An Analysis of the Effect of Reserve
Activation on Small Business

by
John B. Hope, Douglas B. Christman
and Patrick C. Mackin
SAG Corporation
Annandale, VA 20003
for


Under contract no. SBAHQ-07-F-0306

Release Date: October 2009
This report was developed under a contract with the Small Business Administration, Office
of Advocacy, and contains information and analysis that was reviewed and edited by
officials of the Office of Advocacy. However, the final conclusions of the report do not
necessarily reflect the views of the Office of Advocacy.

---

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.

---

Table of Contents

Abstract ....................................................................................................................................... 1
1. Introduction............................................................................................................................. 1
2. Literature Review.................................................................................................................... 3
Retention Effects..................................................................................................................... 3
Reservist Family and Community Effects .............................................................................. 4
Employer Effects .................................................................................................................... 5
Firm Demographics ................................................................................................................ 6
Employee Absenteeism........................................................................................................... 6
Findings................................................................................................................................... 7
3. Data Description ..................................................................................................................... 7
Personnel Data Overview ....................................................................................................... 8
Personnel Demographics ...................................................................................................... 10
Activation Data Overview .................................................................................................... 12
Employment Data Overview................................................................................................. 14
Firm Data .............................................................................................................................. 15
4. Hypothesis............................................................................................................................. 17
5. Model Description ................................................................................................................ 17
6. Results................................................................................................................................... 19
7. Conclusion ............................................................................................................................ 26
8. References............................................................................................................................. 27
Appendix A – Demographics and Length of Activations for Reservists.................................. 29
Appendix B – Alternative Analysis to Check Consistency ...................................................... 44

Table of Figures

Figure 1: Number of Reservists by Service and Year..................................................................... 9
Figure 2: Number of Reserve Activations by Month.................................................................... 13
Figure 3: Total Employees of Combined HQ and DUNS by Firm Size....................................... 16
Figure 4: Firm Sales of Combined HQ and DUNS by Year......................................................... 17
Figure 5: Percent of Total Force by Service and Year.................................................................. 29
Figure 6: Education Level by Service........................................................................................... 30
Figure 7: Percent Female per Service ........................................................................................... 31
Figure 8: Enlisted Personnel’s Marital Status by Grade............................................................... 31
Figure 9: Officers’ Marital Status by Grade ................................................................................. 32
Figure 10: Warrant Officers’ Marital Status by Grade ................................................................. 32
Figure 11: Enlisted Personnel’s Dependents by Grade................................................................. 33
Figure 12: Officers’ Dependents by Grade................................................................................... 33
Figure 13: Warrant Officers’ Dependents by Grade..................................................................... 34
Figure 14: Enlisted Personnel’s Pay Grade by Year..................................................................... 34
Figure 15: Officers’ Pay Grade by Year ....................................................................................... 35
Figure 16: Warrant Officers' Pay Grade by Year ......................................................................... 36
Figure 17: Length of Activation – Air National Guard ................................................................ 36
Figure 18: Length of Activation - Air Force Reserve................................................................... 37
Figure 19: Length of Activation - Navy Reserve.......................................................................... 37

---

Figure 20: Length of Activation - Coast Guard Reserve .............................................................. 38
Figure 21: Length of Activation - Army Reserve......................................................................... 38
Figure 22: Length of Activation - Marine Corp Reserve.............................................................. 39
Figure 23: Length of Activation - Army National Guard ............................................................. 39
Figure 24: Length of Activation - All Services ............................................................................ 40
Figure 25: Total Activations by Service ....................................................................................... 40
Figure 26: Reservists’ Employment by Industry Group and Self-employment Status................. 41
Figure 27: Reservists’ Employment by Industry Group ............................................................... 42
Figure 28: Reservists’ Employers by Firm Size ........................................................................... 42
Figure 29: Length of Reservists’ Employment ............................................................................. 43

Table of Tables

Table 1: Ratio of Active to Reserve Component Personnel ........................................................... 9
Table 2: Change in Reservist Count by Component and Year ..................................................... 10
Table 3: Percent of Total Force by Component and Year ............................................................ 10
Table 4: Education by Service ...................................................................................................... 11
Table 5: Length of Activation by Component .............................................................................. 13
Table 6: Count of DUNS and HQ DUNS Numbers ..................................................................... 15
Table 7: Firm Industry Distribution.............................................................................................. 20
Table 8: Variable Description and Means .................................................................................... 21
Table 9: Change in Sales - Filtered Interactive............................................................................. 22
Table 10: Change in Sales – Filtered ............................................................................................ 23
Table 11: Probit Bankruptcy......................................................................................................... 24
Table 12: Probit 5% Loss.............................................................................................................. 25
Table 13: Crosswalk of Variables for Education.......................................................................... 30
Table 14: Percent of Total Activations by Component ................................................................ 41
Table 15: Extended Variable Descriptions ................................................................................... 44
Table 16: Change in Sales............................................................................................................. 45
Table 17 Change in Sales Lagged................................................................................................. 45
Table 18: Change in Sales – 180................................................................................................... 46
Table 19: Change in Sales - 180 Lagged ...................................................................................... 46
Table 20: Change in Sales - 180 Year .......................................................................................... 47
Table 21: Change in Sales – Interactive ....................................................................................... 47
Table 22: Change in Sales – Filtered 180 ..................................................................................... 48
Table 23: Probit Bankruptcy 180.................................................................................................. 48

---

1. Introduction

Since 2003, thousands of military reservists have been called to active duty; whether that duty
station is overseas or near home, these reservists are not available to work at their civilian jobs
during the period of activation. Businesses that experience employee absences due to Reserve
component activation must adjust staffing and are likely to experience productivity losses and
decreased performance. To the extent that these lost employees possess unique job-specific or firm-
specific skills or work in a team environment, the loss of productivity may substantially exceed the
employee’s foregone wages and lower the firm’s profitability. Moreover, the firm may face
increased costs due to mandatory requirements of the Uniformed Services Employment and
Reemployment Rights Act of 1994, also known as USERRA.1 These costs can include the
continuation of certain benefits for the absent employee and higher management costs necessary to
reallocate work and ensure that a comparable position is available upon the employee’s return.

The costs of job absence also fall on the activated employee; absences may adversely affect career
progression, for example. However, Reserve participation is a voluntary decision on the part of the
employee. Other things being equal, those who volunteer for Reserve duty are in jobs where such
absences are less likely to affect career progression and earnings. Firms have little latitude or choice
regarding Reserve participation by their employees. Indeed, employers are prohibited by Federal
law from discriminating against employees who participate in the Reserves.

Reservist activation may disproportionately affect small businesses in at least two ways. First, a
higher number of activated reservists may be drawn from small firms than would be expected given
their relative share of the total civilian workforce. Second, the average activation may have a greater
relative impact on smaller firms in terms of productive hours lost.

Previous work has shown that productivity losses from absenteeism are larger for small firms than
larger firms. Employees who have no perfect substitutes within the local production unit cause
productivity losses to their firms in excess of their wages.2 In firms with a small number of
employees, those employees may be more likely to perform a large number of tasks and be more
difficult to replace.

The main challenge in examining this issue is obtaining data. The Department of Defense (DOD)
maintains a database with employee-provided information on their employers. Reporting for this
database is now mandatory and includes a unique identifier called a DUNS number assigned by the
Dun and Bradstreet Corporation (D&B) for most of the firms. DOD estimates that its database
currently covers about 90 percent of members of the Selected Reserve, that portion of each service’s
Reserve components that trains regularly and is expected to be ready for quick mobilization at any
time. This database can be linked to other DOD data that identify which reservists have been
activated and for how long.

The Dun and Bradstreet Corporation has provided credit information on private businesses and
corporations since 1841. D&B is best known for its Data Universal Numbering System (DUNS) and

1 38 U.S.C. § 4301 et seq.
2 See Nicholson, et al. (2006).

---

its database with over 140 million business records with approximately 1,500 data elements. The
DOD employer database includes DUNS numbers that can be matched to D&B’s database.

Using these two data sources, we can construct a database of firms to identify differences in the
characteristics of employers of reservists. Compared with the employers of non-activated reservists
and otherwise similar civilians, are the employers of activated reservists more or less likely to be
small businesses? Are activated reservists concentrated in particular occupations? Are there any
differential impacts on firms in particular industries or regions of the country?

Small businesses are expected to face greater difficulties adjusting to extended absences of activated
reservists because they have fewer employees. They must either work shorthanded or hire
replacements, and there is an additional problem redistributing work when reservists return. The
model of firm performance will determine if there are measurable differences between the
productivity effects of small and large firms in terms of income and thus profitability when reservist
employees are called to active duty.

Since the first Gulf War began in 1990, the DOD has increased the operational tempo of the
Reserve components of the armed forces. In 2004 alone, over one-third of the service members
deployed in Iraq and Afghanistan were from the Reserves. Generally speaking, the department
includes National Guard components with the individual services’ Reserve components.

The Uniformed Services Employment and Reemployment Rights Act of 1994:

 Guarantees the right of reservists to be reemployed by their civilian employers after
serving on active duty;
 Prohibits employers from discriminating against individuals in any aspect of
employment because of their service in the Reserves; and
 Mandates some continuation of benefits to reservists who have been activated.

There is little information about the type and magnitude of the disruption that firms experience
when their reservist employees or reservist owners are activated.

The Congressional Budget Office (CBO) conducted a study in 2005 using survey information
collected by the Departments of Defense and Labor and by other organizations. CBO found that
most employers were unaffected by the activation of reservists; in fact, only about 6 percent of
business establishments employed reservists, and fewer than one-half of one percent of self-
employed people served in the Reserves. While this number may seem small when viewed from the
perspective of the economy as a whole, for those firms that are affected by Reserve activation, it is
important to know the degree of inconvenience to both the reservist and the firm.

Several studies have postulated that Reserve call-ups will be more severe for:


Small businesses that lose essential (key) employees;

Businesses that require workers with highly specialized skills; and

Self-employed reservists.

2

---

Among reservists who held civilian jobs, including government jobs, CBO found that small
businesses with fewer than 100 employees employed about 18 percent, and 35 percent were
employed by businesses with fewer than 500 employees or were self-employed. The CBO also
estimated that, of the 860,000 members of the Selected Reserves (who constitute the majority of
active Reserve personnel), 8,000 to 30,000 held key positions in small businesses and an additional
55,000 were self-employed.

The existence of laws such as USERRA could in fact cause additional difficulties for employers
with activated reservist employees. Not only could they experience a loss of productivity, but they
may also incur ongoing costs to maintain benefits for the absent employees. Policymakers are
required to balance conflicting goals in order to help DOD recruit and retain military personnel
while minimizing the effect on the civilian economy and avoiding harm to small business.

This analysis will proceed with a brief literature review chapter followed by descriptive analysis of
the data used. Next, we state the hypothesis to be tested, present an economic model of firm
performance, and report the results of our econometric analysis. Finally, the report provides a set of
conclusions and recommendations for further analysis.

2. Literature Review

The economic effects of National Guard and Reserve activation and deployment on small business
employers (or any size business for that matter) have not been extensively studied. Few, if any,
studies exist that directly or indirectly examine this relationship. This dearth of previous analysis is
partially attributable to the fact that large-scale activations of reservists did not occur prior to the
early 1990s. Before Desert Shield/Desert Storm (sometimes referred to as the first Gulf War), the
last major mobilization of reservists occurred nearly 40 years previously during the Korean conflict.
The Selected Reserve components before 1990 had practically no experience with any large-scale
force mobilization, and the likelihood of activation or deployment probably played almost no role in
enlistment and retention decisions.

Since that time, there have been few studies of the economic effects of Reserve deployment. These
studies tend to fall into one of three categories: retention effects; personal and economic effects on
the reservist, their family and community; and only recently, the economic effects on employers of
activated reservists. In addition to these military-specific studies, there are also some studies on the
general economic effects of long-term employee absenteeism for reasons other than military
activation – usually sick leave or sabbaticals – that would, in theory, have similar economic effects
on employers.

Retention Effects

The Department of Defense (DOD) commissioned the first survey of reservists’ attitudes toward
their military service in 1986: the Status of Forces Survey (Reserve) SOFS(R). In 1992, DOD
included in the SOFS(R) questions relating to how mobilizations affect reservists. DOD cited three
reasons for this focus:

3

---

1. Increased reliance on the Guard and Reserves had increased the likelihood of
mobilization.
2. Little empirical research existed concerning the effects that mobilizations have on
reservists' attitudes and those of their employers and families.
3. Reservists' stay/leave decisions are dependent on their own attitudes and perceptions and
those of employers and families.

Kirby, et al. (1992) studied the 1986 and 1992 SOFS(R) and found a shift in the motivation for
staying in the Guard and Reserve from purely pecuniary reasons to more of an emphasis on the
educational benefits available. This was accompanied by an encouraging favorable shift when asked
to describe their immediate civilian supervisors’ overall attitude toward their participation in the
Guard/Reserve.

The study used a sample from the 1991 Guard/Reserve Survey of Officers and Enlisted Personnel of
3,269 part-time enlisted reservists with 4-12 years of service, of whom 1,752 were mobilized. These
survey responses were matched to individual-level administrative data from the Quarterly Master
Personnel Files of the Reserve Common Component Personnel Data System (RCCPDS) maintained
by the Defense Manpower Data Center (DMDC). Kirby et al. examined differences in retention
among various subgroups and estimated the net effect of different variables on the probability of
retention. They found no statistically significant difference in retention between mobilized and non-
mobilized reservists.

Reservist Family and Community Effects

Angrist and Johnson (2000) analyzed the effects of deployment on military families in the first Gulf
War. Using DOD’s 1992 Survey of Officers and Enlisted Personnel, they found that deployments
increased the divorce rate for both male and female soldiers. Deployments also decreased spousal
employment. Children were not found to have an increase in disability due to parental deployment.

In a later study by the Rand Corporation, Loughran, et al. (2006) examined the impact of activations
and deployments on the lives of reservists, their families, and their communities. The authors
hypothesized that activation of reservists affects the supply of labor to the community. In addition,
the demand for goods and services decreases when reservists are removed from the community. In
order to test this, the authors generated counts of activated and deployed Reserve and active-duty
personnel from the DMDC Global War on Terrorism (GWOT) Contingency File. The research
showed that, although employment declines initially during the first month of activation, it recovers
over the subsequent three months. There was no conclusive evidence that Reserve activation
disproportionately drew from small communities. An additional interesting result was that police
employment decreased after activation, but the magnitude was difficult to gauge.

In a separate Rand study, Loughran, et al. (2006) examined the effect of activation on reservist
earnings using administrative data from DMDC and the Social Security Administration. Contrary to
earlier findings based on survey data, the authors found that most reservists did not suffer earnings
losses attributable to activation, but actually had earnings increases, particularly when the value of
certain tax advantages was considered. Still, about 17 percent of reservists did experience a loss in
earnings, and 11 percent experienced a loss of more than 10 percent of their base-year earnings.
4

---

Employer Effects

Doyle, et al. (2004) attempted to measure the costs incurred by businesses as a result of Reserve
activation. Using estimates of costs based on analysis of provisions in the Uniformed Services
Employment and Reemployment Rights Act (USERRA), and supported by interviews of a non-
random sample of 8 companies applying for the Small Business Administration’s Military Reservist
Economic Injury Disaster Loans, or MREIDLs, (selected precisely because they had suffered
serious economic consequences), they identified several sources of these costs. These included not
only direct financial costs associated with activated reservists’ retirement and health benefits and
transition costs for their return to the job they left, but also the cost of replacement employees, such
as hiring and training costs, and costs from lost productivity due to delays. Among the costs that
they were able to objectively measure were health and retirement benefits which they extrapolated
using the 2000 Survey of Reserve Component Personnel and data from the Employee Benefit
Research Institute (EBRI) using a weighted average by sector and firm size. They concluded that
the employer cost for a reservist’s retirement plan averaged $372 per month and that for an
employer-provided health insurance plan averaged $215 for singles and $550 for families. Other
long-term costs that may also result from the loss of key personnel include losing clientele and
effects on other employees in a team environment. They found that small businesses were
particularly affected by Reserve activations. The authors concluded that further research was
necessary to more specifically define the costs associated with Reserve mobilization and the
businesses incurring those costs.

Golding, et al. (2007) examined the combined effects of reservists’ activation and federal job
protections on civilian employers using interviews with 19 employers of 28 reservists; 12 of these
employers were MREIDL recipients. Using evidence from Nieva (1999), which found that about 6
percent of businesses employed reservists, along with anecdotal evidence, the authors concluded
that small businesses that lose essential employees, businesses that employ highly skilled workers
and self-employed reservists were most strongly impacted. Of the 860,000 reservists in the Selected
Reserves, about 85,000 were in positions where activation could severely affect their employers or
themselves as self-employed individuals. The study presented four options to mitigate the effects of
Reserve activation: compensation of affected businesses through tax credits or direct payments,
subsidized loans to employers, provision or subsidization of call-up insurance, and exemption of
certain reservists from call-up.

Gotz (2003) examined available information on general employer support of the Guard and Reserve
based upon the assumption that employer opinion of Reserve activities affects the decision of
current and potential Reserve participants. He also found a lack of hard statistical evidence but
plenty of anecdotal evidence. In order to improve access to relevant data, Gotz recommended that
DOD institute five initiatives to increase employer support for Reserve participation:

1. Establish a mandatory-reporting employer database linked to Reserve Personnel file.
(This policy was approved in August 2004 when the voluntary DOD Reserve
Employer Survey became mandatory.)
2. Obtain timely information for early warning of problems.
3. Provide more timely information to employers.
4. Decrease uncertainty about call-up frequency and duration.
5. Offset employer costs.
5

---

As a direct result of these recommendations, the Civilian Employer Information (CEI) database was
created, and it is this repository that we hope will provide crucial data in our study of Reserve
activation effects.

Firm Demographics

The Dun and Bradstreet Corporation has provided credit information on private businesses and
corporations since 1841. Often referred to as D&B, it is best known for its Data Universal
Numbering System (DUNS) and its database with over 140 million business records with
approximately 1,500 data elements.

Evans (1987) examines the relationship between firm growth, firm size, and firm age for a sample
of manufacturing firms between 1976 and 1982. Data from the Small Business Data Base (SBDB),
which was originally collected by D&B for credit reporting, provides data on firm age, number of
employees, sales, and various corporate demographics. Evans found several problems with the data:
under-representation of very small firms; data that did not apply to the file year for which the data
were held; and a lack of information on firm acquisitions and mergers. As with most survey
datasets, the SBDB data had various errors due to firms reporting bad data, firms’ misunderstanding
questions, or D&B making various clerical errors. Overall, Evans found that the quality of the data
was reasonable compared with other datasets used by economists. The paper found that firm growth
decreases with firm age and size.

Struyk (1972) examines the impact of industry location on the Boston, Cleveland, Minneapolis, and
Phoenix metropolitan areas over a 32-month period. He found that industry locations significantly
changed the existing spatial configuration of the metropolitan areas. This research was conducted
by using D&B DUNS Market Identifier files to determine Standard Metropolitan Statistical Areas
(SMSAs) and which firms either relocated within an SMSA, entered the SMSA, ceased operating
within the SMSA or remained at their current location. He found that there were some biases toward
firms that ceased to operate in certain SMSAs, though other SMSAs provided highly accurate
accounts of entry and exit through crosschecks against secondary data.

Employee Absenteeism

There exists a small amount of research pertaining to the more general case of employees who are
absent from work for other reasons (whether for a major illness, a sabbatical, or Family Medical
Leave Act (FMLA) reasons). We can extrapolate that the firm effects will likely be similar to the
temporary loss of a reservist employee due to activation or deployment.

De Kok (1997) constructed an explanatory model of involuntary absence from work based on a
telephone inquiry of 900 small and medium-sized (fewer than 200 employees) Dutch firms in 6
different sectors (industry, construction, trade and catering, transport, financial and business
services).3 Due to a small survey sample that increases the possibility of biased parameter estimates,
he modeled the impact of a firm’s precautionary measures on involuntary employee absences using
a two-stage estimation technique. The first endogenous measurement equation estimated a set of

3 The data were originally collected in a previous study, Bosch and de Kok (1997).
6

---

Document Outline

• ÿ
• • ÿ
  • ÿ
  • ÿ
• ÿ
  • ÿ
  • ÿ
• ÿ
• ÿ
• ÿ
• ÿ

---