1 Differential Reporting Incentives and their Impact on Financing Decisions: A Look at Public Defense Contractors Suzanne G. Morsfield Visiting Assistant Professor New York University—Stern School Christine E.L. Tan Assistant Professor Baruch College—City University of New York June, 2005 Preliminary Draft. Please do not quote or distribute without permission. Please address all correspondence to: Suzanne G. Morsfield NYU—Stern School Department of Accounting 44 W. 4 th St., Room 10-85 New York, NY 10012 USA Email: [email protected]Phone: (212) 998-0046 Fax: (212 995-4004
38
Embed
Differential Reporting Incentives and their Impact on ...w4.stern.nyu.edu/accounting/docs/Differenential Reporting...transaction structures in publicly-traded companies contracting
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
1
Differential Reporting Incentives and their Impact on Financing Decisions: A Look at Public Defense Contractors
Suzanne G. Morsfield
Visiting Assistant Professor New York University—Stern School
Christine E.L. Tan Assistant Professor
Baruch College—City University of New York
June, 2005 Preliminary Draft. Please do not quote or distribute without permission.
Please address all correspondence to: Suzanne G. Morsfield NYU—Stern School Department of Accounting 44 W. 4th St., Room 10-85 New York, NY 10012 USA Email: [email protected] Phone: (212) 998-0046 Fax: (212 995-4004
2
Abstract
This initial study provides evidence supportive of the notion that government defense
contractors may make significantly different financing decisions relative to non-defense contractors—for operating leases and debt, in particular. We also show consistent support that the relation between operating leases, taxes, and government defense contractors is as predicted by theory and as expected given a basic understanding of government contracting reporting and revenue requirements. Capital leasing decisions are not consistent with any of our predictions. Our results are robust to additional controls for the firm’s ex-ante financing decision, industry, the firm’s tax reporting incentives and the firm’s financial reporting incentive. The implication of our results indicates that the additional reporting requirements faced by defense contractors influence their asset financing decisions in a way that maximizes their government-related revenues. Given the heightened scrutiny of government contractors in general and the recent high-profiled scandals involving certain government contractors, these results are of interest to the government, regulatory bodies, and the tax-paying society at large.
1
Introduction
This paper examines whether differential reporting for government defense contractors is
related to variation in their respective financing decisions. Recent popular business media stories
draw significant attention to potential financial misreporting and billing, or to questionable
transaction structures in publicly-traded companies contracting with the federal government—
e.g., Halliburton’s over-reporting and billing of expenses in Iraq (King and Simpson, 2004), and
Boeing’s synthetic lease arrangement with the Air Force. Given these discussions are often set
in the context of political concerns, but in addition have wide implications for the general tax-
paying society, this study attempts to inject some initial and objective evidence into the
discussions regarding the potential impact of contractors’ reporting incentives on some of their
financial choices.
We specifically test whether government defense contractors and non-government
contractors make different asset financing decisions with regards to the use of long-term debt,
capital leases, and operating leases. To our knowledge prior accounting-related government
contractor research has only focused on cost-shifting driven by the reporting regime differences
(Lichtenberg 1998; McGowan and Vendrzyk 2002). Our study examines actual financial
transactions. And although the finance literature has examined the debt, operating, and capital
lease financing choices in some detail, it has predominantly focused on the tax reporting aspects
of these transactions (Fama and French 1998; see Graham, Lemmon, and Schallheim 1998 for a
complete review).
We provide a research design that attempts to alternately examine the effects of the U.S.
tax, GAAP, and DCAA (Defense Contractor Auditing Agency) reporting regimes. We also try
to empirically isolate the impact of the government reporting incentives on the financing choices
2
for our sample of firms. Finally, we examine whether the macro factors such as post-9/11
changes in the U.S. tax code and defense spending for example, are detectable in the financing
decisions of our sample of firms. The cumulative financing decision measures and basic model
are drawn from the seminal Graham, Lemon, and Schallheim’s (1998) study of the role of tax
status in these decisions.
Our study shows consistent evidence that government contractors make significantly
different financing decisions compared to non-contractors. Once we control for both the GAAP
and tax financial reporting incentives, our findings on the defense contractor’s incentives remain
strong. We utilize a sample of the 41 publicly-traded defense contracts taken from the top 100
defense contractors list published by the Department of Defense (DoD) each Congressional
Budget year. The top 100 list is ranked by total contract revenue awarded and are publicly
available on the DoD’s procurement website. In addition, we also include in our sample other
defense contractors not in the top 100 list by conducting a search on ‘defense contract’ in the
SEC Edgar database that have the necessary Compstat data. Our final sample of defense
contractors is 103 firms. We match these firms with a sample of 103 non-government companies
based on year 2000 beginning debt/equity and industry (i.e., two-digit SIC). By controlling for
beginning of year debt and industry norms, we are able to begin to separate differences in the
underlying firms’ economic from those of purely differential reporting incentives.
In general we find that defense contractors consistently report lower levels of long-term
debt and higher levels of operating leases than their non-government reporting counterparts.
These findings hold in the presence of numerous additional controls for micro and macro-
economic effects, additional U.S. GAAP financial reporting and U.S tax reporting incentives.
We also note that decreases in long-term debt and increases in the use of operating leases are
3
positively correlated with defense firms vis-a-vis non-government companies. Finally, we find
no evidence of a structural shift in these associations during or after the post-9/11 reporting
periods.
The study of government contractors provides a natural experiment regarding the
potential impact of significant differences in reporting, governance, and contracting
environments on specific accounting, tax, and strategic business decisions. Publicly-traded firms
that contract with the U.S. government are subject to both additional audits (conducted by the
Defense Contract Audit Agency - DCAA) and additional reporting requirements (i.e., the Federal
Accounting and Reporting (FAR) guidelines). In some cases, these firms are also subject to
different tax reporting rules—e.g., such as the calculation of taxable profits from government
contracts and the disallowance of accelerated depreciation for assets leased to the government.
Finally, recognized revenue streams derived from federal contracts are often generated by adding
a standard, government-mandated mark-up of allowable costs, rather than by a general
equilibrium market price. Taken together, these points suggest that the government contracting
context provides many promising empirical studies.
We choose to examine how the different reporting incentives may affect asset financing
as the primary focus of research for several reasons. First, the federal government issues clear
public guidance regarding its standards for the reporting of these transactions and for the
recoverability of these costs as part of the contractor’s revenues. Second, the federal government
also issues asset acquisition and disposal tax reporting requirements specifically for contractors.1
Third, both the Financial Accounting Standards Board (FASB) and Congress have made lease
transactions a key area for reforms in their respective regulatory regimes – generally accepted
accounting principles (GAAP) for the former, and the U.S. tax code in the latter case. Fourth, an 1 These guidelines and manuals are available to the public on www.dcaa.mil.
4
extensive body of empirical research exists to suggest initial research designs and hypotheses
tests and these studies indicate that the reporting and contracting environments of the lessee firm
matters to the financing decision outcome. Finally, the extant accounting and tax literatures have
yet to examine this setting which has many rich and unique reporting, accounting choice, and
auditing attributes.
The findings in this study have a number of contributions. First, we exploit this
unique setting whereby there is an acute difference in financial reporting environments across
two groups of firms. In this context, we are able to show that DCAA financial reporting
incentives play a significant role in how firms structure their transactions, even in the presence of
extensive controls for other reporting regime requirements/incentives and various micro and
macro economic factors. U.S. Second, this study draws attention to a little-studied group of
firms in the accounting literature, government defense contractors, and leverages the political
costs faced by these firms to increase the power of our tests. Since these firms are under heavy
scrutiny, we expect that the political costs incentives they face to bias downwards our results.
This study also contributes to the extant leasing literature which continues to document
ambiguous findings for the tax incentive in the asset financing decision (see Ryan et al., 2001 for
a review). Finally, our results may have general policy setting implications. We demonstrate that
government defense contractors systematically make asset financing decisions that minimize
their unreimbursable expenses and maximizes their reimbursable expenses. This would suggest
that DCAA reporting guidelines influence the way firms structure their transactions in order to
maximize their income. Although this is economically rational for the individual firm, it may
not lead to the transaction that is overall the most efficient use of an economy’s or total
taxpayers’ resources.
5
The rest of the paper proceeds as follows. Section one provides basic background
information on lease-types and on the reporting context of government contractors.
Section two briefly reviews the existing research on these two topics. In section three, we
describe the testable hypotheses and model. Section four describes the sample and the research
design. Results are presented in section five, followed by conclusions and suggestions for future
research.
II. Background
Financing Decisions
To further understand the impact of the various reporting regimes on a sample firm, see
Figure 1. The differences in financial reporting incentives between government and non-
government contractors as they relate to the asset financing decision are as follows:2
(1) Debt
a. For basic GAAP purposes, the balance sheet and income statement effects are
the same, because government contractors are required to follow U.S. GAAP
with respect to their basic financial statement reporting.
b. For basic DCAA purposes, the interest on the debt utilized to purchase the
asset is defined by law as a non-recoverable or non-allowable cost. Since
revenue is often determined on a “cost-plus” basis for government contracts
(i.e., the government is billed for allowable costs incurred by the defense firm
plus a standard mark-up), government contractors may be less likely than non-
2 Synthetic leases area fourth type of financing choice not included in the current study. They utilize a slight difference between the tax and financial reporting definitions of an operating lease to allow the creation a lease that simultaneously provides optimal tax and financial treatment for the lessee. However, since synthetic leases provide for an operating lease treatment for financial reporting purposes, the rent expense recognized in the income statement is an allowable expense for government contractors. This suggests that government contractors may be more likely to utilize synthetic leases than non-government contractors. Due to data limitations we do not study synthetic leases in detail at this time.
6
government contractors to utilize debt because interest is not reimbursed with
a mark-up and therefore using debt would lower GAAP reported revenues,
ceteris paribus). Depreciation from owning the asset is reimbursable or
recoverable from the government; however, a firm would not be reimbursed
for any bonus or accelerated amounts used for tax purposes.
c. For basic U.S. tax purposes, there are no significant differences between
defense and non-defense firms. The key tax differences in incentives between
government and non-government contractors lie with the treatment of
depreciation only if the firm is the lessor.3 However, financing theory
indicates that high marginal tax rate firms are expected to value the interest
and depreciation tax shields, and we should expect the use of debt to be
increasing in a firm’s MTR (Fama and French, 1998).
d. A testable hypothesis that may be derived from these differential reporting
effects is that when we control for GAAP and tax reporting status, we expect
government firms to use less debt financing than non-government firms.
(2) Capital leases
a. For basic GAAP purposes, capital leases are similar to debt in all aspects,
except that the firm often will not end up owning the asset. Capital leases are
defined for DCAA purposes uses GAAP standards.
b. For DCAA purposes, the portion of the lease payment that is implicitly
interest is not recoverable from the federal government, similar to explicit
interest above. Again, since interest is not an allowable cost and since
3 In general, if the asset is leased to the government, then the bonus or accelerated depreciation provisions in the tax code are not permitted accessible to government contractors. Non-government contractors can apply bonus or accelerated depreciation on their assets if they are the lessor.
7
depreciation recoverability is limited to the amount reported on the financial
statements, government contractors are less likely to utilize capital leases than
non-government contractors. A testable hypothesis is that when we control
for GAAP and tax reporting status, we expect government firms to use less
capital lease financing than non-government firms.
c. For U.S. tax purposes, there are again no significant differences between
defense and non-defense contractors. However, financing theory indicates
that high marginal tax rate firms are expected to value the interest and
depreciation tax shields also associated with capial leases. We should also
expect the use of capital leases to be increasing in a firm’s MTR (Fama and
French, 1998).
d. A testable hypothesis that may be derived from these differential reporting
effects is that when we control for GAAP and tax reporting status, we expect
government firms to use less capital lease financing than non-government
firms.
(3) Operating leases
a. For basic GAAP purposes, there is no direct balance sheet effect and the rent
expense reduces EBITDA for both government and non-government
contractors;
b. For DCAA purposes, rent expense is an allowable cost for government
contractors and would be allowed in defense firms’ recognized revenue as a
marked-up cost.
8
c. For basic U.S. tax purposes, there are no significant differences between
defense and non-defense firms. However, financing theory indicates that low
marginal tax rate firms are expected not to value the interest and depreciation
tax shields. Fama and French (1998) and industry experts note that these firms
have the opportunity to essentially sell the tax shields back to the lessor in the
form of lower lease payments. Hence, we should expect the use of operating
leases to be decreasing in a firm’s MTR (Fama and French, 1998).
d. A testable hypothesis that may be derived from these differential reporting
effects is that when we control for GAAP and tax reporting status, we expect
government firms to use less debt financing than non-government firms.
Lease-type decisions may be best initially understood in the context of a lessee’s choice
between a pure capital lease versus a pure operating lease. Pure capital leases provide the lessee
with the associated tax deductions of asset ownership and thereby lower the lessee firms’ taxable
income. However, these tax benefits may be offset by the associated financial reporting
disadvantage of requiring full on-balance-sheet recognition of a significant lease-related liability
and asset. In contrast, operating leases provide off-balance-sheet reporting, but the lessee is
normally expected to forgo the related accelerated depreciation and interest tax deductions in
exchange for relatively lower rent payments. Operating leases as defined in U.S. GAAP are also
known as “true leases” in the tax code.4 These leases are typical lease arrangements whereby the
4 According to the bright line standards of SFAS No. 13, operating lease treatment on the lessee’s financial statements is appropriate when all four of the following requirements are met (if any one of the guidelines are violated, the lease automatically qualifies for capital lease reporting):
• There is no bargain purchase option in the contract • There is no transfer of ownership title to the lessee in the contract. • The lease term is less than 75% of the economic life of the asset.
9
lessee uses an asset for a relatively short duration and pays only for this limited use in the form
of rent. 5 The advantages of this lease-type for financial reporting purposes are clear, however a
tax deduction for rent expense is normally lower than the sum of the asset-ownership deductions
available is a capital lease were used instead. A pricing equilibrium between the lessor and the
lessee does not exist because extreme competition in the leasing market forces lessors to price
capital leases lower than the lessor’s expected value of their foregone asset ownership tax
deductions.
A third type of lease to note is a synthetic lease. These hybrid leases were introduced
shortly after the SFAS No. 13 lease reporting standard became effective in 1975. Synthetic
leases utilize a slight difference between the tax and financial reporting definitions of an
operating lease to allow the creation of a lease that simultaneously provides optimal tax and
financial reporting treatment for the lease. Due to data limitations at this time, these lease are
beyond the scope of this study. For a detailed description, see either Altamaro (2004) or
Morsfield (2004).
• The present value of the minimum rent payments and guaranteed residual value is less than 90% of the asset’s current fair market value (“90% FMV rule”), hereafter.
5 The tax reporting requirements differentiate a “true” lease from what is essentially considered a conditional sale of the asset from the lessor to the lessee. There are no explicit lease-type guidelines in the tax code itself, rather their delineation is expressed in various rulings and the ultimate definition of a “true lease” is definitively established by the six guidelines listed in Section 4, Rev. Proc. 75-21:
• The lessor must have at least 20% of the total acquisition cost of the asset invested and at risk. • The leased asset’s estimated residual value must be equal to at least 20% of the initial costs. • The remaining life of the asset at the end of the lease term must be equal to the longer of one year or 20% of its
original estimated useful life. • There is no bargain purchase option. • The lessee may not loan to nor guarantee the loans of the lessor to purchase the asset. • The lessor must be able to demonstrate an expectation of profits to be derived from the lease.
While the actual tax reporting requirements are slightly different from those found in SFAS No. 13, in most cases a plain vanilla operating lease on a lessee’s financial statements is also a “true lease” and not a conditional sale according to the tax authorities. Both the tax and the financial reporting requirements at present attempt to define an operating lease by determining which party to the lease contract retains most of the risk and rewards of the asset in question. In the case of a plain vanilla operating or “true” lease, the requirements establish that the majority of risks and rewards of the asset remain with the lessor.
10
Government Contractors
Government contractor financing decisions are best understood in the context of their
unique and very structured reporting and revenue generation requirements. To generate federal
government-related revenues, these contractors must collect and report allowable project
expenses plus a standard government-mandated mark-up. Expenses that are allowable are
mandated and defined in detail by federal contractor manuals (throughout the rest of the paper,
we will refer to these as the ‘DCAA reporting guidelines’). These expenses are closely audited
for their allowability, reasonableness and correct allocation to projects by a specially-designated
independent auditor (the Defense Contract Audit Agency – DCAA) paid for by taxpayers, not by
the government contractors. Since the government contractors in our study are also publicly-
traded, their financial reports are also subject to additional standard audits required by the
Securities and Exchange Commission (SEC).
Most government contracts provide revenue on a type of “cost-plus” basis. That is, the
government is billed for allowable costs plus a standard mark-up. In general, then, managers of
government contracting firms have incentives to maximize allowable and recoverable expenses
and to minimize those that do not qualify as such. Hence, given a choice between specific
financing methods, they would be expected to choose the one that generates allowable expenses
over a similar purpose transaction that does not. Relatedly, certain tax reporting requirements
are directed at government contractors. For example, government contractors are generally not
allowed to use accelerated depreciation methods for tax reporting for those assets acquired or
leased for use in a government project. The managers in this instance also have tax reporting
incentives to choose an operating lease over an asset purchased with debt (or with cash).
11
Publicly-listed government contractors must comply with the government contract
standards in addition to GAAP standards. The DCAA audits the financial statement information
according to generally accepted auditing standards (GAAS) mandated by the American institute
of Certified Public Accountants (AICPA) and the financial reporting standards defined by
GAAP. The DCAA also audits cost reporting, allocation, reasonability, and recoverability using
government-specific guidelines, originally mandated by Congress (a.k.a, “the Yellow Book”).
With respect to financing costs, of particular note is that interest on contractors’
borrowings is simply not a recoupable cost, while depreciation and rent are recoverable if
reasonable in amount and if allocated correctly. The DCAA places an additional restriction on
the deprecation allowance by limiting it to the amount reported on the financial statements
regardless of the method used for tax purposes. This implies, for example, if a contractor utilizes
the accelerated and bonus depreciation schedules permitted under current tax law, yet uses the
straight-line method for its financial statements, that the recoverable cost on an invoice to the
government is capped at the reported straight-line calculation. Since, for financial reporting
purposes, synthetic leases are treated as operating leases, the DCAA allows only the rent expense
component of the lease to be recovered.
II. Literature Review and Hypothesis Development
Prior research into the financing decisions of publicly-traded firms tends to focus on the
tax incentives or the financial reporting concerns of the lessee. Little or no research examines
how being a government contractor, subject to added and substantial reporting requirements,
affects these choices.
12
Taxes and Financing
Modigliani and Miller (1963) argue that managers are indifferent between lease and debt
choices as long as there are no taxes. In the presence of taxes, however, some managers have
incentives to use an operating lease rather than to buy an asset with debt. Firms that cannot use
the tax shields of ownership are expected to use operating leases to finance assets from high
marginal tax rate lessors that can use these tax shields. Lessee firms essentially sell their asset-
related tax shields to the lessor in the form of lower lease payments (Myers, Dill, and Batista
(1976); Smith and Wakeman (1985); also see Appendix A for examples of lessor disclosures that
confirm this pricing approach).
Capital leases are considered in theory to be equivalent to debt-financed asset purchases.
That is, high marginal tax rate lessees should prefer capital leases over operating leases so that
they can claim tax deductions for interest and depreciation (Graham, et al., (1998). These
propositions suggest the testable hypothesis that high marginal tax rate lessee firms are more
choose capital leases and debt, assuming that they have no financial reporting constraints.
Operating leases should be chosen by low marginal tax rate lessee firms. Synthetic leases should
be chosen by high MTR firms that wish to keep their leases from on-balance-sheet financial
reporting.
Empirical results in this research area are mixed (see Graham, et al (1998) for a
comprehensive review). Graham et al (1998) propose that this ambiguity about the association
between tax status and financing is due the fact that the choice and tax status are endogenous.
When they provide a tax measure eliminating endogeneity, they find the expected relation
between tax and debt, and between tax and operating leases. However, they find no significant
relation between tax and capital lease choice.
13
Morsfield (2004) utilizes a small, privately-acquired sample of 144 actual lease-type
choice contracts to examine the role of tax versus financial reporting incentives. This sample
allows the researcher to identify the actual lease decision rather than be limited to the public data
available—i.e., the cumulative financing measures per firm. An important contribution of this
study is that the sample includes clearly identifiable synthetic leases. Synthetic leases are treated
as operating leases for financial reporting purposes and as asset purchases with debt for tax
reporting purposes. This lease-type was not previously tested in the literature due to difficulties
in identifying these leases from publicly available data.
When using Graham, et al’s (1998) research design and MTR and regressing the actual
lease-type (capital, operating, or synthetic), Morsfield (2004) finds the predicted relation of tax
status on all types of lease decisions. This is the first study to document consistent theoretical
predictions across all types of financing. A key contribution of this research is providing
evidence that strongly suggests that when synthetic leases are not accurately identified in the
sample, ambiguous findings may result, simply due to this measurement error in the public data.
These tax studies control for financial reporting issues, but do not focus on them explicitly, nor
do incorporate the government contracting environment.
Financial Reporting and Governance
Ryan, et al (2001) and Lipe (2001) provide an extensive review of the existing lease and
financial reporting literature. They note the general conclusion over time is that for risk
assessment and valuation purposes, the equity market treats operating leases similar to capital
leases and debt. More recently, Altamuro (2004) conducts an examination of the synthetic lease
choice in particular and finds that managers with bonus plans driven by net income are more
14
likely to use synthetic leases. She also incorporates the strength of the board of directors and
notes that strong boards are able to constrain managers from using synthetic leases for this
purpose. She holds constant the tax incentives, but does not examine government contracting or
reporting incentives. She does note however that alternative reporting and contracting incentives
such as managers’ bonus plans has a direct effect not just on the synthetic lease-type decision.
Government Contracting
Little recent empirical research exists in the accounting literature which captures the
potential impact of the federal government’s reporting, tax, contracting, and governance
concerns on a supplier firms’ own reporting, contracting, and governance choices. McGowan
and Vendrzyk (2002) update prior research regarding excess GAAP profitability and overhead
cost-shifting related to defense contract reporting requirements observed in the late 1980’s and
1990’s. However, we are not aware of any studies that examine how actual transaction choices
are affected by differences in or conflicting incentives of the GAAP versus government reporting
regimes.
In the public policy literature, the lease-versus-buy decision is only examined from the
perspective of the federal government as the potential lessee or buyer. The existing literature
appears to conclude that leasing costs the government more than purchasing, and that the
government contractors capture these rents from the taxpayers. These studies contribute to our
understanding of the government contracting environment, as well as of the macroeconomic
issues that surround contracting between the federal government and the private business sector
(e.g., Masse, Hanrahan, and Kushner, 1987, 1988; Blose and Martin,1989)
15
We develop hypotheses based on extant leasing theory and on predictions derived from
basic knowledge of the federal allowable cost principles guidelines noted earlier. The predicted
associations assume the tax, financial reporting, and governance incentives are either similar
across the two comparison samples—government contractors (GOV) versus non-government
contractors (Non-GOV)—or are adequately controlled for in a multivariate research setting
(Refer to Figure 1, as necessary).
III. Hypotheses and Empirical Model
Figure 1 provides a summary of the reporting differences that a typical defense contractor
may face when making an asset-related financing choice. This figure and the background
information provided earlier in the paper also serve as a basis for forming our testable
hypotheses. Firms financing their asset acquisition with debt will report an asset and the
corresponding liability on the balance sheet. For tax reporting purposes, these firms can utilize
larger tax deductions associated with the acquisition in the form of interest and accelerated
depreciation. However, the DCAA reporting guidelines set a limit on the reimbursement of the
depreciation for the government contractor from the government to the amount reported in the
financial statements. In addition, interest expense is not a recoupable cost for the government
contractor. Hence, we would expect that government contractors have less incentive than non-
government contractors to utilize debt for asset financing. The following hypothesis is proposed:
Debt Hypothesis
H1: If government contractors are constrained by the DCAA reporting guidelines in
their ability to be reimbursed for debt-related costs, we expect government contractors to utilize debt financing less than non-government contractors, ceteris paribus.
16
Firms choosing to finance their asset acquisitions using operating leases do not report an
asset or liability on the balance sheet. Rent expense associated to the lease payment is
recognized, thereby reducing EBITDA. Relative to debt and capital leases, operating leases lead
to lower tax deductions because only the rent expense is a deductible whereas both interest and
depreciation expenses are deductibles for debt and capital lease transactions. However, the
DCAA reporting guidelines permit rent expense to be a reimbursable cost, thereby providing an
additional incentive for government contractors to utilize operating leases to finance asset
acquisitions. This leads to the following proposition:
Operating Lease Hypothesis
H2: If government contractors are allowed by the DCAA reporting guidelines to either utilize operating leases or to be reimbursed for operating lease-related costs, we expect that government contractors to utilize operating lease financing more than non-government contractors, ceteris paribus.
When firms finance their asset acquisitions through capital-lease type arrangements, then
the asset and associated liability appears on the balance sheet of the firm. Similar to debt
transactions, the lessee firm is able to utilize larger tax deductions in their tax reporting. Similar
to debt transactions, government contractors are constrained by the DCAA reporting guidelines
in their ability to be reimbursed for the capital lease-related costs. The following hypothesis is
proposed:
Capital Lease Hypothesis
17
H3: If government contractors are constrained by the DCAA reporting guidelines in their ability to be reimbursed for capital lease-related costs, we expect government contractors to utilize capital lease financing less than non-government contractors, ceteris paribus.
In this study, we test our hypotheses using a modified version of the Graham, et al.
(1999) model and research design. Balance sheet data from COMPUSTAT to measure the
accumulated amount of capital leases, operating leases and debt-financed asset purchases.
FINANCING = as described in detail below, various debt or lease-type amounts or
choices: H1: Debt = LT Debt/Fixed Claims (DEBT).
H2: Capital Lease = Capital Lease/Fixed Claims (CAPLEASE). H3: Operating Lease = Operating Lease/Fixed Claims (OPLEASE).
ZSCORE = modified Altman (1968) Z-score as per Graham et al. (1998) MTB = Market-to-book value ratio. This variable proxies for the lessee’s
investment opportunities. COLLATERAL = Net property, plant, and equipment scaled by total assets. This is a
proxy of fixed asset usage or available collateral. GOV= Dummy equal to one if the firm is a defense contractor; zero
otherwise.
GOV is the key explanatory variable of interest. We expect the coefficient on GOV to be
negative, negative and positive when the dependent variable in the regression is DEBT,
18
CAPLEASE and OPLEASE, respectively. MTR is another important explanatory variable and
theory predicts it to be positively associated with the propensity to use capital leases and debt
and negatively associated with the propensity to use operating leases.
A measure of financial distress is ZSCORE. Since this variable captures firms with
ongoing financial difficulty it should be positively associated with the propensity to lease in
general. Its association with debt depends on whether the firm had accumulated a significant
amount of long-term debt that is having trouble repaying before the year being studied. In this
case, the sign on this variable would be negative (i.e., a lower ZSCORE reflects more financial
trouble), However, if the firm has been trying unsuccessfully to raise debt financing due to its
financial difficulties, then the sign would be positive. We expect COLLATERAL to be positively
related to the use of debt and capital leases, since both these financing choices may both require
and add to the net measure of property, plant, and equipment. For a firm with a propensity to
lease, we expect COLLATERAL to be negatively associated with the decision, since collateral is
neither required or created by this type of transaction.
IV. Sample and Research Method
The defense contractor sample is drawn firstly from the top 100 government defense
contractor list as identified from the list published by the Department of Defense (DoD) for the
2000 Congressional Budget year. The top 100 list is ranked by total contract revenue awarded
and are publicly available on the DoD’s procurement website. We include in our sample these
publicly-listed large defense contractor firms because they face intense scrutiny and heightened
political costs. The use of this group of contractors increases the power of our tests because the
political costs faced by these firms would bias against finding results consistent with our
19
predictions. In addition, we also include in our sample other defense contractors not in the top
100 list by conducting a search on ‘defense contract’ in the SEC Edgar database that have the
necessary Compstat data. Our final sample of defense contractors is 103 firms. We match these
firms with a sample of 103 non-government companies based on year 2000 beginning
debt/equity and industry (i.e., two-digit SIC). Our period of interest is 2000-2002.
We perform a series of separate regressions whereby the dependent variable is DEBT,
CAPLEASE or OPLEASE on a sample of government defense contractors matched to non-
government contractors on industry and debt-to-equity. The mean total assets for government
contractors is approximately $14 billion (not tabulated) compared to the mean total assets for
non-government contractors of $4 billion. This is difference is significant (p = 0.01). In un-
tabulated results, we observe that defense contractors tend to cluster in certain industries –
transportation equipment (2-digit SIC code 37), business services (2-digit SIC code 73) and
electronic and electrical equipment (2-digit SIC code 36).
V. Results
Descriptive Statistics and Univariate Results
Table 1 presents the descriptive statistics and univariate results for the sample of defense
contractors matched to non-defense contractors on industry and debt-to-equity (based on fiscal
year-end 1999 Compustat data). We observe that there is no difference in the marginal tax rates
of defense and non-defense contractors (p = 0.6190), market-to-book (p = 0.8416) and the level
of financial distress (p = 0.1898). Defense contractors are significantly larger in firm size (p =
0.000) but own significantly lower collateralized fixed assets (p = 0.0610). Our univariate
results suggest that defense and non-defense contractors make similar financing decision except
20
for capital leases whereby non-defense contractors are more likely to use capital leases than
defense contractors (p = 0.0379). However, these results do not control for other factors that
may affect financing decisions between defense and non-defense contractors.
Multivariate Results
Next, we examine the financing choice between defense and non-defense contractors,
using fixed effects ordinary least square regressions to control for a number of other factors that
may affect this financing decision. Table 2 presents the results of the regression of financing
choice (cumulative debt, cumulative capital lease or cumulative operating lease – i.e., ‘levels’
regression) on a dummy variable for defense contractors, Defense, and various other control
variables for a sample matched on two-digit SIC code and (beginning) debt-to-equity.6 We also
re-run the regression on a separate sample matched on two-digit SIC code, beginning debt-to-
equity, and marginal tax rate (to control for the tax incentives). In addition, we also match on
two-digit SIC code, beginning debt-to-equity and revenue (to control for the financial reporting
incentive).
For the full sample of defense contractors (n=103) matched to non-defense contractors on
two-digit SIC code and beginning debt-to-equity, consistent with H1 we expect Defense to be
significantly negative for the regressions where LT DEBT is the dependent variable, significantly
positive where Opg Leases is the dependent variable (H2) and significantly negative where Cap
Leases is the dependent variable (H3). Consistent with H1, H2 and H3, Defense is significantly
negative (p = 0.000), negative (p = 0.076) and positive (p = 0.000), respectively. Overall, the
results suggest that defense contractors are less likely to use debt and capital leases to finance
6 In addition, we also regress changes in debt, capital lease or operating lease (i.e., ‘changes’ regression) in order to capture the financing decision for a given year on a dummy variable for defense contractors and the various control variables for a sample matched on two-digit SIC code, beginning debt-to-equity, marginal tax rate and revenue. Our results (not tabulated) show that there is no difference in financing choice between defense and non-defense contractors.
21
their asset acquisitions than non-defense contractors, after controlling for MTR and other factors
correlated with the financing decision. Our results also suggest that defense contractors use
operating leases more so than non-defense contractors to finance their asset acquisition.
Consistent with our expectations, MTR is significantly positive in the regression where
LT Debt is the dependent variable suggesting that firms with higher marginal tax-rates are more
likely to use debt to finance their asset acquisition. In addition, our results are consistent with
our priors that firms with higher marginal tax rates are more likely to utilize operating leases to
finance their asset acquisitions (p = 0.000). However, our results for MTR are mixed in relation
to capital leases. This is consistent with the ambiguous findings in the finance literature
documenting the association (or lack thereof) between MTR and capital lease financing.
ZSCORE is not significantly associated with financing choice as with. Consistent with our
priors, COLLATERAL is significantly negative in the regression where LT Debt is the dependent
variable (p = 0.086), COLLATERAL is not significant where Cap Leases is the dependent
variable (p = 0.129) and COLLATERAL is significantly negative where Opg Leases is the
dependent variable (p = 0.001). This is consistent with the expectation that the propensity of
firms to lease is associated with the financing decision. Our results are consistent when we also
control for the financial reporting incentive (income statement effect) (EBITDA) in the
regression. The adjusted-R2s across the various regressions are comparable to prior leasing
studies.
Overall, our results so far are consistent with expectations that defense contractors are
more likely than non-defense contractors to use operating leases and less likely than non-defense
contractors to use debt and capital leases to finance their asset acquisitions. These results
suggest that the non-reimbursable component of interest expenses for defense contractors make
22
debt and capital lease financing less attractive. Our main results show that after we control for
tax and financial reporting incentives and the ex-ante financing decisions, defense contractors
and non-defense contractors make different financing decisions. We suggest that the differential
financial reporting incentives between government contractors (driven by DCAA guidelines and
GAAP) and non-government contractors (driven only by GAAP) provides an explanation for this
difference.
To provide further insight to our main result and to increase the power of our tests, we re-
run our levels regression and increase the power of our tests by matching on two-digit SIC code,
beginning debt-to-equity, MTR and also commercial revenue. We employ the one-to-one
nearest neighborhood matching method using the estimated propensity score for each firm. The
propensity score (the conditional treatment probability) is estimated using the two-digit SIC
code, debt-to-equity, marginal tax rate and the commercial revenue. We do this in order to
control further for GAAP financial reporting incentives in the financing decision, and thus any
observed significance on Defense can be clearly attributed to the DCAA reporting requirements.
We match on commercial revenue (i.e., non-government revenues) in order to provide another
control for the financial reporting incentives between the defense and non-defense contractors.
Table 3 provides the results for this regression. Consistent with H1 and H2, we observe
that Defense continues to be negatively (p = 0.009) and positively (p = 0.007) associated with LT
Debt and Opg Leases, respectively. H3 is not supported as in prior tests. We generally conclude
that, as reported in Morsfield (2004) the measurement error in COMPUSTAT data regarding the
all effects of synthetic leases, may be confounding the capital lease results. ZSCORE and MTB
are not significantly associated with the various financing choices. Collateral is negatively and
positively associated with the operating and capital lease choice decision, respectively. Size is
23
positively associated with both the debt and capital lease choice decisions and negatively
associated with the operating lease choice decision. Overall, we increase the power of our tests
by controlling for the ex-ante financing choice (debt-to-equity, the marginal tax rate (i.e., the tax
incentive), industry and commercial revenue (i.e., the financial reporting incentive), such that
any observable difference between defense and non-defense contractors can be attributed to the
additional reporting requirements of the DCAA faced by defense contractors. Our results
regarding the debt and operating lease choice decisions are consistent with our hypotheses and
the main results reported, suggesting that the additional reporting requirements by the DCAA
faced by defense contractors influence their asset financing choices. This result has not been
previously documented in the extant literature.
VI. Conclusions and Future Research
We examine whether differences in financial reporting incentives between defense and
non-defense government contractors influence the asset financing decision. Very recent media
and political attentions have been drawn to the financial reporting indiscrepancies evident in
government contractor firms, in addition to their participation in questionable transaction-
structuring. This study attempts to provide some empirical evidence to shed light to the
discussions surrounding the impact of government contractors’ financial reporting and tax
incentives on the structure of their asset financing.
Overall, the results consistently document that the additional financial reporting
guidelines outlined by the Defense Contract Audit Agency explains the differential asset
financing choices between government contractors and non-government contractors. We also
provide consistent evidence that tax incentives play a significant role in the asset financing
24
decision. Our main results are robust to controls for industry and the firm’s ex-ante financing
decision. In additional analyses, we increase the power of our tests by controlling for the
additional GAAP financial reporting incentives and the firm’s tax reporting incentives such that
any difference observed in the asset financing choice between defense and non-defense
contractors can be attributed to the DCAA reporting requirements. Again, we find that defense
contractors make financing decisions that are different to non-defense contractors in a manner
consistent with the additional DCAA reporting incentives they face.
This study contributes in the following ways. First, we are able to exploit a unique
setting to demonstrate that differences in financial reporting environments influence how firms
structure their transactions. One implication of this is that it draws attention to the behavioral
impact on firms by financial reporting requirements. Second, we examine a group of firms that
have received relatively little attention in the literature despite heavy scrutiny by the media and
political players. Third, we systematically show that financial reporting incentives and not tax
incentives influence the asset financing decision. Finally, our results have implications for
government policy setting.
25
REFERENCES
Altamuro, Jennifer L.M. The Economic, Financial Accounting, and Corporate Governance
Determinants of Synthetic Leasing. Working Paper 2004. The Massachusetts Institute of Technology.
Barclay, M., and C. Smith, 1995, The priority structure of corporate liabilities, Journal of Finance 50, 899-917. Blose, L. and J. Martin, 1989, Federal Government Leasing: Costs, Incentives, and Effects,
Public Budgeting & Finance, Summer, 66-75. Callahan, J, 1981, The Lease versus Purchase Decision in the Public Sector, National Tax
Journal 34, 235-240. Financial Accounting Standards Board (FASB), 1996, Accounting for leases: a new approach,
Special Report (Stamford, CT). General Accounting Office. 2002. International taxation – Information on Federal Contractors
with Offshore Subsidiaries. United States General Accounting Office Report to Congressional Requesters (GAO-04-293).
Graham, J., 1996a, Debt and the marginal tax rate, Journal of Financial Economics 41, 41-73. ________, T. Lemmon, and J. Schallheim, 1998, Debt, lease and taxes and the endogeneity of
corporate tax status, Journal of Finance 53, 131-162. King, N. K., Jr. and G. R. Simpson, 2004, Pentagon asks Justice to Join Halliburton Probe; Move
Suggests Investigators See Grounds for Penalties Over Billing for Fuel in Iraq, Wall Street Journal, March 11, 2004, p. A.1.
Lipe, R. 2001, Lease Accounting Research and the G4+1 Proposal, Accounting Horizons,
Sep(15), 299-311. Masse, Hanrahan, and Kushner, 1987, The Lease versus Borrow Decision from the Public Sector
Perspective, National Tax Journal 40, 271-274. Masse, Hanrahan, and Kushner, 1988, The Effect of Changes in Tax Legislation on the
Purchase/Lease Decision in the Public Sector Perspective, National Tax Journal 41, 123-130.
McGowan, A., and V. Vendrzyk., 2002, The Relation between Cost Shifting and Segment
Profitability in the Defense-Contracting Industry, The Accounting Review, Oct(77), 949-
26
969. Modigliani, F., and M. Miller, 1963, Corporate income taxes and the cost of capital: a correction,
American Economic Review 53, 433-443. Morsfield, 2004, A new look at the effect of lessee tax rates on the lease-type decision: The role
of synthetic leases. Working Paper, New York University. Myers, S., D. Dill, and A. Bautista, 1976, Valuation of financial lease contracts, Journal of
Schipper, Catherine Schrand, Douglas Skinner, and Linda Vincent. Evaluation of the Lease Accounting Proposed in G4+1 Special Report. Accounting Horizons (2001) 15: 289-299.
Smith, C., 1993, A perspective on accounting-based debt covenant violations, The Accounting
Review 68, 289-303. _______, and L. Wakeman, 1985, Determinants of corporate leasing policy, Journal of Finance
40, 895-908.
1
Figure 1 Financing Decisions Descriptions and Hypotheses: Financial & Tax Reporting Effects for Owner/Lessee
Ceteris Paribus with respect to MTR and GAAP reporting concerns: H1: We expect Defense firms to use less debt than non-defense (NON) firms, because all “costs of borrowing” will not be reimbursed by the federal government. H2: We expect Defense firms to use more operating leases than NON firms, because operating rent expense is a fully reimbursable cost. H3: We expect Defense firms to use less capital leases than NON firms, because all costs of borrowing, including, the implicit interest, in capital leases as defined by SFAS No. 13 will not be reimbursed in federal contract revenues.
Financing Decision
LT Debt
(H1)
Operating Lease
(H2)
Capital Lease
(H3) Financial Reporting(GAAP) -Applies to both defense & non-defense contractors
• On Balance Sheet, • Will own the asset, • Depreciation and Interest
expense, • Stronger EBITDA, • Lowers current tax
expense, • Increases deferred tax
expense, • DTL for depreciation.
• Off Balance Sheet, • Will not own, • Rent expense, • Reduces EBITDA, • No depreciation and
interest expense, • No related deferred tax
expense, unless cash flows vary from recognized expense.
• DTL only if cash flows issue above
• On Balance Sheet, • Will not own, • Depreciation and Interest
Expense, • Stronger EBITDA, • Lowers current tax
expense, • Increases deferred tax
expense, • DTL for depreciation.
Financial Reporting (Defense) -Applies to only defense contractors
• Interest not recoverable, • Depreciation recovery is
allowed, but limited to GAAP amount.
Rent is recoverable, if reasonable.
• Interest not recoverable • Depreciation recovery is
allowed, but limited to GAAP amount.
Tax Reporting (TAX) -Applies to both defense & non-defense contractors
• Larger Tax Deductions:
• Interest, MACRS, & Bonus Deprn.
• Smaller Tax Deductions:
• Rent only.
• Larger Tax Deductions:
• Interest, MACRS, & Bonus Deprn.
2
Table 1 – Descriptive Statistics and Variable Definitions
Panel A—Descriptives Matched Sample—on Industry and Debt/Equity at FYE 1999 T-test Defense Contractors
(n = 103) Non-Defense Contractors
(n = 103) p-value
General Marginal Tax Rate (MTR) per Graham (1999)
0.2730
0.2785
0.6190
Total Assets (ln) 7.0732 5.6674 0.0000 Collateral 0.1804 0.2042 0.0610 Market-to-Book (MTB) 4.1515 4.0365 0.8416 Z-Score 3.8760 -13.9219 0.1898 Financing Decisions Long Term Debt/FC 0.4817 0.466 0.5961 Operating Leases/FC 0.5029 0.5093 0.8255 Capital Leases/FC 0.0153 0.0243 0.0379
1
Table 1 – Descriptive Statistics and Variable Definitions, continued
Panel B Variable Definitions General Defense Equal to one if firm top 100 government contractor in 2000 zero otherwise. Sales, Net Data12 EBITDA Data 13-Data15-Data16 Marginal Tax Rate (MTR) Graham’s (1996) before-financing tax rate, courtesy of John Graham. Total Assets ln(Data6) Collateral/FC Data8/Fixed Claims (see below) PV Operating Leases Data47 +PV((Data95/5) at 5% average rate for our sample) Fixed Claims (FC) Data9+Data84+PV OperatingLeases Market Value (MV) Data6-Data60+((D25*Data199)+PVOpgLeases)) Market-to-Book (MTB) MV/(Data60+PVOpgLeases) Z-Score Modified Altman’s Z-score, per Graham, et al (1998) Debt-to-equity Data9/Data60 Financing Decisions LT Debt Data9 – Data84 Capital Leases Data84 Operating Rent Data47 PV Operating Leases Data47 +PV((Data5/5) at 5% average rate for our sample) Fixed Claims (FC) Data9+Data84+PVOpgLeases Long Term Debt/FC Calculations per above. Operating Leases/FC Calculations per above. Capital Leases/FC Calculations per above. Synthetic Leases Inadequate data points.
2
Table 2 – OLS Regression of Financing Choice on Government Contractor and
Control Variables—Samples Matched on Industry and Debt/Equity The sample consists of 103 defense contractors from 2000 matched to non-defense contractors as noted below for the years 2000-2002. Using year-fixed effects techniques, we regress various financing decisions on an indicator variable for government contractor and various control variables. See the variable definitions in Table 1, Panel B.
Table 3 – OLS Regression of Financing Choice on Government Contractor and Control Variables--Samples matched on Industry, Debt/Equity, Tax Reporting (MTR), and Financial Reporting (Revenue)
The sample consists of the top 100 government contractors from 2000 matched to non-government contractors as noted below for the years 2000-2002. Using year-fixed effects, we regress various financing decisions on an indicator variable for government contractor and various control variables. See the variable definitions in Table 1, Panel B.
(H1)
(H2)
(H3)
LT Debt
Opg Leases
Cap Leases
Matched on:
Industry, Beginning D/E, MTR &Commercial Revenue
Industry, Beginning D/E, MTR &Commercial Revenue
Industry, Beginning D/E, MTR &Commercial Revenue
Intercept
0.2311 (0.000)
0.1924 (0.000)
0.0335 (0.000)
Z-Score 0.0555 (0.367)
0.0116 (0.549)
0.0293 (0.554)
MTB 0.0061 (0.073)
-0.0023 (0.038)
0.0009 (0.133)
Collateral 0.1202 (0.165)
-0.0479 (0.080)
0.0565 (0.001)
Size (Ln Assets) 0.0567 (0.000)
-0.0128 (0.000)
-0.0041 (0.000)
Defense -0.0716(0.009)
0.02327 (0.007)
-0.0022 (0.669)
N 29 29 29Adj. R2 0.3292 0.2286 0.0669
4
Appendix A: Examples of Lease Descriptions
Example 1: www.LeaseExperts.com
True Lease or Operating Lease Also known as fair market value leases. The most notable feature of this type of lease is that its structure does not contemplate a full payout of the cost of the equipment as is the case in a "Finance" type lease. Two of the common tests are:
The term of the lease is generally not greater than 75% of the equipment's anticipated useful life.
The present value of the lease payments should not exceed 90% of the fair market value of the equipment using the lessee's incremental cost of borrowing.
A significant benefit is that the monthly payments are also less than on a finance type lease (above) or even a bank loan. Typically the lessee either returns the equipment at the conclusion of the lease or may be granted the opportunity to purchase the equipment from the lessor for "the fair market value." Payments under this kind of lease structure are treated (by the I.R.S.) as rental payments and therefore are 100% tax deductible operating expenses. Also, as rental payments, neither the asset nor its corresponding liability need to appear on the company's balance sheet. The lessor retains the right to depreciate the equipment.
Capital Lease / Finance Lease / $1 Buyout May also be referred to as a nominal or ($1) dollar-buyout lease. These leases share the advantage of fixed monthly payments but with the guaranteed option to purchase the equipment for a nominal price at the conclusion of the lease. With this type of lease there is no uncertainty about the value of the equipment at the conclusion of the lease as the buyout terms are generally a part of the initial agreement.
Finance type lease may not qualify under I.R.S. regulations for true lease deductibility [authors’ note: it is then treated as a installment plan purchase with debt (see next Example 2 below)] The lessee is considered the owner of the equipment (unlike an FMV lease) and maintains full control of the residual value.
The lessee can depreciate the equipment.
Lessees records the equipment as an asset and the lease payments as liabilities on their balance sheets.
5
Example 2: Old National Bank website
The two major types of business leases are operating leases and capital leases:
• Operating leases. An operating lease is a lease that often allows the lessee to make smaller periodic payments than with a term loan. Smaller lease payments are the result of the lease not being fully amortized over the lease term. As a result, the leased equipment has a residual value at the end of the lease term. An operating lease requires the lessor to service and maintain the equipment. Costs for these services are added to the lease payments. An operating lease also allows the lessee to cancel before the end of the lease term.
• Capital leases. A capital lease amortizes the lease over the span of the lease term. The lessee receives the tax benefits of a capital lease, such as the tax savings earned on the depreciation expense of the equipment. At the end of the lease term, the lessee owns the equipment or faces a much smaller residual value. Since capital leases are considered equivalent to loans, you are required to show any capital leases on your balance sheet. Lease payments on a capital lease are larger than payments on an operating lease. In fact, capital-lease payments are comparable to loan payments. As a result, a lease-versus-buy analysis is more appropriate when evaluating operating leases.
While equipment leasing and real estate leasing have their respective unique terms (triple net in real estate leasing, for instance), operating leases have similar features, including:
• Lower payments. Since operating leases only partially amortize a lease, monthly payments are generally lower than payments on a term loan, which fully amortizes the loan.
• Residual value. At the end of a lease or loan term, the asset is likely to have some resale value. This value is called the residual, or salvage, value. If the lessee renews the lease, the residual value will be the amount that is renewed.
• Purchase option. Instead of renewing a lease or walking away, a lessee may decide to exercise the option to purchase the equipment at the residual value. A purchase option is a standard feature of an operating lease.
Leasing offers certain advantages over buying. There are certain trade-offs with either decision. With leasing, we discussed the benefit of smaller payments and flexibility that comes with not owning the equipment for any longer than you wish. However, buying (using a loan to supplement any cash down payment) has advantages that come with ownership. As a result, the lease-versus-buy decision can be a complex one. At a minimum, you should use a discounted cash flow (DCF) analysis to show expected cash inflows and outflows of either financing decision.
6
APPENDIX B: List of Defense Contractors
AT&T CORP ACACIA RESEARCH- COMBIMATRIX AEROFLEX INC AFFILIATED COMP SVCS -CL A AIRTRAN HOLDINGS INC ALLIANT TECHSYSTEMS INC ANALEX CORP BAE SYSTEMS PLC -ADR BALL CORP BEARINGPOINT INC BOEING CO CET SERVICES INC CNF INC CACI INTL INC -CL A COMARCO INC COMPUTER SCIENCES CORP CONOCOPHILLIPS CUBIC CORP DRS TECHNOLOGIES INC DIRECTV GROUP INC DYNAMICS RESEARCH CORP EDO CORP ELECTRONIC DATA SYSTEMS CORP ENGINEERED SUPPORT SYSTEMS ESTERLINE TECHNOLOGIES CORP EXXON MOBIL CORP FAIRCHILD CORP -CL A FEDEX CORP FREQUENCY ELECTRONICS INC GP STRATEGIES CP GSE SYSTEMS INC GTSI CORP GENERAL DYNAMICS CORP GENERAL ELECTRIC CO GOODRICH CORP HALIFAX CORP HALLIBURTON CO HARRIS CORP
7
HARSCO CORP HEALTH NET INC HONEYWELL INTERNATIONAL INC HUMANA INC ITT INDUSTRIES INC IDENTIX INC IMAGEWARE SYSTEMS INC INTEGRAL SYSTEMS INC/MD INTL SHIPHOLDING CORP IRVINE SENSORS CORP JMAR TECHNOLOGIES INC JACOBS ENGINEERING GROUP INC JOHNSON CONTROLS INC L-3 COMMUNICATIONS HLDGS INC LOCKHEED MARTIN CORP MACE SECURITY INTL INC MANTECH INTL CORP MARCONI CORP PLC -ADR MAXIMUS INC MECHANICAL TECHNOLOGY INC METALS USA INC MICROLOG CORP MOTOROLA INC NETPLEX GROUP INC NORTHROP GRUMMAN CORP OPTELECOM INC ORACLE CORP ORBITAL SCIENCES CORP OSHKOSH TRUCK CORP PAR TECHNOLOGY CORP PHILIP MORRIS COS -PRE FASB PROCTER & GAMBLE CO RAYTHEON CO REINHOLD INDUSTRIES -CL A ROCKWELL COLLINS INC SM&A CORP SSP SOLUTIONS INC SATCON TECHNOLOGY CORP SCIENCE APPLCTNS INTL SECURE COMPUTING CORP
8
SENSYTECH INC SIERRA HEALTH SERVICES SIGNAL TECHNOLOGY CORP SPACEHAB INC SPIRE CORP STEELCLOUD INC STEWART & STEVENSON SERVICES SUPERCONDUCTOR TECHNOLOGIES SURMODICS INC TAYLOR DEVICES INC TEKNOWLEDGE CORP TELECOMMUNICATION SYS INC TETRA TECH INC TEXAS INSTRUMENTS INC TEXTRON INC THERMO ELECTRON CORP TITAN CORP URS CORP UNISYS CORP UNITED INDUSTRIAL CORP UNITED TECHNOLOGIES CORP VSE CORP VERSAR INC VIASAT INC WASHINGTON GROUP INTL INC