Collective Action under the Articles of Confederation

Published by EH.NET (April 2002)

Keith L. Dougherty, Collective Action under the Articles of

Confederation. New York: Cambridge University Press, 2001. xii + 211 pp.

$50 (hardback), ISBN: 0-521-78209-0.

Reviewed for EH.NET by Robert A. McGuire, Department of Economics, University

of Akron.

In this relatively short book, Keith Dougherty, assistant professor of

political science at Florida International University, begins by noting that

under the Articles of Confederation the thirteen states did not fully

contribute to the public finances of the central government because of the

system of unenforced requisitions. Under the Articles, Congress determined its

fiscal and military requirements and then requested each state to contribute

its share of men and/or money to the central government, a share that was based

on the state’s size (white population). Each state was to assess its citizens

an amount necessary to meet the requisitions, collect the taxes at the

state-level, and forward the requisition to Congress, using the tax revenues to

pay for the men or the money requested. Existing scholarly treatments of

Confederation public finance concentrate on the fact that the Confederation

government struggled to meet both its military and its domestic and

international fiscal responsibilities because the states contributed so little.

Dougherty, however, emphasizes the conundrum that the states contributed so

much of the requisition, voluntarily without the incentive of enforcement. So

his quest is to explain why the states contributed to the central government

when there was no enforcement mechanism under the Articles of Confederation.

Drawing on Mancur Olson’s The Logic of Collective Action (1965),

Dougherty offers a public goods approach to the issue of Confederation public

finance. He indicates that the overwhelming proportion of Confederation

expenditures prior to 1783 was used to provide national defense and the

overwhelming proportion after 1783 was used to pay off the public debt incurred

because of the Revolution. National defense and paying off the public debt can

be viewed as public goods because once the central government provided them

they were nonexcludable and nonrival goods — any individual could be a

nonpayer, free ride on their provision, and consume the goods, and the same

amount of the goods still existed for others to consume. Once produced it is

impossible to exclude any individual from receiving the benefits of the good

(nonexcludability) and consumption of the good by any one individual does not

diminish the amount of it available to any other individual (nonrivalry).

Dougherty applies this analysis to the state-level under the Articles of

Confederation, arguing that no state could be excluded from receiving benefits

and consumption by any one state did not diminish the amount available to any

other state once the Confederation government provided the public goods.

Because the central government was providing nonexcludable and nonrival goods,

a state would have had little incentive to pay for their production and would

have attempted to free ride on the payments of other states. Yet because of

this free-rider problem, there would have been a shortfall in the number of men

and money contributed by the states to the central government. As a result, the

socially optimal amount of the public goods would not have been produced. The

standard solution to the lack of voluntary provision of public goods is to have

governmental provision, as it has the legal authority to enact compulsory

taxation to raise the resources necessary to produce the goods. However, under

the Articles of Confederation, while the central government was authorized to

provide national defense and public debt payments, it had no authority to enact

compulsory taxation; it could only request the men and money from the states.

So Dougherty asks: Why did any state contribute at all? His answer is that

states contributed parts of their requisitions because the Confederation

government was not actually producing pure public goods. According to

Dougherty, the Confederation government produced “joint products” that provided

both “public” benefits and “private” benefits. The national defense

produced during the war and payments on the public debt after the war provided

both nonexcludable and excludable benefits. When a particular state expected to

receive excludable private benefits from the military defense or debt payments

of the Confederation government, while other states were not expected to

receive those benefits, the state had an incentive to contribute resources to

produce the goods. Dougherty maintains that because national defense during the

Revolutionary War was neither spatially nor temporally homogeneous, some state

or states during some time period were protected from the British while others

were not. Because the public debt was not homogeneously held across the states,

debt payments during any time period were received by citizens in some state or

states and not by others. As a result, according to Dougherty, a state was more

likely to contribute its requisition when it expected to receive these

excludable private benefits.

This view of Confederate public finance is supported with historical and

quantitative evidence, including the results of a couple of estimated

regression models. The regression findings indicate that a state contributed

more men to the central government the closer existing Confederation troops

were to the state (it was being protected more) and the greater the number of

British troops in America (a greater threat to the state). The regression

findings also indicate that a state contributed more money to the central

government when its citizens received more of the Confederation debt payments

(more “private” benefits). Dougherty also maintains that two historical case

studies offer additional support for his “joint products” hypothesis: 1)

Virginia was more willing to contribute men to defend against Indian

hostilities in 1786 than were other states because the natives were a greater

threat to Virginia’s territorial claims and its citizens. 2) A state

contributed more men in response to Shay’s Rebellion the more it expected to

benefit from quelling the rebellion.

Although it is always easy to be critical, in the case of Collective Action

under the Articles of Confederation, there are evidential and theoretical

aspects of the book that I seriously question, including the basic methodology

and public goods conceptualization. The approach taken throughout the book is

in terms of what a state did or should have done not what any individual did.

Dougherty’s unit of analysis is a state not an individual. As a result, all

discussion of the benefits received from the actions of the Confederation

government is in terms of whether the benefits are excludable or nonexcludable

and rival or nonrival to a state and whether a state considers the benefits to

be private or public. As a consequence, Dougherty’s use of the concepts of

excludability and rivalry and his discussion of the existence of private or

public benefits are misplaced. The concepts should refer to whether an

individual can be excluded or not from consumption, whether an individual’s

consumption reduces the amount of a good, and whether only the individuals who

are a party to an action benefit (private benefits) or whether citizens as a

whole benefit from the action of others (public benefits). The concepts of

excludability, rivalry, and private benefits are not appropriately applicable

to an entire citizenry of a state. (In one instance, Dougherty does recognize

that the behavior of the state legislators is important as he acknowledges that

public securities-holding state legislators would have specifically benefited

from Confederation debt payments. As a result, he attempts to control for the

effects of these specific benefits, albeit, with a couple of rough proxy

variables in an alternative regression. See p. 96, note 15.)

A more appropriate approach to Confederate public finance would be to examine

the choices of the individuals involved in the decisions to contribute the

requisitions, to examine the choices of the presumably hundreds of state

legislators responsible for deciding to contribute men and/or money to fund the

requisitions. (Dougherty implicitly dismisses this approach in a technical

appendix. See p. 183, note 1.) With state legislators as the unit of analysis,

the question would then be: What were the benefits to a state legislator if he

supported fully funding the requisition? Of course, the benefits to a

legislator would take into account the benefits of the public goods to the

legislator’s constituents. Furthermore, the more appropriate concepts to use in

the examination of the goods produced by the Confederate government are the

concepts of “local” public goods versus “global” public goods rather than

“private goods” versus “public goods.” What the Confederation government

produced was overwhelmingly “local” public goods rather than “global” public

goods, but public goods nonetheless. Drawing on the theory of clubs, goods that

confer nonexcludable and nonrival benefits over a limited group of individuals

rather than over an entire population can be referred to as “local” (or

“quasi”) rather than “global” public goods. But military defense that protected

only one state, or a small group of states, still is a public good. Once

produced the consumption of it is nonexcludable and nonrival to the residents

of the limited number of states. Such military defense does not confer

“private” benefits in the way Dougherty indicates. (The problem is that

throughout the book Dougherty interchanges the use of the words “private,”

“state,” and “local” in reference to the benefits a state received.)

An issue that appears to have been slighted is that given military technology

that existed in the late eighteenth century it is not appropriate to refer to

late-eighteenth century “national” defense in the sense that we think of today.

Technology in the late eighteenth century made military defense “local”

defense. Yet defense of a local group of individuals is still a nonexcludable

and nonrival public good. The benefits of the defense were in no way restricted

to merely the individuals involved in the action, excluding others from

receiving them. They were not private benefits. As a result, legislators in

states whose citizens received the benefits would be more willing to assess

taxes on their citizens to fund such local or quasi- public goods. Furthermore,

as technology changes over time, military defense that at one time was a

“local” public good can evolve into a “global” public good. Thus later, under

the Constitution, the concept of a “national” defense produced efficiently by a

central government is then more appropriate. (Finally, even if the

Confederation government actually produced goods that had both private and

public benefits, the goods generally would be described as “impure” public

goods rather than as “joint products,” which in economics has another meaning.)

Even setting aside the methodological issues, the evidence presented still is

problematic. For example, the data provided by Dougherty on states’ compliance

with the requisitions in one case suggest very little, if not almost trivial,

compliance because of the manner in which quarterly compliance rates between

June 1782 and March 1789 are calculated, compliance as a proportion of a

running total of the requisitions (p. 42 and table 3.1). But later the same

data suggest fairly high compliance because of a different manner in which

compliance is calculated, total compliance as a percentage of total

requisitions for the period June 1784 and March 1789 (pp. 94-95 and table 5.3).

Dougherty acknowledges that the two sets of compliance figures are not

comparable but offers no convincing explanation for including both sets (p.

43). Moreover, relying on the earlier figures that indicate generally trivial

compliance rates, he argues that compliance was unusually high. This conclusion

is reached because he employs a third way of looking at compliance, he notes

that twelve of the thirteen states or 90 percent of them contributed some men

or funds at least one time during the entire period. This 90 percent figure is

said to be unusually high because contemporary experimental data on the

voluntary provision of public goods indicates that only 55 to 65 percent of all

participants contribute voluntarily (p. 45, note 9). Perhaps a more appropriate

figure to compare to the experimental data percentages would be the percent of

all state legislators who voted to fully fund the requisitions. In another

instance, the results of one of the estimated regressions surely are

questionable. Using OLS, Dougherty estimates a simple regression of state

compliance rates on public debts held within a state without a constant term,

with one observation for each of the thirteen states. The reasons for omitting

the constant are: 1) compliance would be zero if public debts held (the

“private” benefits) was zero; 2) all possible degrees of freedom are needed

when there are so few observations; and 3) in any event, when the regression

model was estimated with a constant term it was not significant anyway (see

pp.93-94).

Despite any of the problems with the book, I still found it informative and

enjoyed reading it. And there is the new twist. It is refreshing to see

Dougherty’s focus on why the states contributed a nontrivial portion of their

requisitions rather than — as so many others have done — merely bemoaning the

fact that they contributed so little.

Robert McGuire is author of To Form a More Perfect Union: A New Economic

Interpretation of the United States Constitution (Oxford University Press

2002) and “The Confederate Constitution, Tariffs, and the Laffer Relationship,”

Economic Inquiry (July 2002) with Norman Van Cott. He also has primary

research interests in examining the impact of parasitic diseases on economic

history and growth and is author of “Biology, Diseases, and Economics: An

Epidemiological History of Slavery in the American South,” Journal of

Bioeconomics (1999) with Philip Coelho.