Politics and Patronage:
The Function of Dysfunctional Regional Trade
Agreements
Julia Gray1
Assistant Professor
University of Pittsburgh
April 23, 2010
1 Paper
prepared for conference on “The Politics of Preferential Trade Agreements,” Princeton
University, April 30-May 1, 2010.
1
Introduction
This paper addresses two related concerns: how can we compare the functioning of trade
agreements across regions, and why might dysfunctional agreements persist? The first
concern is one that more and more social scientists are tackling. Until recently, scholars
looking to evaluate the effects of regional trade agreements (RTAs) have treated them
as like units. Many empirical studies have tended to operationalized RTA membership
as a dichotomy: a country is in one or it isn’t, and the relative capacity of different
agreements is ignored. Yet we know that wide variation exists across trade agreements,
in terms of their ability to accomplish their goals and their structure — and many of the
hundreds of agreements on offer today sit at the low end of that spectrum. Given that
many regional agreements — such as the European Union (EU), the North American Free
Trade Agreement (NAFTA), or even to some extent Mercosur — are quite effective in
promoting trade and fulfilling the goals they set out for themselves, where plenty others
are not, how can we capture these differences using comparable measures across regions?
In terms of the second concern, the existing theories do not sufficiently explain when
and why regional trade agreements fail to promote trade or why leaders would even bother
to negotiate agreements that are likely to be ineffective. We observe many types of RTAs
across the world — highly ambitious but potentially ineffective ones, such as the proposed
African Free Trade Area mentioned above; ambitious and effective ones in Europe and in
North America; and no shortage of unambitious and ineffective ones that exist in name
alone, such as the Latin American Integration Association, the Mano River Union, and
the Economic Community of the Great Lakes Countries, to name a few. How can it be
an equilibrium that countries continue to put money into the staffing and maintenance
of these agreements, if they are truly ineffective?
One class of arguments in the international organizations literature states that the
persistence of ineffective agreements is a type of principal-agent problem, where bureaucratic drift results from principals’ inability to monitor their agents (Dreher, Sturm and
Urspung 2008, Finnemore 2005, Thompson 2009). These arguments treat international
1
organization as a dysfunction that has somehow slipped beyond the control of the governments that created them. However, particularly in developing countries, leaders are
actually very involved with the workings of these agreements. Summits and diplomatic
working groups occur several times a year, and the activities of the agreements are often
above-the-fold news items in member countries’ periodicals. Thus, it seems improbable
that leaders are unaware of or unable to control these agreements.
This paper proposes a new theory to explain the continued existence — and, indeed,
the growth — of trade agreements that do not seem to actually promote trade. I draw on
the literature in bureaucratic politics to argue that even if these agreements are relatively
ineffective in increasing trade, they can serve valuable political or patronage functions for
their member states. Leaders benefit from pooling resources at the supranational level
and utilizing those resources to enrich themselves or their cronies.
I use two new sources of data to test this argument. The first is an original dataset
collected from expert surveys on the many attributes that make for a successful agreement
(Gray & Slapin 2009). Expert surveys are a valuable way of operationalizing concepts
where no other good data exist, and this dataset is a tremendous contribution to our
understanding the dynamics of different RTAs. There is no shortage of anecdotal evidence
of the effectiveness or lack thereof of different agreements, but social scientists are left
with few indicators that can be applied cross-regionally. The expertise of officials in the
trade agreements themselves as well as external observers at the multilateral level can
shed much light on the actual functioning of these RTAs, beyond what the treaties and
trade flows alone can tell us.
The second is new and detailed data on the annual budgets of these organizations.
As part of a larger data-collection project, I have gathered historical data on the annual
expenditures of five regional trade organizations in Latin America and Africa, with five
more (in Europe and Southeast Asia) to be collected later in 2010. These data offer valuable insights into the priorities of these organizations, the contributions of each member,
and their day-to-day operations. Most of these organizations keep highly detailed records,
2
and access to their annual budget items gives us a valuable way of making comparisons
across organizations in like units.
This paper proceeds as follows. The next section discusses the gaps in the extant literature on regional agreements and their effectiveness. I bring in insights from the literature
on comparative bureaucracies and comparative pork to present an original theory that
explains the persistence of agreements that seem ineffective in terms of promoting trade:
these agreements survive because they allow member governments to exercise patronage.
2
If RTAs don’t promote trade, why are there so
many of them?
Regional trade organizations have been on the rise since the 1990s, and they have become increasingly important with the failure of the Doha rounds of multilateral trade
talks. Fewer than 20 RTAs were reported to the GATT in 1990, but by July 2005, a
total of 330 had been notified to the WTO, and only one WTO member — Mongolia
— was not additionally a member of a regional trade agreement. The WTO estimates
that over 400 RTAs will be in existence by 2010, including those in existence that have
not been notified, those that have been signed but are not yet in force, those presently
being negotiated, and those in the pipeline. GATT’s Article 24 allows RTAs to be set
up as an exception to the most-favored nation rule that members must uphold, as long
as those RTAs complement the multilateral trading system. To that end, much of the
initial literature on the formation of regional trade agreements focused primarily on their
economic effects in the context of multilateralism (Panagariya 1999). Since the boom in
popularity of RTAs coincided with the Uruguay Round, and the greater institutionalization of the multilateral agenda with the transformation of the GATT to the WTO, most
researchers at the time were concerned with whether RTAs would have a trade-diverting
effect or would potentially derail multilateralism (Bhagwati 1991, Bhagwati, Greenaway
& Panagariya 1998, Krueger 1999).
3
Only recently have researchers begun to study RTAs as institutions in and of themselves, applying the theories from the literature on international organizations. However,
many of the empirical studies treat membership in RTAs as a dichotomy — either a
country is a member of an RTA, or it is not (Mansfield & Reinhardt 2008, Milner &
Büthe 2008). Indeed, a country could be a member of several RTAs (as Kenya is a
member of SADC, COMESA, and the EAC), one dysfunctional one (Libya in the Arab
Maghreb Union), or one highly functional one (Germany in the European Union) and
receive the same empirical treatment. Clearly, vast differences exist across these different
possible scenarios, yet researchers are only beginning to try to weigh RTAs according to
their differing characteristics (Stinnett 2007, Kim 2008, Kim & Hicks 2008, Mansfield &
Reinhardt 2009, Haftel 2007).
The issue of effective institutional design has long been a concern of international
relations scholars — though the institutions under study have primarily been at the
multilateral level. Many have argued that deeper legalization — that is, treaties and
procedures that put the functioning of the institution in the hands of third parties, rather
than leaving them at the mercy of heads of state — makes for more effective forms of
cooperation (Keohane, Moravcsik & Slaughter 2000, Smith 2000). Empirical tests of
these propositions have tended to focus primarily on the design of the treaty itself and
the language employed (Koremenos 2005, Rosendorff 2005). Furthermore, these studies
have tended to focus primarily on multilateral or the more ambitious of the regional
institutions, such as studies of the WTO (Rosendorff 2005) or the European Union (EU)
(Moravcsik 1998). Among the virtues of those organizations for social scientists is the
relative abundance of data they make available. Thanks to strict reporting procedures and
data standards, it is relatively straightforward for researchers to conduct empirical studies
of those organizations’ effect. However, a comprehensive test of the effects of legalization
would examine not just organizations with high levels of legalization and documentation,
but also those with medium- or low levels as well. Indeed, those organizations most
frequently studied represent only a small fraction of the myriad organizations in the
4
world today.
Another difficulty with focusing on levels of legalization or the language of the treaties
in question is that the simple presence of a court or of a highly legalized agreement does
not necessarily mean that the language is enforced or that the courts are active. Many
countries, in practice, sidestep their own regional dispute settlement mechanisms, preferring instead to seek redress through the WTO’s dispute-settlement mechanism (LacarteMuró & Gappah 2000, Busch 2007, Bown & Hoekman 2005). The Andean Community’s
court — lauded by at least one scholar (Alter 2009) as a beacon of institutional design
on par with the institutions of the European Union — has in fact only seen fewer than
a dozen cases of states taking other states to court in the 15 years of its inception.1
Coding the language of the treaties can also be deceptive, since it is extremely easy to
lift impressive-sounding language from other agreements. organizations that do function
well. Take, as just one example, Article 30 of the courts of COMESA and the EAC (two
different RTAs in Africa), which establishes the relationship between national courts and
the DSM: “National Courts or tribunals may refer a matter to the Court or request the
Court to give a preliminary ruling on the application or interpretation of the Treaty or
the validity of regulations, directives and decisions of the Common Market.” Though the
exact language differs, this borrows heavily from the EU’s preliminary ruling procedure
(Article 234 of the Consolidated European Treaties). This article is notable because it
allows national courts to refer questions about the compatibility of national law with
European obligations to the European Court of Justice, an unprecedented move that was
critical for EU legal integration (Stone Sweet & Brunell 1998, Alter 1998).
Even if they do not have the intended consequence in terms of effectiveness, some
scholars have already pointed out that such levels of bureaucratization inevitably afford
opportunities to inefficiency at best and corruption and pathology at worst (Bryant and
Lindeberg 2001, Dijkzeul and Beigbeder 2003, Barnett and Finnemore 2006). Weber
(1978) notes that even the most minimal bureaucracy holds authority of providing public
1
Author interview, Elba Roo, Andean Community Secretariat, 6 July 2009.
5
goods such as infrastructure. Control over that process means that public goods can be
exploited for private profit. Thus, as intergovernmental organizations, RTAs can potentially create private rents through a well-staffed secretariat that provides jobs, prestige
and other perqs to member state nationals.2 There may be incentives to provide private
rents through RTAs when a general environment of corruption and the personal exploitation of public office exists at home. This view is supported by Vaubel (1986) — who
discusses how politicians can use opportunities for international coordination in a collusive manner, to increase their prestige and to stifle criticism of policy — and extended
by Vaubel, Dreher, and Soylu (2007), who show that international organizations have
fewer employees when member states have more oversight capacity. This implies that
domestic law-makers who scrutinize government spending would frown upon joining a
high-cost regional organization that lavishes bureaucrats with perqs. These arguments
support a discussion of the benefits of international organizations is incomplete without
an awareness of the ability of those organizations to reward not just member states, but
also individual politicians.
There exists a substantial literature in comparative and American politics on patronage and pork in bureaucracies. However, many of these works are concerned with
explaining the overall effectiveness of policy, not on the motivations for making ineffective
appointments. For example, Lewis, Krause and Douglas (2006) examine the relationship
between political appointments and policy effectiveness. Ting’s bureaucratic redundancy
theory (2008), building off Landau (1969), argues that principals choose multiple agents
in a bid to increase the effectiveness of an organization. His setup, however, assumes
that principles always want to attain “good” policy outcomes, and that appointments are
never an end inof themselves.
2
Debate surrounds the role that corruption plays in economic development. Some argue that corruption can actually be useful in circumventing bureaucratic and rigid state structures (Bates 1981,
Huntington 1968). Others claim that bloated bureaucracies can impede development in the long term
(Hutchcroft 2002). In this context, I argue that it is possible to have an RTA whose bureaucracy generates private rents, but that does not necessarily preclude an effective RTA. Rather, the process driving
the creation of an RTA that is broad in scope differs from the process that drives the creation of an
effective RTA.
6
Some scholars, however, have focused on how policy can be used to accomplish political goals. McGillivray (2004) argues that industrial policy can serve as a means of
redistribution from one group to another that is politically efficient. Lewis (2008), in a
book that extensively reviews patronage in American politics, points out that “the vast
majority of federal [patronage] jobs were located outside of Washington, D.C. They were
an important political resource and were viewed proprietarily by congressmen who sought
to distribute patronage to local and state machines that brought them to power” (1314). He also notes that “federal patronage has also historically been a means by which
Congress secured control over federal administration locally. ... Appointed positions also
help members to the extent that they help their party or individual election prospects.”
These works give clues to how outcomes that seem inefficient from a policy standpoint,
but are efficient — and even irreplaceable — to individual politicians.
3
A Theory of Patronage in Regional Organizations
How can we explain the disproportionate operating costs or infrastructure that surround
what are widely agreed to be ineffective agreements? It should be noted that the baseline
for the amount of extra maintenance — in terms of a support staff or other bureaucracy
— that a trade agreement needs is practically zero. For 14 years the Washington, DCbased secretariat for NAFTA — the largest of the three secretariats, with the other two
in Ottowa and Mexico City — consisted of three full-time staff, which only increased to
five staff members in 2009.3 The GATT took over office space that had been allocated by
the UN to house the failed ITO and had very few staff members at the outset. (Hudec
1998). The myriad preferential trade agreements negotiated by Washington DC and
Brussels, for example, are administered out of the USTR and the Directorate General for
Trade, respectively — no one rents separate buildings. Similarly, Mercosur’s staff consists
primarily of legal experts and economists, and the organization exists in rent-free space
3
Interview, Marsha Iyomasa, NAFTA secretariat, March 8 2010.
7
provided by the United Nations.
The answer lies in thinking about the other benefits that secretariats with broad scope
in issue areas and numerous staff members might provide to governments. Whether or
not they increase trade, the bureaucracies surrounding RTAs can offer substantial private
rents for member state governments. It should be noted that our discussion of private
rents refers not to the benefits accrued by domestic interest groups, lobbyists, or sectors, as
described by Tullock (1967) and Krueger (1974). Rather, it designates the private benefits
that politicians can squeeze out of thickly institutionalized RTAs. The bureaucracy of
an intergovernmental agreement can provide politicians with several forms of private
rents: for example, the opportunity to engage in clientelistic behavior, by rewarding their
cronies with posts; the ability to secure lavish per diems at conferences and meetings;
and diplomatic passports, as just a few examples.
Leaders will have an incentive to support the existence of supranational organizations
if those organizations give them access to sources of patronage and rents that extend
beyond what is available to them at the national level. Since many of these organizations are funded externally — organizations such as the EU and USAID, as well as the
Scandinavian development programs, provide up to 50 percent of the budgets of many
of these regional agreements — it is in member states’ benefit to pool their resources
at the supranational level, since they will be able to tap into more resources than what
would have been available to them if they simply acted alone. Countries that have high
domestic environments of corruption, then, will be more likely to feed into supranational
organizations that also serve this function.
I argue that regional agreements can extend private rents to individuals in several
ways, though at present this paper will only examine one of these causal mechanisms
empirically. First, through salaries and per diems, they can generate significant sources
of funds for those involved in their activities. For example, in 1997, COMESA’s secretary
general, Bingu Mutharika, was fired for misappropriation of funds. An internal report
stated that “there is ample evidence of financial malpractices. The Secretary General
8
has used COMESA funds to finance missions which cannot be confirmed to be official
and beneficial to COMESA. The Secretary General has also used COMESA resources
for personal activities.” These included misappropriating COMESA funds for “on such
items as air tickets, daily subsistence allowances, office furniture, motor vehicle repairs
and spare parts etc.”4 Mutharika also in 1996 tried to hire one of his cronies in the
post of Executive Secretary of the Clearing House in a manner that “ignor[ed] the laid
down procedures. This attempt was made notwithstanding the fact that several other
applicants from various Member States had applied for the job. The applications were
retained in the custody of the Secretary General and [the candidate] was not even one of
the applicants.”
5
These benefits can trickle down to staff as well. One expert describes SADC programs
by saying that on the books, “senior economists took home $30 per months, but pocketed a
flat per diem of something like $280 per day when they were traveling, regardless of actual
expenses – even if their expenses were fully covered. One colleague attended a six-week
[training] course and used his per diems to go into the minibus business.”6 These funds can
be proportionally huge, particularly in environments where private-sector opportunities
are low. For illustration, annual salaries for drivers of SADC were around three times the
average local salary; upper-level officials at COMESA received housing allowances of 84
percent of their base salaries (which were already around four times greater than average
local salaries); and upper-level secretaries at SADC were allocated Mercedes Benz S320s
in 2003.
How might we know a patronage administration when we see it? States where the
environment is more conducive to the use of public office for private rents would tend to
be more likely to use supranational organizations as another form of patronage. Thus,
we could expect that supranational agreements would serve as likely repositories for pa4
“Report of the Special Committee on the Operations of COMESA, 1992 to date.” 04 March 1997
The committee also confirmed the “There is no mechanism in place to ensure the effective implementation of the decisions of the policy organs. Since 1992, several major decisions remain unimplemented
because of the Secretary General’s failure to find time to address the critical areas that need attention.”
6
Author interview, Bruce Bolnick, chief economist at Nathan Associates, 30 January 2009
5
9
tronage agreements in the following scenario:
• H1 States with higher levels of corruption are more likely to support organizations
that exist for patronage reasons.
There are other mechanisms by which RTAs can serve as efficient means of political
and private benefits for the countries involved, though this paper will not examine these
mechanisms in detail. Second, they can offer opportunities for prestige and recognition
abroad, and an assertion of their place in the international system (Chayes & Chayes 1995,
Hutchcroft 2002). By offering extensive travel opportunities (SADC’s average annual
travel budget is around $300,000, which includes trips not only for staff members but
also for ministers of member countries), these organizations can help legitimize politicians
and even countries in the eyes of domestic publics. Third, RTAs can serve as havens
for political figures who have fallen out of favor. Every one of Argentina’s deposed
presidents following the 2001 financial crisis subsequently left the country to take up
posts in either Mercosur or the Andean Community, where they could collect salaries and
remain active in political life outside of the country for a few years. Finally, RTAs could
serve as important political goals that have nothing to do with trade. In his memoirs, Lee
Kuan Yew noted that “the unspoken objective [of ASEAN] was to gain strength through
solidarity ahead of the power vacuum that would come with an impending British and
a possible US withdrawal. Indonesia wanted to reassure Malaysia and Singapore that ...
its intentions were peaceful and that it had abandoned Sukarno’s aggressive policies.”7
Similarly, in Africa, one expert noted that even if organizations such as the Mano River
Union and the East African Community are not great trade-promoters, “the fact that
they have Eritrea and Ethiopia meeting regularly at the same table, or Tanzania and
Uganda, is an incredible accomplishment.”8 Thus, RTAs that might not promote trade
might help reduce conflict. For the moment, however, I leave these propositions aside,
since testing them is beyond the scope of this paper.
7
Yew 2000.
Author interview, Douglas Brew, Trade Relations with African, Caribbean and Pacific countries,
European Commission, 12 May 2008.
8
10
The following sections tests these arguments using different sources of original data
on regional trade agreements.
4
First Test of Hypothesis: Survey Data
How might we operationalize the degree of patronage within organization? Many works
to this end examine biographical data on staff to arrive at assessments of those workers’
competencies (Lake 2000, Lewis 2008, Krause and Joseph 2010, Alexiadou 2010). The
organizations here, however, do not keep such data on file. They do, however, keep data
on many other aspects of their functioning, and there are other ways in which data on
these organizations can be gathered.
One such strategy is conducting expert surveys. This section presents a portion of a
new dataset on 24 distinct characteristics of RTAs the world over, as well as an overall
scoring of whether the RTA is strong or weak (Gray and Slapin 2010). The data collection process involved conducting expert surveys, a practice that is widely used in political
science to aid in comparisons where data do not already exist (Laver 1998, Whitefield,
Vachudova, Steenbergen, Rohrschneider, Marks, Loveless & Hooghe 2007, Hooghe, Marks
& Wilson 2002, Benoit & Laver 2006, Marks, Hooghe, Steenbergen & Bakker 2007, Busch,
Reinhardt & Shaffer 2009). Expert surveys have been shown to offer substantial leverage
in operationalizing the political positions of parties, in situations where their own manifestos are suspect and little other available data, such as voting records, can be leveraged.
This mirrors the situation with the actual functioning of regional trade agreements, where
founding documents can only be trusted to a limited extent, and little other data on their
actual progress are available. Expert surveys can provide us with substantial leverage
for capturing concepts that may be hard to operationalize using other types of data. All
responses were given anonymously and represent the views of individuals. They in no
way represent the views or opinions of the respondent’s institution or of their member
states.
11
Though they are not without limitation, expert surveys represent the best method
for obtaining comparable data across numerous RTAs for a wide variety of dimensions.
Others have coded data on legalization (including information on dispute mechanisms)
by analyzing the contents of treaties. This, however, assumes that what is written in the
treaty is what is done in practice. As discussed earlier, this is clearly not always the case.
For other dimensions, hard data are difficult to come by. Thus, these data represent a
rich spectrum of the possible angles through which RTAs can potentially be studied.
Our sample includes 25 respondents, many of whom filled out the survey for more
than one RTA, leading to a total of 3,924 individual data points encompassing 35 different
agreements. Though the number of total respondents may seem small, it should be noted
that expert surveys have relatively few respondents. For example, the Chapel Hill expert
survey on party positions in Europe obtained 98 survey responses (34 percent of the
total solicited)—and parties are far more visible and public than regional agreements
(Steenbergen & Marks 2007, Marks et al. 2007). It is surprisingly rare to find experts
who are able to comment on the many aspects of the actual day-to-day workings of
RTAs. Perhaps as a function of the previously discussed tendency for academics to think
of RTAs as like units, relatively few academics have expertise on the details of individual
RTAs. The majority of the respondents came from practitioners, specifically the WTO’s
Regional Trade Agreements unit and the European Union’s Directorate General for Trade,
which employs the main trade negotiators for the EU. These include issue area as well as
regional specialists. We attempted to balance the representation from the developed and
the developing world by soliciting responses through email from experts from Southeast
Asia and Africa, and from in-country visits to Zambia, Botswana, South Africa, Peru,
Argentina, and Uruguay, as well as to various organizations in Washington, DC; the WTO
offices in Geneva; and the Directorate General for Trade in Brussels. The final sample
includes one respondent from Africa; six from Latin America, and 18 from Europe. To
eliminate the biases that respondents might have in evaluating their own organization,
we dropped those observations where respondents scored their own institution.
12
Our survey instrument asked participants to rank as many RTAs on which they felt
comfortable commenting along a simple scale, from one to 10, for each different dimension.9 In addition, the respondents were asked to report the importance of each dimension
to the particular RTA, from one to 10. We included this “salience” dimension so that
RTAs that scored low on a particular issue but for whom that issue was unimportant
would not be unduly penalized; for example, an RTA may score low on “international
influence” but may have no such aspirations. Similarly, an RTA that scored low on “trade
in services” but to whom trade in services was important would be treated differently than
an RTA that had no intention of liberalizing trade in services. Details on the dimensions
used in this article are provided in the appendix. One of the authors was present when
the survey was completed, enabling us to clarify the meaning of the various questions
where necessary. Surveys were conducted between June 2008 to July 2009.
The WTO’s website lists 190 RTAs that have been reported to the organization.
Of that list, however, only 34 agreements are more than just bilateral arrangements.
Our survey listed those 34 RTAs as prompts, and we also encouraged respondents to
comment on any additional RTAs that they felt should be included or felt qualified on
which to comment. Respondents were also encouraged to only score those dimensions
that they felt qualified to evaluate, meaning that participants did not necessarily score
every dimension for every RTA that they chose to rank. In aggregate, respondents at the
EU coded attributes of 25 different RTAs, while respondents at the WTO coded a total of
30 different RTAs. Unsurprisingly, the better-known RTAs received more responses; the
EU received 445 data points, with EFTA, ASEAN, CARICOM, and NAFTA receiving
between 223 and 278 entries. On the low end, a few of the bilateral FTAs were scored, but
of the larger RTAs, the Mano River Union, the East African Community, and the Arab
Maghreb Union received the lowest number of respondents (between 8 and 39 total).
Although we received some responses that scored bilateral agreements, for the purposes of this analysis we are restricting our sample to include only multilateral RTAs of
9
Much of the design of the survey instrument comes from Benoit and Laver’s (2006) expert surveys
on party positions in modern democracies.
13
neighboring states that have or could potentially have a secretariat or a dispute settlement
mechanism. This stands in contrast to many other studies of RTAs (Smith 2000, Kim &
Hicks 2008). Although bilateral agreements would not necessarily be excluded from our
theory — since they represent possibilities for interstate cooperation with minimal levels
of institutionalization — data limitations make inclusion of the whole spectrum of potential RTAs infeasible. The relatively few number of observations for each of the scores
of bilaterals would make our estimates unreliable. Thus, we limit our empirical analysis
to the RTAs that notify to the WTO as more than bilateral agreements. It should be
noted even within this subsample, substantial variation exists on every dimension coded,
so selection on the dependent variable should not be a concern.
Table One in the appendix shows mean scores for these dimensions on aggregate as
well as for each RTA, along with overall scores for the RTAs in question. It should be
noted that the standard deviations around these responses are relatively small, indicating
the substantial degree of expert consensus on the actual performance of these agreements.
This should give confidence in the degree of intra-item reliability of these measures. Table
Two in the appendix shows the degree of correlations among many of these items. Note
that the degree of correlation between an agreement’s scope and its ability to meet its
goals is relatively low (.30), indicating that many of the more expansive agreements do
not live up to their ambitions.
To operationalize the proposition listed above, I test whether the presence of a highrent environment leads to the creation of RTAs with greater scope. Of course, many effective agreements also have broad scope, but the two are not synonymous. This is borne
out by a look at the relationship between survey respondents’ rankings of whether “The
agenda items over which PTA members negotiate go beyond traditional trade issues,”
for the “Scope” question, and “Has a well-trained staff and well-organized bureaucracy”
for the “Quality of Infrastructure/Secretariat” question. Organizations that have broader
scope can potentially offer more opportunities for patronage appointments or enrichment,
while the quality variable reflects on the competencies of the staff that are appointed.
14
FIGURE ONE ABOUT HERE
Many agreements have a bureaucratic quality that matches the scope of their goals,
such as the EU on the high end, NAFTA in the middle, and SPARTECA on the low end.
But still more RTAs tend to have relatively high-scope agreements that are perhaps illserved by poor bureaucracies (the CIS, APEC, and the SCO). Interestingly, the converse
seems not to be true; that is, few low-scope agreements have high-quality bureaucracies.
As discussed in the theory section, I anticipate that states that tend to create vast public
bureaucracies of low capacity at home — that is, bureaucracies that serve the purpose
of patronage rather than efficiency — will tend to replicate those institutions on the
international level as well.
I test this argument by examining the determinants of RTAs to be high in scope but
bureaucratically inefficient. I capture this by dividing the scope variable by the secretariat
quality variable (to capture the fact that high values in the “secretariat” variable represent
a well-functioning bureaucracy, whereas we are interested in large but poorly functioning
agreements). Thus, higher values should mean agreements whose competencies fall short
of their, where lower values represent more efficient organizations. I log this variable to
normalize the distribution.
I regress this variable on two separate variables that model the exogenous environment
for private rents, as described in the theory above. The first is the level of transparency
in member states, as gathered by Transparency International. States where obfuscation
persists would be expected to partake in agreements where the opportunities for rentextraction were high. I invert this scale to reflect this argument: high levels of obfuscation
(that is, low levels of transparency in the original scale) should be associated with inefficient organizations (that is, high values on the dependent variable), so the expectation is
positive. As a robustness check, I also include a variable recorded from the International
Country Risk Guide (ICRG) on corruption levels in the country. Again, higher levels of
15
corruption should be associated with inefficient agreements – another expectation for a
positive relationship.
I also include several variables to control for rival explanations. One is the average
level of development among member states, logged to normalize the distribution (GDP ).
We can expect that states with higher domestic economic capacity would have less need
to resort to supranational patronage, so the expectation is negative. I also examine public
employment as a percentage of total employment among member states, averaged across
RTAs for the year at the time of RTA formation (or the first year of available data, if missing). These data were collected from the International Labor Organization. Higher levels
of public employment should indicate that the government is in the habit of providing
relatively more opportunities for its citizens than are available in the private sector. However, such opportunities are not necessarily sources of corruption in and of themselves;
thus, we must examine these effects while controlling for levels of corruption. Holding
corruption constant, we might expect no independent effect of public employment.
An alternate hypothesis would be that these organizations spend the amounts of
money that they do not as a means of enrichment or patronage, but as necessary expenditures to promote a well-functioning organization. I capture this argument by including
variables for yearly changes in the volume of trade of the states within the RTA, relative to the total that those states trade externally (intra-RTA trade), as well as yearly
changes in tariff levels. If these variables were significant, it would indicate that the relative efficiency of these organizations was simply a justifiable function of extensive their
trade-promoting activities. Because these surveys were conducted in the past two years
and had no time variance, I constructed all these independent variables from member
state data 2007, which was the most recent year for which data were available for the
majority of countries in question.
A final variable is the level of legalization in the treaties, as coded by McCall Smith
2000 (legalization). Many have argued that the level of legalization determines a treaty’s
overall effectiveness; thus, higher levels of legalization would be associated with more
16
efficient secretariats. Because of the small number of observations, I include this last
variable in a separate table. Table Three depicts the results of these specifications.
TABLE THREE ABOUT HERE
The results generally conform with the expectations of the theory described. More
corrupt environments as well as environments where the state offers relatively more opportunities for employment tend to also produce trade agreements that have broad scope
but do not have the bureaucracies to effectively carry out those missions. Less corrupt
groups of states (recalling that higher values on the corruption variable mean less corruption) tend to produce smaller-scope and more efficient bureaucracies in their trade
agreements. RTAs composed of relatively rich member states are less associated with
inefficiency; this variable is statistically significant and negative. Note that the two measures of trade-promotion within RTAs — that is, annual changes in trade volumes and
tariff levels — are not statistically significant, which offers further support to the idea
that these organizations spend money on activities that have little to do with the stated
goal of trade expansion.
It should be noted that the legalization variable — one that has been a hobbyhorse
among many scholars — is not statistically significant, and in fact the direction of the
coefficient is positive. This indicates that more legalized agreements may actually be
associated with inefficiency, rather than effectiveness, as much of this literature claims.
5
Second Test of Hypothesis: Budget Data
This section examines the hypotheses established above with the use of empirical budget
data. Though the number of organizations here is far smaller than in the estimations
above — five compared with 34 that received scores in the expert survey — this is a
rich source of data from the organizations themselves, and can examine this argument
17
from a different level of analysis. Between 2009 and 2010, I collected data from the
secretaries of the Common Market for Eastern and Southern Africa (COMESA) in Lusaka,
Zambia; the South African Development Community (SADC) in Gaborone, Botswana;
the Latin American Integration Association (LAIA) and the Common Market for the
South (Mercosur), both in Montevideo, Uruguay; and the Andean Community, in Lima,
Peru. I obtained annual data on budget allocation for each of these organizations since
its inception. These data include both income and expenditures at the yearly level.
Organizing these data into comparable, aggregate variables made it necessary to collapse many subcategories that were not consistent across organizations. This is a regrettable loss of detail, as many of these organizations, though, keep extremely extensive
records of their programs and budget line items. In 2001, SADC, for example, spent
$250,00 on calendars, $30,000 on an essay competition,10 and nearly $60,000 on the commissioning and recording of a SADC pan-national anthem. In 1992, COMESA spent
$7,000 providing ministers who attended a meeting on the PTA with leather briefcases.
Nonetheless, for the purposes of comparison across regions, some level of generalization
is inevitable.
Table Four lists averages of the budget data across the most mutually compatible
categories.
TABLE FOUR ABOUT HERE
Note that although COMESA has the highest operating budget overall, at an average $14 million per annum, it is also an organization with a relatively large number of
members. The average annual costs per member are the highest for the Andean Community, at $1,031,241.80; COMESA and SADC are the next most costly, at $647,826.09
and$610,369.00 per member, respectively; LAIA’s average costs are $396,797.92 per member; and Mercosur’s run $248,293.85 per member. Of course, these costs are not dis10
The topic was, “What has SADC achieved and what more would you like it to do to bring the
economies and the people of southern Africa together?”
18
tributed equally across members. It is usually the case that the member states’ contribution are proportional to their income — and external organizations tend to foot a portion
of the bill as well. Nonetheless, this gives a sense of the costs of the organizations relative
to the number of member states included.
To analyze these data with respect to the theory outlined above — that groups of
countries where public corruption is already widespread are more likely to use these organizations as forms of patronage — I extract two variables to operationalize organizations
that have high levels of patronage. The first is average salaries in the organization as
a ratio to local salaries for central government employees. The higher this number, the
better-paid are employees in the RTA relative to government workers at home; smaller
numbers mean that the discrepancy between local salaries and RTA salaries is less. I
obtained local salary data from the International Labor Organization’s database. Collapsing the regional data into average salaries misses some important levels of detail;
some organizations (LAIA and SADC) record the different salary levels at different levels
of seniority among staff, but the others do not. Aggregating these data into yearly averages also is deceptive; Mercosur, for example, tends to employ fewer support staff and
more expert staff, so their salaries seem higher; COMESA employs more support staff
than senior staff; and SADC is the opposite. Thus, higher salaries for experts (such as
economists and lawyers) offset the lower salaries of support staff, but these extremes are
unfortunately lost in the averages. Additionally, this level of detail tends not to exist
in national-level statistics. Because coverage for many of the countries in question was
erratic, I used averages for all available countries in the organization. If anything, we can
anticipate that this might be biased in favor of countries that pay their local employees
better, since poorer countries tended to have more missing data. Thus, this is likely a
conservative estimate of the gap between local and regional pay. In general, salaries in
Mercosur are only slightly higher than those in central governments in member states;
COMESA salaries are twice as high; salaries in the Andean Community are three times
as high; in LAIA, nearly twice as high; and 50 percent more in SADC. Thus, it is highly
19
possible that these organizations are a cushier job than those at home, and are more
probably associated with patronage appointments.
The second is program costs, per diems, and personal expenses as a share of the
overall budget. Program expenditures can be significant sources of revenue. For example,
a 1996 SADC program on illicit cross-border on drug trafficking and crime cost a total of
$237,571. The program hired out four local consultants, along with honoraria of $300/day
for 150 days, $18,000 in travel costs, and subsistence funding of $48,124. Administrative
costs for the program (including secretarial services, stationery, distribution of reports,
and local transportation) ran $30,000, with $30,000 in printing costs for the document,
along with “contingency” funding of $23,974. Higher ratios of these expenses to the
overall budget indicate greater amounts of patronage.
Tables Six and Seven depicts the result of OLS regressions on our two measures of
patronage. The sample covers five organizations over time; data actually range from 1971
(for the Andean Community) to 2009, but since Mercosur budgets only extend back to
1997, the year coverage in practice is from 1997 to 2009.I use similar independent variables as those described in the first sets of models. McCall Smith’s legalization variable
does not vary over time, but the level of legalization of a treaty does not change, and
thus this variable is included as a constant value for all years in the specification, so as
to preserve the number of observations.
TABLES SIX AND SEVEN ABOUT HERE
This is a smaller sample, with only five organizations, but consistent with the results
using the expert survey data, both sets of specifications indicate that the trade-promoting
activities of the organizations (as represented by the changes in tariffs and trade levels)
have little to do with the amount of patronage. Both measures of corruption are associated
with higher staff salaries relative to local salaries, as well as with higher amounts of slush
funding in the organization, as represented by the program and other expenses. None of
20
the other variables are statistically — including the levels of income in a country. This is
perhaps not surprising, since the previous sample included agreements from the developed
and the developing world alike, and levels of patronage are to be expected in the latter
group of countries.
6
Conclusion
This paper has introduced a novel theory to explain the persistence of what appear to
be dysfunctional trade agreements. Furthermore, I have tested this theory with two new
sources of data. As social scientists struggle to find ways to compare these agreements,
these data offer consider leverage, providing levels of detail and perspective where none
previously existed, and giving us the tools to evaluate agreements across regions. I have
demonstrated that member states that have high levels of corruption tend to be associated
with agreements that provide considerable opportunities for patronage. The levels of
inefficiency in these organizations cannot be explained by the level of legalization of the
agreement — an explanation that has gained considerable purchase in political science
— or by levels of development in the organization. Instead, countries with higher levels
of corruption tend to be associated with more inefficient organizations, in terms of their
scope in relation to the quality of their secretariats, as well as in how well-paid its staff
are with respect to local salaries, and the amount of slush funding in their budgets.
This finding paves the way for further investigation into the day-to-day workings of
these agreements, and the budget data offer considerable opportunities to do just that.
These data will shed further light on the ways in which RTAs can serve the private
interests of their constituencies. Additional data-gathering trips in 2010 (to CEFTA,
EFTA and the EU in Europe, as well as to ASEAN and APEC in Asia) will expand the
sample even further and give us a better perspective of the workings of these agreements
in a comparative focus.
These new tools for studying regional trade agreements will have important impli-
21
cations not only for scholarship in the area of economic integration, but also for policy
work. Many trade policymakers are quick to acknowledge not only the wide variation
across different RTAs, but also the frequently observed gap between the commitments
announced in summits and actual fulfillment of those commitments. I am hopeful that
the expert survey as well as the budget data will provide a consistent metric for assessing
the way these RTAs actually work in practice. This note has only presented a small
part of our rich survey dataset. Future work is required to understand the relationship
between the many different dimensions of RTAs.
These data will prove extremely useful in exploring, for example, the political functions of regional agreements; the level of influence, or lack thereof, that they have at home
and in the world; and their actual contributions to economic welfare. Given the recent
surge in regional integration, an understanding of the myriad different levels on which
RTAs function, or fail to function, is a critical tool for examining this new era of globalization. Many EU and US officials have expressed frustration in their dealings with these
agreements. Trade negotiators have no independent authority that allows them to make
deals without the explicit of government officials (many of whom are present at trade
negotiations); agreed-on provisions are not implemented; and overlapping memberships
in multiple agreements with conflicting priorities make country obligations unclear. An
understanding of what these agreements actually provide to members is a critical first
step in understanding the variation in these agreements and their impacts on a variety of
outcomes.
22
References
Alter, Karen J. 1998. “Who Are the ”Masters of the Treaty”?: European Governments
and the European Court of Justice.” International Organization 52(1):121–147.
Bates, Robert H. 1981. Markets and States in Tropical Africa. Berkeley: University of
California Press.
Benoit, Kenneth & Michael Laver. 2006. Party Policy in Modern Democracies. London:
Routledge.
Bhagwati, Jagdish. 1991. The World Trading System at Risk. Princeton, NJ: Princeton
University Press.
Bhagwati, Jagdish, David Greenaway & Arvind Panagariya. 1998. “Trading Preferentially: Theory and Policy.” Economic Policy pp. 1128–1148.
Bown, Chad P. & Bernard M. Hoekman. 2005. “WTO Dispute Settlement and the Missing
Developing Country Cases: Engaging the Private Sector.” Journal of International
Economic Law 4(8):861–890.
Busch, Marc. 2007. “Overlapping Institutions, Forum Shopping, and Dispute Settlement
in International Trade.” International Organization 61(4):735–761.
Busch, Marc, Eric Reinhardt & Gregory Shaffer. 2009. “Does Legal Capacity Matter? A
Survey of WTO Members.” World Trade Review 8(4):559–577.
Chayes, Abram & Antonia Handler Chayes. 1995. The New Sovereignty: Compliance with
International Regulatory Agreements. Cambridge, Massachusetts: Harvard University Press.
Haftel, Yoram Z. 2007. “Designing for Peace: Regional Integration Arrangements, Institutional Variation and Militarized Interstate Disputes.” International Organization
61(1):217–237.
23
Hooghe, L., Gary Marks & C Wilson. 2002. “Does left/right structure party positions on
European integration?” Comparative Political Studies 35:965–989.
Huntington, Samuel P. 1968. Political Order in Changing Societies. Yale University Press.
Hutchcroft, Paul D. 2002. “The Politics of Privilege: Assessing the Impact of Rents, Corruption, and Clientelism on Third World Development.” Political Studies 45(3):639
– 658.
Keohane, Robert O., Andrew Moravcsik & Anne-Marie Slaughter. 2000.
“Legal-
ized Dispute Resolution: Interstate and Transnational.” International Organization
54(3):457–488.
Kim, Moonhawk. 2008. “Adjusting to Institutionalized Trade Integration: Preferential
Trade Arrangements, the GATT/WTO and Pressures for Competitiveness and Compensation.” Working paper.
Kim, Soo Yeon & Raymond Hicks. 2008. “Commitment, Signaling, or Flexibility? The
Effectiveness of PTAs in the Asia-Pacific.” Paper presented at the annual meeting of
the International Politcal Economy Society, University of Pennsylvania, Philadephia,
PA, November 12, 2008.
Koremenos, Barbara. 2005. “Contracting around International Uncertainty.” American
Political Science Review 99(4):549–565.
Krueger, Anne O. 1999. “Are Preferential Trading Arrangements Trade-Liberalizing or
Protectionist?” Journal of Economic Perspectives 4(13):105ñ124.
Lacarte-Muró, Julio & Petina Gappah. 2000. “Developing Countries and the WTO Legal
and Dispute Settlement System: A View from the Bench.” Journal of International
Economic Law 3(3):395–401.
Laver, Michael. 1998. “Party Policy in Britain 1997: Results from an Expert Survey.”
Political Studies 46(2).
24
Mansfield, Edward D. & Eric Reinhardt. 2008. “International Institutions and the Volatility of International Trade.” International Organization 62(4):621–652.
Mansfield, Edward D. & Eric Reinhardt. 2009. “International Institutions and Terms of
Trade Volatility.” Working paper.
Marks, G., L. Hooghe, M.R. Steenbergen & R. Bakker. 2007. “Crossvalidating data on
party positioning on European integration.” Electoral Studies 26(1):23–38.
Milner, Helen & Tim Büthe. 2008. “The Politics of Foreign Direct Investment into
Developing Countries: Increasing FDI through International Trade Agreements.”
American Journal of Political Science 52(4):741–762.
Moravcsik, Andrew. 1998. The Choice For Europe. Ithaca, NY: Cornell University Press.
Panagariya, Arvind. 1999. “The Regionalism Debate: An Overview.” The World Economy 22(4):455–476.
Rosendorff, B. Peter. 2005. “Stability and Rigidity: Politics and Design of the WTO’s
Dispute Settlement Procedure.” American Political Science Review 99(3):389–400.
Smith, James McCall. 2000. “The Politics of Dispute Settlement Design: Explaining
Legalism in Regional Trade Pacts.” International Organization 54(1):137–180.
Steenbergen, Marco R. & Gary Marks. 2007. “Evaluating Expert Surveys.” European
Journal of Political Research 46(3):347–366.
Stinnett, Douglas. 2007. “Depth, Compliance, and the Design of Regional Trade Institutions.” Paper presented at the annual meeting of the Midwest Political Science
Association, Palmer House Hotel, Chicago, IL, April 12, 2007.
Stone Sweet, Alec & Thomas Brunell. 1998. “Constructing a Supranational Constitution: Dispute Resolution and Governance in the European Community.” American
Journal of Political Science 92(1):63–81.
25
Whitefield, Stephen, Milada Anna Vachudova, Marco R. Steenbergen, Robert Rohrschneider, Gary Marks, Matthew P. Loveless & Liesbet Hooghe. 2007. “Do expert surveys
produce consistent estimates of party stances on European integration? Comparing
expert surveys in the difficult case of Central and Eastern Europe.” Electoral Studies
26:50–61.
26
100
Figure 1: The Relationship Between an Agreement’s Scope and the Quality of the Secretariat
EU
80
CIS
SCO
OECS
ECOWAS
60
EAEC
COMESA
UEMOA
CEFTA
APEC
Scope of
Agreement
40
ECCAS
CEMAC
Mercosur
GCC
NAFTA
SADC CARICOM
EFTA
20
ANDEAN
SACU
ASEAN
0
MSG
SAFTA
Bangkok
SPARTECA
0
20
40
60
Quality of Secretariat
27
80
100
Table 1: Expert Survey: Descriptive Statistics for RTAs
Number of Mean
RTA
Responses Score Minimum Maximum
AMU
39
3.59
1
10
ANDEAN
83
4.65
1
10
APEC
73
4.11
1
9
ASEAN
258
5.23
1
10
Bangkok
70
1.84
1
9
CACM
33
5.27
4
7
CARICOM
224
6.41
1
10
CEFTA
98
7.05
2
10
CEMAC
48
5.36
1
10
CIS
60
4.65
1
10
COMESA
47
7.81
2
10
China-ASEAN
10
3.70
2
7
EAC
25
7.44
1
10
EAEC
21
5.00
1
8
ECCAS
49
5.52
1
10
ECOWAS
82
6.68
1
10
EEA
21
8.10
1
10
EFTA
278
7.35
1
10
EPA
41
7.46
1
9
EU
445
8.97
1
10
GAFTA
36
5.61
1
10
GCC
169
6.17
1
10
India-Singapore
2
2.00
2
2
Japan-Mexico FTA
3
5.00
1
10
KORUS FTA
7
7.14
5
10
Korea-Chile FTA
27
5.59
2
10
LAIA
34
2.88
0
10
MRU
8
1.50
1
5
MSG
62
5.50
1
10
Mercosur
201
5.78
1
10
NAFTA
223
7.17
1
10
OECS
83
7.15
1
10
PICTA
34
6.21
1
9
SACU
100
6.29
1
10
SADC
138
6.45
1
10
SAFTA
63
4.29
1
10
SCO
21
3.81
1
9
SPARTECA
48
2.19
1
9
Singapore-Japan
2
2.00
2
2
UEMOA
74
6.60
1
10
28
29
Ability to Achieve Goals
Ambition
Enforcement
Promotes Trade
Scope
Reduces Tariffs
DSM
Monitoring
Legalization
Secretariat
Ability
1
0.39
0.80
0.65
0.03
0.58
0.73
0.80
0.47
0.70
1
0.41
0.74
0.61
0.65
0.52
0.53
0.64
0.73
1
0.74
0.23
0.70
0.89
0.86
0.60
0.71
1
0.15
0.92
0.81
0.80
0.72
0.80
1
0.02
0.19
0.22
0.25
0.44
1
0.79
0.77
0.76
0.74
1
0.90
0.71
0.75
1
0.71
0.84
Table 2:Enforcement
Correlation
Matrix of
Dimensions
of RTAs
Promotes Trade
Scope
Reduces Tariffs
DSM
Monitoring
Ambition
1
0.70
Legalization
1
Secretariat
Table 3: The Effect of Private Rent Environment on Inefficient Bureaucracy∗
1
2
3
4
Constant
2887.00*** 5181.81*** 2419.67*** 3885.68***
(957.85)
(1371.26)
(734.73)
(1182.94)
Corruption (ICRG) 170.18*** 216.59***
(61.57)
(113.72)
Corruption (TI)
73.35***
38.49***
(32.67)
(13.25)
GDP
-147.70*** -218.54*** -120.93*** -153.82***
(41.49)
(54.13)
(31.22)
(39.560)
Public
-3.25
-30.51*
-2.74
-22.09*
Employment
(3.39)
(14.61)
(3.22)
(10.47)
δ Intra-RTA
2839.23
3584.04
2904.39
4156.81
Trade Volumes
(1729.21)
(2671.20)
(1650.45)
(2075.15)
δ Tariffs
1729.11
1832.29
1738.21
1792.87
Within RTA
(1423.42)
(1185.21)
(1074.23)
(1023.94)
Legalization
-96.703
-129.11
(81.31)
(104.88)
N
32
17
34
19
2
R
.56
.80
.74
.89
∗
Dependent variable is aggregated expert survey rankings of the scope of agreements
divided by the quality of RTA secretariats. OLS regression. Standard errors in
parentheses. * p<.10, ** p<.05, *** p<.01
30
Table 4: Budget Data: Descriptive Statistics for RTAs
No. of Members
No. of Staff
Salaries
Honoraria
Other
Consultants
Compensation
Support
Personnel
Benefits
Transportation
for Personnel Training
Recruitment
Personal
Expenses
Travel
Expenses
Building Rent
Building
Maintenance
Building Services
Vehicles
Furniture
Office Insurance
Infrastructure
- Misc
Supplies
Secretariat Meetings
Operational expenses- other
Programs:
Workshops
Programs:
Studies
Programs:
Compensation
Programs:
Publications
Programs: Other
Other
Grand Totals
Andean Comm
5
90
2,348,111.00
11,508.06
256,062.30
.
COMESA
23
110
1,122,580.00
84,583.54
1,453,680.00
423,199.30
LAIA
12
45
2,268,210.00
555,103.30
155,492.20
241,919.40
Mercosur
4
18
609,700.00
.
17,600.00
.
SADC
15
77
869,043.30
8,265.82
39,107.00
14,761.27
371,199.80
267,550.20
34,700.00
44,799.27
748,892.50
878,074.30
.
.
.
.
421,593.40
71,162.83
8,441.50
298,104.00
.
487,123.30
.
37,589.38
.
26,013.76
52,309.73
430.00
5,754.46
.
.
1,200.00
153,338.90
1,297.29
78,285.33
.
266,281.50
181,865.60
63,563.33
86,657.45
77,189.39
58,503.64
54,166.97
42,781.34
6,275.13
119,270.60
87,720.56
.
38,009.73
91,740.24
25,395.69
203,016.60
20,883.33
81,890.22
23,662.08
46,968.36
32,504.29
48,862.11
.
.
1,264,181.00
89,611.11
5,361.11
.
9,321.43
94,805.56
28,157.27
.
9,700.27
1,324.27
30,667.27
18,079.13
1,972.07
.
.
144,012.50
68,957.89
64,169.81
218,386.20
.
37,264.99
903,769.80
4,030,692.00
.
147,177.80
123,966.70
313,953.80
.
40,977.45
2,520.00
23,325.09
.
2,474.36
3,453.87
1,994,180.00
210,632.80
.
.
.
.
.
.
775,033
10,800
.
54,085
35,497
11,131
.
.
3,884
123,623
25,245
5,156,209
1,701,265
1,760,769
14,900,000
146,000
95,484
4,761,575
.
1,243.09
993,175
4,514,403
99,840
9,155,535
31
Table 5: Predictors of Organizational Patronage - Disproportionate Salaries∗
5
6
7
8
Constant
5.18*** 8.29*** 4.18*** 3.17***
(2.65)
(3.42)
(1.12)
(1.98)
Corruption of Members (ICRG) 4.18*** 2.22***
(average)
(244)
(1.84)
Corruption of Members (TI)
5.89*
3.28**
(average)
(2.91)
(1.61)
GDP of members
-3.13
-1.28
-1.01
-.85
(average)
(3.78)
(2.31)
(.91)
(.56)
Public
1.28
2.01
1.56
1.97
Employment
(1.56)
(1.88)
(1.78)
(2.13)
δ Intra-RTA
.12
.19
.17
.16
Trade Volumes
(.13)
(.18)
(.14)
(.09)
δ Tariffs
.04
.02
.04
.03
Within RTA
(.04)
(.04)
(.05)
(.05)
Legalization
.12
.21
(.07)
(14)
N
96
96
96
96
2
R
.31
.33
.39
.40
∗
Dependent variables are average salaries in the international organization as a ratio of
average local salaries for central government employees. Likelihood ratio test of
independent equations (ρ = 0): χ2 (1) = 0.39 Prob > χ2 = 0.53. * p<.10, ** p<.05, ***
p<.01
32
Table 6: Predictors of Organizational Patronage - Program Budgets∗
9
10
11
12
Constant
1.40
-6.90*** 2.81 -4.78*
(3.95)
(2.39)
(1.97) (2.21)
Corruption of Members (TI)
-.65***
-.24*
(average)
(.21)
(.13)
Corruption of Members (ICRG)
-.31* -.21*
(average)
(.16)
(.10)
GDP of members
-.13
-.21
-.17
-.23
(average)
(.24)
(.16)
(.11)
(.16)
Public
.16
.15
.18
.13
Employment
(.12)
(.09)
(.11)
(.09)
δ Intra-RTA
.03
.02
.04
.03
Trade Volumes
(.05)
(.02)
(.05)
(.02)
δ Tariffs
Legalization
N
R2
∗
.001
(.02)
(.02)
96
.26
.0003
(.006)
.01
(.05)
96
.27
.0001
(.04)
-
.002
(.008)
.03
96
.28
96
.27
Dependent variable is programs, honoraria, consultant expenses, travel expenses, and
personal expenses as a share of operations. * p<.10, ** p<.05, *** p<.01
33