/clerk will receive Rs. 11,500/month. At the end of the fiscal year, a "Performance Evaluation Committee" divides circles in this group into three equal size groups -- top, middle, and bottom -- based on a holistic assessment of performance, including recovery, increase in net demand, customer satisfaction, accuracy, and a subjective "director's rating". All circle staff get paid 1.5 times their base salary top-up, but the top group gets an additional bonus equal to 0.75 times their salary, while the bottom group loses 0.75 times their salary (though slightly less in year 2).
Experimental Design Details
Randomization Method
Public lottery conducted by a computer.
Randomization Unit
Tax circles in effect as of July 2011
Was the treatment clustered?
Experiment Characteristics
Sample size: planned number of clusters
482 tax circles
Sample size: planned number of observations
482 (tax circles)
Sample size (or number of clusters) by treatment arms
Revenue - ~70
Revenue Plus - ~70
Flexible Bonus - ~70
Control - ~272
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
IRB Name
MIT Committee on the Use of Humans as Experimental Subjects
IRB Approval Date
IRB Approval Number
Post Trial Information
Study Withdrawal
Is the intervention completed?
Intervention Completion Date
June 30, 2013, 12:00 AM +00:00
Is data collection complete?
Data Collection Completion Date
Final Sample Size: Number of Clusters (Unit of Randomization)
Was attrition correlated with treatment status?
Final Sample Size: Total Number of Observations
Final Sample Size (or Number of Clusters) by Treatment Arms
Data Publication
Data Publication
Is public data available?
Program Files
Program Files
Reports, Papers & Other Materials
Relevant Paper(s)