CAB D0014440.A1/Final June 2006 Manning Under AIP Peggy A. Golfin 4825 Mark Center Drive • Alexandria, Virginia 22311-1850 Approved for distribution: June 2006 Henry S. Griffis, Director Workforce, Education and Training Team Resource Analysis Division CNA’s annotated briefings are either condensed presentations of the results of formal CNA studies that have been further documented elsewhere or stand-alone presentations of research reviewed and endorsed by CNA. These briefings repre- sent the best opinion of CNA at the time of issue. They do not necessarily represent the opinion of the Department of the Navy. Clear For Public Release; Distribution Unlimited. Specific authority: N00014-05-D-0500. Copies of this document can be obtained from the Defense Technical Information Center at www.dtic.mil or from the CNA Document Control and Distribution Section by calling (703) 824-2123. Copyright © 2006 The CNA Corporation Executive summary The Navy began offering Assignment Incentive Pay (AIP) for hard-to-fill billets in three geographic locations in June 2003. At that time, N13 asked CNA to analyze various components of the experiment. Reference [1] describes our analysis. In this CNA-initiated project, we analyze different aspects of the experiment that are now possible because it is no longer in its initial stages. In particular, sufficient time has elapsed for Sailors who were initially assigned under AIP to roll to their new assignments, as well as to make retention decisions. In the first requisition cycle that AIP was offered, there were 222 jobs, with 100 applications. Through the 3 March 2006 cycle, 7,503 jobs have been filled—just 5 percent of these filled with Sailors involuntarily assigned. The application rate per job posted has increased appreciably since AIP first began, going from .45 application for every job in the first cycle, to 0.69 application up through the most current cycle, representing an increase of about 50 percent in 3 years. The rationale for AIP was that it would benefit the Navy in two ways: (1) higher retention and lower attrition due to greater volunteerism and (2) a reduction in endstrength from converting Type 3 locations to Type 6 locations. Our earlier analysis of AIP determined that the savings from converting Type 3 to Type 6 billets (the second benefit) amounted to approximately $2,200 per month, per billet . The average bid of Sailors to date, $382 per month, has remained well below this break-even estimate. Our present analysis finds further evidence of the second benefit of AIP—specifically, improved sea manning. This was made possible by a sizable decrease in the number of sea duty billets available to Sailors applying for sea duty jobs, at the same time that the supply of Sailors available to fill those jobs remained fairly constant. Hence, overall sea duty manning improved, with benefits accruing to the Navy in a number of ways, some of which are captured with the break-even estimate of $2,200 per job per month. Using this estimate and the average bid to date, we calculate a potential savings of over $114 million each year in the conversion of Type 3 billets alone. The loss of Type 3 billets, however, means that Sailors rotatingto sea duty have fewer options than before AIP, due to the near elimination of the onlytype of sea duty that does not require Sailors to deploy for long periods of time. We foundthat the relative retention of single parents rolling to sea duty in both Zone A and Zone B,compared with the retention of all other Sailors rolling to sea duty in each zone, has decreased slightly since AIP began. In absolute terms, however, the retention of all Sailors rolling to sea in Zone A and Zone B decreased sharply in FY04, regardless of gender or single-parent status. We speculate that a number of factors may have contributed to the overall decline in reenlistments, such as Perform To Serve (PTS), and a reduction in the unemployment rate. 1 AIP also has implications for shore duty. We noted in our analysis that one hypothesized benefit of AIP, an increase in retention resulting from an increase in volunteerism, was impossible for us to assess. Volunteerism, especially in terms of assignments, has previously been unmeasured and has garnered little high-level attention—leaving us with no basis for comparison, either within locations before and after AIP or across AIP locations after AIP began. What we can conclude, however, is that the attrition and retention behavior of Sailors assigned to AIP shore locations, both voluntarily and involuntarily, is similar to that of Sailors assigned at the same time to other shore duty jobs. It is impossible to know how many of these Sailors would have attrited or failed to reenlist had they been assigned to these same AIP locations in the absence of an incentive, especially since they were either already less desirable as Type 6 than other overseas shore jobs (Misawa) or because they became less desirable when they converted from Type 3 to Type 6 (Naples and Sigonella). It is not surprising that manning did not improve in shore jobs,especially in AIP locations; unlike sea duty jobs, the supply of shore duty jobs increased with AIP without a comparable increase in the supply of Sailors available for these jobs. Exacerbating the shore manning difficulties is the fact that some of the early AIP locations had a large number of IT jobs. Our analysis showed that overall IT manning has decreased considerably since AIP began, and this is especially true for shore duty jobs. Specifically, there were only enough IT Sailors E4 and above in inventory in September 2002 to fill 91 percent of Billets Authorized (BA); their numbers decreased to 87 percent by September 2005. Given these shortages, it appears that the Navy has chosen to man Type 2 and Type 4 sea duty jobs fairly close to 100 percent, while manning shore billets at a much lower level; although CONUS shore duty manning decreased 6 percent since 2002, combined Type 3/6 manning decreased 18 percent. This helps to explain why IT manning in Naples and Sigonella decreased since AIP began, in spite of the incentive, and why the Navy has not chosen to involuntarily assign more IT Sailors to these jobs. In fact, in relative terms, IT manning in Misawa fared well compared with other shore locations. We found that, while there has been no significant change in the quality of Sailors assigned to AIP locations, in terms of AFQT and “fast promoters”—the latter defined as those promoting to E5 as fast as the first 25 percent to promote to E5 in their rating and accession cohort— there was a slight improvement in both metrics in a control group of Sailors rotating to Type 6 duty in Rota and Atsugi since AIP began. Also relative to our control group, we found an increase in the proportion of male Sailors in Naples/Sigonella and a decrease in the proportion of Asian/Pacific Islanders in Misawa. Finally, both of our AIP location groups (Naples/Sigonella combined, and Misawa separately) as well as the Rota/Atsugi control group experienced a decrease in the proportion of Sailors who are married and have children since AIP began, but the AIP locations experienced less of a decrease,suggesting that AIP may have helped reduce the loss of these types of Sailors in these locations. We found a few other changes in the characteristics of Sailors coming on board after AIP began, but these changes were also found in our control group. 2 Therefore, we conclude that these changes are not attributable directly to AIP. In particular, we noted that in most of the locations under investigation there has been a general shift toward more Sailors in the E4 and/or E5 paygrades, with fewer E6s. While some of the shift can be explained by relatively small changes in the requirements of billets in these locations, the magnitude of many of these changes is far greater than the change in requirements. Another finding is a fairly large decrease in the experience of Sailors in the mid- career paygrades (E4 through E6) and an increase in the experience of Sailors in paygrades E7 through E9. Some of the loss of experience in Naples and Sigonella could have been due to the fact that these locations changed sea/shore duty type with AIP, and Sailors within the same paygrade typically roll tosea duty versus shore duty at a different length of service (LOS). Even so, we found this phenomenon in other locations that did not experience a comparable change in duty type, so we pursued this further. We concluded that the loss of experience of mid- career Sailors is largely due to the 452-percent increase in BA for Sailors in the Master-at-Arms (MA) rating since the September 11th terrorist attacks, which has also caused the average experience of mid-career MAs to decrease considerably. In particular, the average LOS of all E4 and E5 MAs has decreased roughly 25 percent since the end of FY02, while the LOS of MAs at higher paygrades remains relatively unchanged. The increase in the requirements for MAs and the decrease in their length of service throughout the Navy explains only part of the overall loss in experience in the AIP and non-AIP locations we analyzed. While it is beyond the scope of this study to determine what has caused a general decrease in the experience of E6s and an increase in paygrades E7 and above that appear to affect more than just AIP locations, we speculate that phenomena related to the drawdown may be partly responsible. The loss of experience is one reason why there are fewer marriedSailors with children in the OCONUS locations we looked at, but even when we control for this, the loss of these types of Sailors persists, which could have significant implications for day care, schools, housing, and so on, in these overseas locations. We urge the Navy to monitor these manning changes to see if this is a longer term trend. Finally, the purpose of this research was to look at AIP as currently designed and implemented. We think it is important to note, however, that AIPhas a number of features that could help the Navy to achieve its “Strategy for Our People” goals. For instance, because AIP is not an across-the-board incentive for all Sailors, but instead allows Sailors to state the level of incentive they would be willing to accept to take a particular assignment, AIP might be a more cost-effective sea duty incentive or selective reenlistment bonus (SRB). Therefore, we urge the Navy to investigate additional ways in which AIP could be used effectively. 3 Manning Under AIP Peggy Golfin June 2006 Assignment Incentive Pay (AIP) is additional monthly pay that Sailors receive for being assigned to hard-to-fill billets. Sailors submit a bid for a job that has been designated as AIP-eligible on the Job Advertising and Selection System (JASS). Each job requisition cycle, detailers select the Sailor with the lowest bid from all qualified applicants for each AIP job. The Navy began offering AIP to three geographic locations in June 2003. Two locations, Naples and Sigonella, included Type 3 sea/shore billets that were converted to Type 6 under AIP.1 The third, Misawa, included billets that were Type 6 both before and after AIP. At that time, N13 tasked CNA with analyzing various components of the experiment. Our analysis is described in [1]. In this CNA-initiated project, we analyze different aspects of the experiment that are now possible because it is no longer in its initial stages. In particular, sufficient time has elapsed for Sailors who were initially assigned under AIP to roll to their new assignments, as well as to make retention decisions. ____________ 1For reference, Type 1 jobs are continental United States (CONUS)shore duty, Type 2 jobs are CONUS sea duty, Type 3 jobs are outside CONUS (OCONUS) shore jobs that are given sea duty credit, Type 4 jobs are OCONUS sea duty, and Type 6 jobs are OCONUS shore duty. 5 Background • CNA reported analysis of first year of AIP in June 2004 – First three locations: Misawa, Sigonella, Naples – Our conclusions: • AIP very cost effective, especially in Sigonella and Naples • NCTAMS jobs require higher caps – Cost effective up to $2,200 – Above that, Navy may need to revert them to Type 3 • Type 3 locations should be next AIP candidates, in order of desirability • JASS use does not differ by location or type of duty • It is too early in the experiment, then, to determine manning consequences AIP was created to help alleviate shortages in hard-to-fill billets. Other incentives existed for this purpose, such as offering points toward promotion, choice of next assignment, and sea duty credit for shore assignments (Type 3 billets). These incentives were found to be inflexible and costly. In particular, Type 3 billets reduce readiness and require a higher endstrength than would be required if those billets were shore duty. Another method to fill these and all other types of billets has been to involuntarily assign Sailors, a practice that negatively affects retention [2, 3]. Hence, the rationale for AIP was that it would benefit the Navy in two ways: (1) higher retention and lower attrition due to greater volunteerism and (2) a reduction in endstrength from converting Type 3 locations to Type 6 locations. Expanding on the methodology developed in [2, 3], our earlier analysis of AIP determined that the savings from converting Type 3 to Type 6 billets (the second benefit from above) amounted to approximately $2,200 per month, per billet. As long as the average bid in formerly Type 3 locations was below this break-even point, AIP was cost effective. In fact, the average bids for non-NCTAMS1 jobs in Naples and Sigonella were both well below this level, and we calculated that the benefits from AIP were about 7 times greater than the costs in those locations. While the ratio of benefits to costs for NCTAMS jobs (2.8) was lower, AIP was ____________ 1 NCTAMS stands for Naval Computer and Telecommunications Area Master Station. These AIP jobs, which mostly involve Sailors in the IT rating, have been the most difficult to fill. 6 still cost effective. However, we noted then that the costs were based on voluntarily assigning Sailors, with something less than a 100-percent fill rate for these jobs. The cost to involuntarily assign Sailors in order to achieve a 100-percent fill rate (or whatever rate the Navy determined was optimal) would most likelybe higher. Further, we noted that the retention benefits of AIP in Type 6 locations,and Misawa in particular, were not as easy to calculate because of a lack of any experience with the impact of such an incentive on retention. At that time, however, the average bid for Misawa was generally below $150 for all ratings, with just 4 exceptions, involving 15 Sailors. Our other major findings from [1] are summarized in the background slide on page 6. 7 AIP Details • Now expanded to include: – Australia, Azores, Bahamas, Bahrain, Belize, Bolivia, China, Cuba, Egypt, England, Gibraltar, Greece, Guam, Guinea, India, Israel, Indonesia, Italy, Jamaica, Japan, Korea, Kuwait, Lithuania, Malaysia, Morocco, Oman, Panama, Peru, Portugal, Qatar, Russia, Senegal, Singapore, Spain, Sri Lanka, Tanzania, Thailand, Turkey, United States, Ukraine, Uruguay, Virgin Islands – 36,845 total jobs posted, 25,466 applications as of 3 Mar 2006 cycle – Increase from 222 jobs,100 applications in first cycle • 7,503 jobs filled as of 3 Mar 2006 – Average bid = $382 – 5% involuntarily assigned • Budget increased from $1.0M in FY03 to $26.5M in FY06 Since we conducted our original analyses, AIP has expanded to additional locations and to include Type 2 and Type 4 billets. We list the current geographic locations in this slide. Note, however, that not all billets of a particular type are eligible for AIP in each location; AIP may be offered in only certain Unit Identification Codes (UICs) and for selected ratings and paygrades. Appendix A provides a summary of the locations and ratings, and the start date for the location. In the first requisition cycle in which AIP was offered, there were 222 jobs, with 100 applications, resulting in 0.45 application for every job posted. Through the 03 March 2006 cycle, there has been 0.69 application for every AIP job, an increase of about 50 percent in 3 years.1 The increase in the number of eligible jobs has also required additional funding; the budget for AIP has increased from just $1 million in FY03 to $26.5 million in FY06 (PERS-401). As we noted previously, we estimated that the cost-effective break-even point for AIP bids was $2,200 and that the average bid early on was well below that. That continues to be true. According to PERS-401, the average bid for Naples, Sigonella, and Guam (the other formerly Type 3 duty location selected for AIP so far) through the 18 May 2005 JASS cycle (the last date for which we have ____________ 1 Our statistics are derived from monthly updated AIP reports distributed by PERS-401.Due to Hurricane Katrina, JASS was down for 4 cycles (of the 59 since AIP began). Information pertaining to the number of jobs posted and applications made during that time period are not included in these figures. 8