The US Army Projects the Effect of Merit Pay on Payroll Growth

  • Authors:
  • Guy C. Huntsinger;Roger Chapman Burk;Timothy E. Trainor

  • Affiliations:
  • Department of Systems Engineering, United States Military Academy, West Point, New York 10996;Department of Systems Engineering, United States Military Academy, West Point, New York 10996;Department of Systems Engineering, United States Military Academy, West Point, New York 10996

  • Venue:
  • Interfaces
  • Year:
  • 2012

Quantified Score

Hi-index 0.00

Visualization

Abstract

The National Security Personnel System (NSPS) was designed to provide greater opportunity for merit-based pay raises than the preexisting civil service system. However, a US Army organization was concerned that over time the accumulation of merit raises would make the total payroll unaffordable. We were asked to investigate the conditions under which this situation would be likely to happen. We used the Vensim® modeling software to produce a model of the influences and events that would cause payroll to rise or fall from year to year. Our analysis showed that the merit pay policies in place at the start of our study were very likely to lead to undesirable growth in total pay and a concentration of all employees near the maximum allowed salary, defeating the purpose of NSPS—to reward better service with more pay. The most important factors were the fraction of merit pay that was given as merit raises (as opposed to bonuses) and the salaries of people newly hired to replace those who left the organization. We showed how departing senior employees should be replaced with less-senior ones to avoid long-term payroll growth. We used the model to find the combinations of merit pay and new-hire policies that would balance the effect of these two major influences and produce little or no long-term payroll growth. We also showed that some other influences that had concerned the client were actually of small effect compared to the stochastic uncertainty in payroll growth. Another influence, attrition, produced a more substantial effect. The interactive nature of the model allowed real-time exploration of alternatives with the client and helped build confidence in the robustness of the results. These results highlight potential problem areas needing careful analysis in implementing similar merit pay systems.