SAG Publications
include journal articles, papers, and reports authored and co-authored by SAG staff members.

Click on the PDF icon to download publications.
An Analysis of the Effect of Reserve Activation on Small Business
Douglas B. Christman, John B. Hope, & Patrick C. Mackin
October 2009
Abstract 

The authors use Department of Defense (DOD) data on the employment and activation of military Reserve personnel and Dun and Bradstreet data on sales and firm size to examine the impact of Reserve activation on employers. The study includes an analysis of the size and industry characteristics of firms employing activated reservists as well as the impact of the length of activation on the firm. The authors find that small businesses are disproportionately affected at the margin. Adding one more employee increases the percent activated in small firms from 0.36% to 2.6% but increases the percent activated in large firms by only 0.05% to 0.06%. There is a corresponding negative effect of 0.30% on the change in sales for small firms, about 15 times greater than the 0.02 percentage for large firms.

The authors also find that the length of activation has a small but significant negative effect on the firm’s revenues. The econometric model found a 1.9 percentage point decrease in sales for small firms relative to larger firms for those with reservist employees activated 30 days or more and a 3.0 percentage point decrease in sales for those with employees activated 180 days or more.


The Relationship Between Employee Turnover and Employee Compensation in Small Business
John B. Hope & Patrick C. Mackin
July 2007
Abstract 

This study explores the relationship between employee turnover and firm size as it relates to compensation using the National Longitudinal Survey of Youth (NLSY). The purpose of this study is to examine whether employee turnover differences between small and large firms are the result of differences in wages and benefits or of some form of self-selection where employees of small businesses are simply more prone to high turnover rates than those in larger firms.


A Model of Reenlistment Behavior: Estimates of the Effects of Army’s Selective Reenlistment Bonus on Retention by Occupation
Patrick C. Mackin, Paul F. Hogan, Javier Espinosa, & Peter M. Greenston
June 2005
Abstract 

A logit model was applied to estimate the effect of selective reenlistment bonuses (SRBs) on the retention rates of Army Soliders. The model was estimated separately by occupational group and by first (zone A), second (zone B) and third term (zone C) reenlistment decisions. An “annualized cost of leaving” (ACOL_ variable was constructed to estimate the net financial returns to reenlisting in the Army compared to leaving for the civilian sector. The model was estimated using data on actual reenlistments from the period FY1990 through FY2000.

The effects of SRBs on reenlistments at Zones A, B, and C were estimated at three levels of occupational aggregations – all Army, CMF, and MOS. After out-of-sample testing, we re-specified and re-estimated the model. In general, the results for Zone A at all levels of occupational aggregation indicate that reenlistment bonuses have a positive and statistically significant effect on Zone A reenlistments. The magnitude of the effect varied by occupation, but a one-level increase in SRB at Zone A typically increases the reenlistment rate by three to seven percentage points, depending upon the occupation. The results for Zone B are also solid at both the CMF and MOS levels. Results for Zone C, where reenlistment rates are typically very high, were reasonably solid but not as good as the Zone A and Zone B results. We were unable to obtain positive, statistically significant ACOL parameter estimates for a small number of occupation groups. Statistically significant effects for demographic control variables and labor market conditions were also obtained.


Econometric Analysis of 2003 Data on the Post-Service Earnings of Military Retirees: Methodology Report
Patrick C. Mackin & Kimberly L. Darling
September 2004
Abstract 

This study describes the methodology used to evaluate work effort decisions and estimate earnings of military retirees based on data from the 2003 Survey of Retired Military (2003 SRM), supplemented with data on the civilian non- institutional population from the March 2003 Current Population Survey Annual Social and Economic Supplement (CPS-ASEC).

This report describes how the estimation datasets were constructed from these two data sources and describes the econometric methodology in detail, including the definition of alternative models designed to address empirical and data issues. The analysis datasets for the 2003 SRM and 2003 CPS-ASEC are discussed first, followed by a description of the basic model used throughout the analysis for both military retirees and comparable civilian workers. Application of this basic model estimate military retiree earnings and results of alternative model estimates are described next. Finally, Appendixes A-D include details about analysis datasets and regression results for enlisted and officer retirees and comparable civilians.


Econometric Analysis of 2003 Data on the Post-Service Earnings of Military Retirees
Patrick C. Mackin & Kimberly L. Darling
June 2004
Abstract 

This study provides estimates of the post-service earnings experience of individuals who retired from active military service during the period 1971-2001. The findings are based on data from the 2003 Survey of Retired Military (2003 SRM), supplemented with data on the civilian non-institutional population from the March 2003 Current Population Survey Annual Social and Economic Supplement (CPS-ASEC).

Post-service earnings are of continuing policy interest for a number of reasons. In particular, the earnings of military retirees may be lower than those of otherwise comparable civilians. A major reason for the shortfall, cited in previous studies, is military experience may not be directly transferable to civilian careers. Retirees may have to accept lower wages initially as they seek to acquire general job skills. Moreover, the discrepancy in earnings may vary depending on the retiree's military occupation; some military jobs may have relatively close civilian counterparts, while others do not. Previous empirical work has shown the retiree earnings gap disappears over time as retirees' wages approach those of their civilian counterparts.

The military retirement system is designed, in part, to offset this negative effect of a military career on an individual's earning capacity. There has been a great deal of attention focused recently on the issue of veterans' disability compensation for military retirees. Congress has acted to gradually repeal the prohibition against concurrent receipt of military retired pay and disability compensation from the Department of Veterans Affairs (VA). This study pays particular attention to the effects of possessing a VA disability rating on both the decision to work full time and on earnings.


Review of Military Death Benefits
Patrick C. Mackin, Richard Parodi, & Mark Dye
April 2004
Abstract 

The Principal Deputy Undersecretary of Defense (Personnel & Readiness) contracted with SAG Corporation to conduct an independent review that assesses the relative value and sufficiency of death benefits provided to survivors of members of the Uniformed Services. The requirement for this report stemmed from the Conference Report on the FY 2004 national Defense Authorization Act.

The study employs a functional categorization of death benefits to compare benefits across military and civilian occupations. Using this taxonomy, the report provides an overview of current military benefits and contrasts them with employer-provided death benefits for other occupations. The report next addresses alternative policy options to bring military death benefits more in line with those benefits received by other occupations, where those benefits were considered more appropriate for the sacrificed individual. The report concludes with recommendations for improvements to the military death benefits system and provides estimated costs for those changes.


Design of an Econometric Module to Support the ODCSPER Strength Management Systems Redesign
Patrick C. Mackin, Paul F. Hogan, & Peter M. Greenston
January 1998
Abstract 

This study assessed alternative approaches to incorporation an econometric module in ODCSPER’s redesigned strength management system. It includes an examination of the operation of the strength management system and an assessment of the analytical needs of strength planners. The study also looked at available econometric methodologies and results available in the literature. These findings were integrated with an evaluation of the major management issues for individual Army personnel communities.

Study findings include a recommendation on the module’s design specification, with description of key module algorithms and identification of appropriate econometric models and parameters. Other results include an evaluation of existing empirical parameters and recommendations for future econometric research to provide the module with new and improved parameters.


Financial Issues of Reserve Service: A Report from the 1992 Reserve Components Surveys
Patrick C. Mackin, Robert F. Lockman, Shelley Perry, & Mary M. Wetlin
July 1997
Abstract 

To better understand and plan for the needs of a changing military force, the Defense Manpower Data Center (DMDC) conducted the 1992 Reserve Components Surveys. Major topics that were addressed in the member survey include: reserve pay and other income sources, military benefits, family housing expenditures, and health and dental care for Reservists. The questionnaire was mailed to 76,783 members with responses received from 47%. The objective of this report was to analyze issues relating to individual and unit readiness focusing on financial aspects. Highlights of the report include the following: a) distance from commissary and exchange locations was the most limiting factor for users; b) 60% indicated they were satisfied with military pay and allowances; c) the percentage of household income spent on housing appears to decline with pay grade group; d) more than half of Reservists own their principle residences; e) 78% of all Reservists rated their medical insurance coverage as excellent or good and 69% gave an equally high rating for dental coverage.


Reserve Component Members: A Report from the 1992 Reserve Component Surveys
Patrick C. Mackin, D. Wayne Hintze, Shelley Perry, & Mary M. Weltin
July 1997
Abstract 

To better understand and plan for the needs of a changing military force, the Defense Manpower Data Center (DMDC) conducted the 1992 Reserve Components Surveys. Major topics that were addressed in the member survey include: amount of compensation and benefits, impact of service on civilian jobs and family life, quality of unit leadership, downsizing of Reserves, and perceptions about skill development and its relationship to civilian jobs. The questionnaire was mailed to 76,783 members with responses received from 47%. The objective of this report was to analyze issues relating to individual and unit readiness for mobilization and deployment. Highlights of the report include the following: a) individual preparedness increased as pay grade increased; b) junior enlisted were less likely to have workable dependent-care arrangements in case of emergency; c) the most common potential work-related problem from mobilization was loss of income; d) the burden placed on the spouse was the most cited family-related problem; e) nearly 90% of respondents reported that the opportunity to serve the country made at least a moderate contribution to their most recent retention/reenlistment decision.


Nuclear Officer Retention: MSR and Beyond
Patrick C. Mackin, Christopher D. Mackie, & Michael K. Nakada
October 1996
Abstract 

To combat low retention in the nuclear officer community, the Nuclear Officer Incentive Pay (NOIP) program was designed. A 1981 study of nuclear submarine officer retention found a significant, positive effect of NOIP on their retention at the end of the minimum service requirement (MSR). In 1996, the study was broadened to include nuclear-trained surface warfare officers. Again, a significant, positive effect of NOIP on the MSR retention rate was found.

This report documents the results of an investigation of historical nuclear officer retention behavior at the MSR decision point and 9 decision points later. It specifies an ACOL-2 model and quantifies the impact of the NOIP retention bonus program on 10 retention decisions

Separate models of retention were estimated for the submarine and surface nuclear officer communities. For both communities, the retention elasticities with respect to the NOIP retention bonus program were small, but significant, indicating that “pay does matter.”

These models can be used to assess the retention and cost impact of alternative NOIP retention strategies.


Estimation of Retention Parameters for the Prototype Officer Personnel Inventory, Cost and Compensation Model
Patrick C. Mackin, Lee S. Mairs, & Paul F. Hogan
October 1995
Abstract 

This research estimated a multiperiod Annualized Cost of Leaving (ACOL-2) model that predicts officer career decisions as a function of economic, demographic, and Army personnel policy (e.g., military compensation) influences. The panel probit estimation yielded statistically significant pay but not unemployment effects. The research also found that fixed, unobserved preferences for military service significant influenced retention behavior. The estimation encompassed up to 13 consecutive annual “decision” points, with data taken from ARI’s Officer Longitudinal Research Database, covering year groups 1979-1992. The retention parameter estimates were embedded in an Officer Personnel Inventory, Cost and Compensation (OPICC) Model. This PC-based prototype model was designed and developed to improve the Army’s ability to effectively manage its officer force by providing policy makers with accurate information about the impact of policy changes, including promotion policy, compensation, and separation incentives. The OPICC model provides estimates of the impacts of policy and economic changes to the Officer Personnel Management Directorate inventory for a 6-year projection horizon. The prototype version does not contain a cost estimation capability. The model was validated by using it to predict actual historical behavior.


A Multiperiod Model of U.S. Army Officer Retention Decisions
Patrick C. Mackin, Lee S. Mairs, & Paul F. Hogan
May 1993
Abstract 

This report describes a study on the determinants of officer retention behavior. Stay-leave decisions for field-grade active-duty officers in the Infantry and Signal Corps branches were examined using multiperiod ACOL-2 models. This effort expanded upon a pilot study involving the Air Defense Artillery branch. A multiperiod model that predicts officer career decisions as a function of economic, demographic, and Army personnel policies (e.g., military compensation) was successfully estimated with longitudinal data from the U.S. Army Research Institute for the Behavioral and Social Sciences Officer Longitudinal Research Data Base. The estimation showed that financial incentives exerted strong behavioral influence on retention, although there is evidence that the strength varies by branch. Civilian labor market conditions, as measured by the unemployment rate, had a significant effect on career decisions. Retention behavior also varied by source of commission, gender, race, and marital statues. Finally, unobserved heterogeneity had a significant impact: as officer cohorts age, the distribution of unobserved tastes for the military becomes truncated and retention rates rise.

This study also explored options for the design of a PC-based Officer Personnel Inventory, Cost and Compensation (OPICC) policy analysis model. The report presents a proposed structure, along with a discussion of modeling tradeoffs necessary to balance policy maker analytical needs and costs. The proposed structure for the OPICC model consists of three integrated modules: (a) a policy module that translates policy and other changes into changes in retention rates using econometric research results; (b) an inventory projection module that reflects the manpower effects of the policy changes; and (cc) a cost module, using the Army Manpower Cost System (AMCOS) methodology, to capture the resources impact of policy changes.


A Model of U.S. Army Officer Retention Behavior
Lee S. Mairs, Patrick C. Mackin, Paul F. Hogan, Robert Tinney
September 1992
Abstract 

This report summarizes the findings of a pilot study of the determinants of officer retention behavior. Stay-leave decisions for a sample of Air Defense Artillery officers were modeled in an ACOL-2 (Panel probit) framework. The estimation showed that the officers were sensitive to changes in civilian and military pay, as well as to the condition of the civilian labor market. Retention behavior also varies by source of commission, gender, race, and marital status. Finally, the panel probit specification confirmed that unobserved heterogeneity had a significant impact. As officer cohorts age, the distribution of unobserved tastes for the military becomes truncated and retention rates rise.

The study also included tests of alternative specifications for the pay variable and the size of the decision window, as well as an evaluation of the applicability of model results for a policy analysis model. The model will allow Army decision makers to track the effects of changes in policy, compensation, and economic conditions on the probability that officers will stay through key career decision points. Follow-on work will include estimating the model for all officers and increasing the number of decisions modeled.