INTRODUCTION
Risk assessment is an essential component in supply chain planning.
While newfound promises from globalization have ushered in a vast
network of countries actively pursuing global trade, uncertainties
remain. As supply chains spread globally, risks of operational
disruptions become costlier and less controllable. Operational risks
abound: maritime attacks in the Gulf of Aden, H1N1 in Mexico, SARS in
Hong Kong, currency crisis and supplier insolvency in Argentina, credit
meltdown in Iceland--just to name a few. These are risks that supply
chain managers consider in making offshoring and trade-partnering
decisions across countries. Consider that even a reputed supplier in
Pakistan will fail to secure a contract because of the operational risks
from a fragile government and the Taliban insurgency in the Swat valley.
In short, the country offers an institutional aegis for any firm intent
on participating in the global economy. Unless country-level assurances
are available, firms would shy away from trading with clients and
suppliers based in a country with high operational risks. Such
assurances include the rule of law, the element of due process, an
established court system, intellectual property protection, financial
liquidity, and currency stability, among others. Perhaps that has been a
reason why firms from certain countries establish corporate (not
operational) headquarters in stable Western economies. As Bryan Squibb,
managing director of Aon Trade Credit, rightly notes,
"understanding the nature of supply chain risk exposures, and where
they occur most frequently, is now a board-level priority" (Supply
Chain Digest 2006).
Operational risks and third-party indices constitute the core of
our discussion. Extending the Basel (Basel Committee on Banking
Supervision [BCBS 2008] definition, we define operational risk in
commercial supply chains as the calculated loss resulting from adverse
shifts across one or more activities in a global supply chain owing to
inadequate or failed assessment of processes, people, systems, and
external events that can disrupt the commercial flow of information and
materials. In global supply chains, operational risks encompass
disruptive threats tied to a trading partner's country of origin.
These operational risks stem from governance problems, geographic
separation and cultural gaps between trading partners, and bipartite
limitations of communications in information transparency (Aron, Clemons
and Reddi 2005). In short, a lack of arms-length transactions, imperfect
and asymmetric information, risk of the loss of intellectual property,
costs of monitoring reliability, quality and threats, and operating in
regulatory mazes are just a sample of the myriad economic,
technological, sociocultural and political uncertainties that can affect
the performance of global supply chains. Therefore, an improved
understanding of country-level operational risks is essential as a
prelude to global sourcing and trading.
Operational risks stymie import and export supply chains by the
dosing of or tight control and quarantine over incoming goods and
services via air, land or water. Note that Canadian trade suffered a
loss of US$515 million in 2003 from SARS; Avian flu led to a loss of
US$10 billion to South-East Asian economies; and Mexico continues to
lose US$57 million per day from operational threats related to the H1N1
virus (Shah 2009). In fact, 2 days after the H1N1 virus was defined as a
pandemic, the United States closed all trucking routes to and from
Mexico. Facing growing uncertainties across stretched commercial supply
chains, third-party indices can aid managers in making important supply
chain decisions, from contracting to selecting suppliers across the
globe. These indices can complement internal supplier assessments by
providing firms with more complete information as a prelude to supplier
selection, thereby supplementing existing efforts in supply chain risk
management.
As assessment yardsticks, measures for the internal environment of
a country can provide crucial information as harbingers to operational
risks. Third-party indices provide a multifaceted view of a
country's internal environment (Aron et al. 2005). These
third-party indices provide numerical rating measures for the purposes
of comparative assessment and decision-making as precursors to global
supply chain planning. Moreover, these indices are developed by
independent third-party organizations and agencies espousing unbiased,
universal affiliations.
Merits of these third-party indices have been emphasized in circles
of practice and policy. For example, information and financial
transparency indices have been linked to global trade to an extent that,
during volatile economic conditions, companies rush out of highly opaque
countries (Gelos and Wei 2006). Take, for example, the rampant maritime
piracy in the Gulf of Aden and how that has disrupted commercial supply
chains. Ongoing disruptions and operational risk considerations are
forcing companies to restructure contracts and re-envision supply routes
by foregoing the Suez Canal to sail around the Cape of Good Hope. This
increases costs and lead times by adding approximately 5,500 miles and
20 days. Yet, central to the supply chain disruption from piracy is the
political instability in Somalia, best highlighted by third-party
indices.
While policy groups (e.g., agencies such as UNDP and World Trade
Organization [WTO]) and major firms rely on third-party indices to aid
supply chain-related decision-making and policy formulation, these
indices have received little attention in supply chain research.
Nonetheless, in a flat and uncertain world, it is useful for
practitioners and researchers to enquire: to what extent do third-party
indices explain operational risk in countries, and what consequences do
operational risks have on the "volume" of a country's
import and export supply chain? In answer to our enquiry, we explore an
empirical framework (refer to Figure 1 for a preliminary overview) and
empirically test our contentions using archival data (1) collected on 81
member countries of the WTO, gathered from UN agencies, independent
think tanks, the WTO and the Economist Intelligence Unit (EIU).
[FIGURE 1 OMITTED]
The remainder of this paper is organized as follows:
"Theoretical Underpinnings" discusses the theoretical
underpinnings leading to our proposed model. "Conceptual Framework
and Hypotheses" develops hypotheses core to the proposed model.
"Research Design, Analyses, and Results" discusses the
research design, analyses and results. "Discussion and
Contribution" concludes the study with a discussion of the results
and contributions of this study
THEORETICAL UNDERPINNINGS
We use Transaction Cost Economics (TCE) (Williamson 1985) as the
theoretical underpinning for examining underlying governance mechanisms
and assessing operational risks across global supply chains. TCE
specifically draws attention to the cost of contracting, monitoring and
safeguarding under the threat of opportunism--a trait where parties
leverage information asymmetry to the disadvantage of the other (buyer
or supplier) (Williamson 1985; Rindfleisch and Heide 1997; Van
Oosterhout, Heugens and Kapstein 2006; Datta and Chatterjee 2008).
Because risks from opportunism increase transaction costs (tighter
contracts, heightened monitoring), gauging the risks of doing business
with a country is essential (March and Shapira 1987; Lee and Billington
1992; Landeghema Van and Vanmaele 2002; Cavinato 2004; Enyinda, Ogbuehi
and Briggs 2008). Threats of opportunism and information asymmetry
further contribute to uncertainty, noted as "neither ignorance nor
complete and perfect information but partial knowledge" (Knight
1985, p. 199; Datta and Chatterjee 2008). In the scope of our
discussion, assumptions underlying TCE surface the costs of transacting
under threats of opportunism from countries participating in global
supply chains. Thus, under assumptions of information asymmetry and
opportunism, firms need to rely on credible sources of information that
can alleviate transaction costs by offering crucial and (relatively)
complete information on transacting with (parties operating in/from) a
country.
Williamson (2008) further establishes the relevance of TCE in the
supply chain context. Williamson's research focuses on transactions
in a supply chain as the fundamental unit of analysis. Transactions are
grounded in contracts involving more than one party across the supply
chain. Williamson (2008) points out that supply chain transactions are
driven by opportunism and self-interest that capitalizes on information
asymmetry. An opportune party in the supply chain transaction will try
to create complex contracts that are incomplete or prohibitively
difficult to monitor in global supply chains (Lamming, Caldwell,
Harrison and Phillips 2006). As such, the prevalent market (buy)
contracting in commercial supply chains has to assume that opportunism
exists and that the other party will leverage its position on incomplete
and asymmetric information. After all, "the market-mode features
high-powered incentives, little administrative control and a legal-rules
contract law regime, which combination is well suited to implement
autonomous adaptations but poorly suited to effect cooperative
adaptations" (Williamson 2008, p. 8), In the scope of our
discussion, countries and market suppliers originating from a country
will tend to portray asymmetric and incomplete highlights of their
country's environment to secure contracts.
Apart from peremptory "muscular" buyers who can create
stringent contracts to their own advantage, benign and credible buyers
are likely to make suboptimal decisions in relation to their supply
chain contracting (Williamson 2008). As implied by Williamson, in
commercial supply chains, information asymmetry about a country can lead
to unintended operational risks from unforeseen contingencies, myopia in
contracting and undue supply chain complexities. The calculus of
credibility therefore requires firms to rely less on the supplier's
promotion of their country and more on independent third-party
assessments of a country for hazard mitigation, transparency, relatively
credible commitments (2) and most of all, reducing information asymmetry
in an attempt to mitigate opportunism.
Opportunism based on information asymmetry is a form of deception.
In market transactions, opportune parties in
"business-seeking" countries will try to capitalize on the
bounded rationality of the other party by revealing incomplete
information to mislead or deceive the party into contracting with them.
For example, in 2008, Argentina was providing deceptive inflation
statistics by removing products whose prices rose too sharply from its
consumer price index (CPI). As a result, the Argentine government
misinformed (deceived) international stakeholders by showing a 0.6
percent inflation when, in reality inflation was in the double digits
(The Economist 2008). While the Argentine Central Bank engaged in
underreporting real inflation for over two years, third-party statistics
(e.g., INDEC) signaled the misinformation. If firms were to transact on
the basis of the inflation rates offered by the Argentine government,
their supply chain transactions would be contracted on deceptive
premises. Given the amorphous nature of self-reported measures by
countries and suppliers to obtain contracts, reliable third-party
information can mitigate deception in competitive environments (Spence
2002). Therefore, for a third-party index to be credible, it has to
provide "reliable" information cues.
Reliable information from third-party indices offers tremendous
decisional benefits in the face of information asymmetry across various
operational facets--including supply chain transactions and
coordination. Reliable information gained from the indices needs to be
more than redundant and reinforcing--it must credibly fill the
information gap created by asymmetric and incomplete information.
Therefore, because third-party indices claim to offer reliable
information to mitigate uncertainty from information asymmetry and
opportunism, their reliability and credibility become core to assessing
operational risks in countries.
CONCEPTUAL FRAMEWORK AND HYPOTHESES
We begin this section with an illustration of the relationship
between a country's internal environment, (3) operational risk and
trade. This paper defines internal environment of a country in light of
(i) information environment (transparency), (ii) business environment
(relational continuity, economic freedom, and financial standards and
compliance) and (iii) socioeconomic environment (political rights and
liberties, governance performance and economic growth).
Information Environment and Operational Risk
The information environment of a country refers to information
transparency related to governance of its internal environment.
Countries with a transparent information environment allow open access
to information, have fewer restrictions on information dissemination,
take significant steps to bridge the digital divide and publish accurate
and reliable information.
As a dimension of transparency, corruption is closely tied to
operational risk. Rose-Ackerman (2006) remarks on bid rigging, tender
forgeries, political pressures and rampant bribery as serious
operational risks and impediments to trade. Similarly, in examining
operational practices among utility companies in South America, Bo and
Rossi (2007) find that corruption adversely affects operations and
limits economic progress, noting that "corruption may ultimately
determine the viability and health of business in any given
country." Timeliness, periodicity and coverage of disclosures
relate to operational risks by continuously conveying useful and
complete information on time. Gelos and Wei (2006) find that
transparency from disclosures allows markets to react on time. Moreover,
disclosures mitigate information asymmetry and reduce operational
concerns and risk.
Finally, given that countries have unique deduction techniques, a
transparent assessment of the way data has been collected and analyzed
toward the disclosures is equally important. For example, the Argentine
government decided to remove sharply rising item prices in its CPI to
understate its inflation. Similarly, at the core of Dubai's current
liquidity crisis is Dubai promoting itself as a haven for opening
businesses while obfuscating the hassles of closing businesses where
"any creditor foolhardy enough to test the regime can expect to
recover just ten cents on the dollar" (The Economist 2009). Given
that countries use unique deduction techniques to conveniently promote
themselves, transparency of the deduction techniques (including data
collection, analysis and information dissemination) and the information
environment is important. This information allows interested parties to
gauge the accuracy, validity and reliability of the information. Unless
reliability of disclosed information can be assessed, operational risks
remain from a lack of assurance on a country's internal
environment. Therefore, we argue:
H1: Countries with more transparent information environments will
have less operational risks than countries with less transparent
information environments.
Business Environment and Operational Risk
The business environment of a country aims at maximizing
sustainability (economic freedom and relational continuity) and
creditworthiness (financial standards and compliance) of a
country's internal environment to reduce operational risks.
Economic Freedom. Economic freedom is the fundamental right of
every human to control his or her capital, labor, property and
enterprise. In economically free societies, governments allow labor,
capital and goods to flow freely (Heritage Foundation 2009). A decade
ago, Colombia was a nation in turmoil; ruled by drug kingpins,
narcofunded leftist terrorists and guerrillas and far-right
paramilitaries. Today, with the "Build Colombia" economic
freedom program, Columbia trains employers in proactively ensuring
compliance with all Colombian labor laws on occupational safety, child
labor, working hours and other issues of concern to Colombian workers.
The International Labor Organization's (ILO) report on the
Colombian government's push toward economic freedom breathed new
life into Columbian trade (Griswold and Hidalgo 2008; Roberts 2009) as
investors perceived less operational risks. We thus posit the following:
H2: Countries with more economic freedom will have less operational
risk than countries with less economic freedom.
Relational Continuity. A fundamental premise of maintaining a
sustaining or going concern is a set of good business rules overseen by
an efficient regulatory system that is accessible to users and simple to
understand. Relational continuity establishes sustainable business
development in a country by considering aspects related to the
incorporation, operation and dissolution of business as one proxy for
assessing a country's internal environment related to economic
governance (Business Indicator Index 2009). For example, in Norway, it
takes five procedures, 5 days and 2.6 percent of annual income per
capita in fees to open a business. In contrast, Argentina is a paradox.
"The failure of Argentina, so rich, so under-populated, twenty
three million people in a million square miles, is one of the mysteries
of our time," wrote the Nobel Laureate V.S. Naipaul. It takes 15
procedures and little more than a month to open a business in Argentina!
Thus, when all economies under the WTO are ordered by their average
percentile ranks, Norway is in 10th place, while Argentina is a distant
111th (Business Indicator Index 2009)--increasing perceptions of
operational risks from bad economic governance. Therefore, we argue:
H3: Countries with higher relational continuity will have less
operational risk than those with lower relational continuity.
Financial Standards and Compliance. Stability of capital markets is
a prerequisite for operations. Failing financial standards and
compliance, including liquidity crises, mismanaged money supply and
capital flow manipulation axe serious operational risk concerns for any
firm intending to establish business relationships in a country.
The index of financial standards and compliance gauges a
country's internal financial environment constituting its (i)
reporting standards, (ii) banking system and (iii) regulatory system. A
country failing in one or more of the aforementioned aspects creates a
risky operational proposition for business. Consider wanton financial
deregulation in Iceland. Iceland, with little regard to its currency,
the Krona, had committed itself to borrowing and investing heavily in
other international currencies--making loans 9-10 times the size of
their own Gross Domestic Product (GDP). Lax regulations, coupled with
marginal reporting and compliance led to the mortgage meltdown, bringing
Iceland to the brink of bankruptcy and overthrowing the government
behind the financial deregulation (The Economist 2008). The financial
meltdown further led to a scuttled pullout of businesses--fearing risks
of operating businesses without credit availability, in financial
turmoil and with uncertainties surrounding money flow.
A global economy hinges on standards related to financial
transactions and reporting that offers a degree of accounting
transparency and ensures accurate information related to a
country's financial environment. Similarly, a stable banking system
with well-defined regulations effectuates free flow of capital. Thus,
financial standards and compliance symbolize a country's
creditworthiness to its investors, and increased creditworthiness
reduces country-specific operational risk. We thus posit:
H4: Countries with higher financial standards and compliance will
have less operational risk than countries with lower financial standards
and compliance.
Socioeconomic Environment and Operational Risk
The socioeconomic environment of a country aims at establishing
freedoms of expression, association and beliefs among its citizens,
thereby enhancing governance performance to stimulate economic growth
and reduce operational risks.
Political Rights and Liberties. Freedom House (2009) defines
political rights and liberties in the light of a democratic political
system in which the government is accountable to its own people, the
rule of law prevails and freedoms of expression, association and belief,
as well as respect for the rights of minorities and women are
guaranteed. For example, companies consider it risky sending their
female executives to countries with laws requiring women to wear burkhas
(eg., Saudi Arabia) or chadors (eg., Iran). The freedom of press in
Pakistan is a case in point. In November 2007, Pakistan ex-President
General Pervez Musharraf declared a state of emergency, suspending the
constitution and dismissing the chief justice. The political unrest
stemming from a state of emergency deterred businesses from setting up
operations in Pakistan. In short, doing business in a region without the
freedom of speech, respect for humanity and presence of constitutional
(rather than religious) law is often deemed risky. Thus, we contend:
H5: Countries with more political rights and liberties will have
less operational risk than countries with less political rights and
liberties.
Economic-Political Governance Performance. Economic-political
governance performance refers to sustainable economic development
supported by a solid institutional framework necessary to fight poverty,
establish equal opportunity and thus compensate for harsh social
conditions (Bertelsmann 2008).
The Bertelsmann Transformation Index ties overcoming absolute
poverty to a more secure market-based democracy. Democratic norms of
accountability help protect the viability and legitimacy of the economic
system from being undermined by distributional conflicts among social
and economic actors that can increase disparity and poverty. A measure
of absolute poverty quantifies the number of people below a poverty
threshold that is independent of time and place. For the measure to be
absolute, the line must be the same in different countries, cultures and
technological levels. World Bank draws the International Poverty line at
US$1.25/day; based on this measure, 42 percent of Indians live below the
poverty line, the highest rate in South Asia. Because the ratio of per
capita consumption to per capita population equates to poverty; it comes
as no surprise that the population of India is a beast to reckon with in
this regard. However, India's efforts toward social justice and
equal opportunity have been regarded as a positive strand in its
economic-political governance. Better economic-political governance
creates a conductive operational environment that can, over time, raise
the overall living standard and alleviate poverty. This allows us to
hypothesize:
H6: Countries with higher levels of economic-political governance
will have less operational risk than countries with lower levels of
economic-political governance.
Economic Growth. Economic growth is closely tied to economic
competitiveness (Porter 2000) and is more than a mere increase in GDP.
Economic growth reflects the economic productivity of a country over
time, typically measured using indicators such as real GDP forecast,
unemployment, education and health (World Economic Forum 2008). Economic
growth has to be calculated in real, inflation-adjusted terms such that
the effect of inflation on economic performance is accurately
discounted.
A case in point is the economic growth in Bangladesh from the
period of 1947-71. Initially founded as East Pakistan in 1947, political
unrest, inflation, severe unemployment, lack of birth control and a
largely illiterate workforce have plummeted economic growth and created
an operational quagmire. Malaysia, on the other hand, with its
established educational institutions, industrial development and
healthcare initiatives, created an environment conducive to
international business operations. As a result, Malaysia enjoys economic
growth relatively higher than its neighbors. Thus, we contend that
higher economic growth enhances stability and reduces operational risk.
H7: Countries with higher economic growth will have less
operational risk than countries with lower economic growth.
Operational Risk and a Country's Import and Export Supply
Chains
We use the terms import supply chain and export supply chain to
point to the volume of merchandise imported by and exported from a
country. We argue that the volumes of country's import and export
supply chains are best manifested in its trade patterns to and from its
shores (notwithstanding its internal supply chain), respectively. The
perception of high operational risk of a country limits other countries
from trading with it. There are risks of insolvency (e.g., Iceland,
post-financial meltdown), embargoes (e.g., North Korea, Cuba) and unfair
expropriation (e.g., Zimbabwe), among others. The sampled examples point
to the fact that operational risks, at the very least, chafe the
patterns of trade flows in countries. Arguing that volumes of
merchandise imports and exports reflect the "volumes" of
import and export supply chains, respectively, operational risks can
truly impede a country's import and export supply chains Figure 2.
[FIGURE 2 OMITTED]
The growth of the Internet has increased information dissemination
and the slightest hint of operational disruptions can risk immediate
spread across the globe and sway business decisions. From disease
outbreaks and credit constraints to electoral violence, terrorist
threats and coercive posturing, operational risk information constantly
shapes supply chain decisions.
While a country's operational risks have a negative bearing on
the volume of its import and export supply chains, it is interesting to
note that the effect is unequal. Consider Mugabe's 2000 coercive
land distribution reform that triggered the country's operational
risk (Clemens and Moss 2005). Robert Mugabe, the president of Zimbabwe,
misused Zimbabwe's 1992 Land Reform Act to nationalize the farmland
and coercively remove existing landowners without compensation or legal
recourse. The land was badly parceled and redistributed to government
members and their families who had no training or interest in
agricultural operations. In a short period, the "bread basket"
of southern Africa plummeted into starvation and famine. Zimbabwe, the
sixth-largest producer and exporter of tobacco, a cash crop, fell to its
lowest levels in 50 years. Investors perceived operational risk and
withdrew their capital, original landowners migrated and the government
forced banks to lend to the new, untrained landowners with high default
risk. Risks of operations affected Zimbabwe's internal production
acutely. Corn production, Zimbabwe's largest export produce that
had helped build its agrarian supply chain, was approximately halved in
a few years. Even more astounding is that Zimbabwe, once a major
exporter of corn, became a net importer (Madslien 2008). Operational
risks often choke internal business and production, some of which are
intended for the country's internal market. If operational risks
deter supply to the internal market, the country has to rely on
importing merchandise to serve its population but fails to secure
international buyers who abstain due to operational risks. A
country's export supply chain therefore remains more sensitive and
vulnerable to internal operational risk shocks than its import supply
chain. Thus, we hypothesize:
H8: Higher operational risks negatively impact a country's
import and export supply chain volume.
H9: Operational risks will affect a country's export supply
chain volume more adversely than the country's import supply chain
volume.
Control Variables
For this study, we have chosen inflation and GDP as our control
variables. Core inflation is measured by the CPI. The measured inflation
rates, coupled with uncertainties in future demand conditions and
control measures, directly impact trading (Bauer, Haltom and Peterman
2004). High inflation typically leads the government to induce measures
that prove to be detrimental to trade. A case in point is Argentina
where, in late 2005, a 6-month ban on beef exports and higher
agricultural export taxes were designed to boost local supplies. On the
other hand, since country-specific GDP is widely used as a source of
information in all the third-party indices used in this paper, its
inclusion as a control variable comes as no surprise.
RESEARCH DESIGN, ANALYSES, AND RESULTS
We empirically investigate 81 member countries of the WTO using
archival data collected from UN agencies, independent think tanks, the
WTO and the EIU. We use a host of seven third-party indices (Tables Ia
and Ib) to gauge a country's internal environment, map those
indices to corresponding country-specific operational risk (data
obtained from EIU) and investigate the consequent effects of operational
risk on trading volume. Our sample of 81 countries out of the 153 WTO
member nations is approximately 52 percent of the WTO members. We have
10 countries from Africa, 18 from Asia, 9 from Eastern Europe, 7 from
the Middle East, all 3 countries in North America, 12 from South America
and 22 countries from Western Europe. It is thus safe to assume that our
dataset is neither regionally nor geographically biased. The sample is
representative of the WTO member set and, hence, unlikely to taint any
inferences in our study. The country list can be found in the Appendix.
We start off the discussion in this section with a detailed
framework of the various indices used, and their respective mapping with
the constructs conceptualized in the previous section (Tables Ia and
Ib). (4)
Consider the measurement for a country's information
environment in Table Ia. We define information environment of a country
in terms of the transparency of information interchange. Since its
inception in 1993, Transparency International, a global civil society
organization, has been leading the fight against corruption.
Transparency International's Corruption Perceptions Index measures
the perceived level of public-sector corruption in 180 countries and
territories around the world. The corruption perception index is a
"survey of surveys," i.e. a meta survey based on 13 different
expert and business surveys (Table Ib). Transparency International ranks
these countries based on their timeliness of disclosure, periodicity and
coverage of information, level of corruption, and the quality of
statistics used to support these standards.
This study does not suffer from common method bias given the
multiple sources of independent archival data collected via different,
comparative methods (refer to Table Ib; Campbell and Fiske 1959). Given
that the data originate from independent sources and observations, there
is no propagated bias from the use of a common method to assess the
antecedents and outcomes. The third-party indices used in the analysis
are world bodies with global data, and are hence devoid of any country
specific bias. The country-specific operational risk data were collected
from fact sheets of the EIU, a premier global source of economic data.
Finally, we used merchandise exports and imports as proxies for the
volume of export and import trade volumes of a country; the data were
obtained from fact sheets of the WTO.
Because the third-party indices have different scaling systems (as
an example, the Heritage Foundation Economic Freedom Index uses a
100-point scaling system, while both the Freedom House Index and World
Economic Forum Global Competitive Index use a seven-point scale), we
normalize all variables in our analysis into a 10-point scale. To
diminish the bias of export or import volume due to size of the country
itself, we normalize the raw merchandise export/import data by taking a
ratio of the per capita export/import to the per capita GDP of
respective countries.
Choice of a Statistical Method
This research is a novel exploration of supply chain challenges in
risky global settings. In exploratory research settings, partial least
squares (PLS) regression has been well established as a suitable
component-based modeling technique (Hulland 1999; Diamantopoulos and
Wink-Ihofer 2001; Hsu, Chen and Hsieh 2006). In particular, we find PLS
to be appropriate because of the following:
First, PLS is fitting for novel exploratory studies where the
research aims to surface relationships rather than confirm theory.
Covariance-based structural equation models (SEMs) such as LISREL or
AMOS are more suited for theory confirmation where the focus is on
parameter estimations.
Second and more importantly, in exploratory studies where the
multivariate model implies logically common causes and correlated
errors, PLS is more appropriate than covariance-based SEMs. While there
are methods such as full-information maximum likelihood (FIML) and
three-stage least squares (3SLS) that consider cross-regression
correlation in order to increase efficiency, they are partly inoperable
because of their failure to provide a k-step estimation procedure when
estimating multiple equations with logically common causes and
correlated errors. While this study has sample size limitations (n=81),
it is not the primary driving logic for using PLS. Although studies have
often cited sample size limitations as the reason to choose PLS (e.g.,
Floe 2008), the appropriateness of PLS in this study centers on the
presence of logically common causes and correlated errors.
Third, the PLS technique is not limited by distributional
restrictions on data while covariance-based SEMs require multivariate
normal distribution and can suffer from factor indeterminacy problems
and inadmissible solutions (Hsu el al. 2006).
We thus report results from a combined analysis using both PLS
regression as well as ordinary least squares (OLS) regression on our
dataset. While the OLS platform provides robustness and seeks linear
functions of the predictors in our mode) to explain as much variation in
each response as possible, the PLS platform adds to the accountability
when the predictors are correlated. In regards to establishing the
effect of mediation, we use the theories put forward by Judd and Kenny
(1981) and Baron and Kenny (1986) to test the mediation effects (5) of
operational risk on the relationship between a country's internal
environment (based on third-party indices) and its volume of import and
export supply chains.
Test of Assumptions
This study assumes linearity (6) between the antecedent and outcome
variables. Following Neter, Wasserman and Kutner (1990), the two
assumptions of linearity are that (i) our dependent variable (Y),
Operational Risk, is a continuous variable and (ii) the independent (X)
variables used in our model are fixed (the values of X are measured
without error; this is a valid assumption based on the authenticity of
data sources for our indices). Therefore, it is appropriate to use a
linear specification to test our hypothesis.
As noted, a linear regression technique using ordinary least
squares serves the purpose of our study. We use the following model
specifications to test our hypotheses with corresponding variable
definitions explained within parentheses in Table 1.
Operational Risk = [[beta].sub.0] + [[beta].sub.1]TICRIndx +
[[beta].sub.2]HFEFIndex + [[beta].sub.3]BIIndex + [[beta].sub.4]Moody +
[[beta].sub.5]FHIndex + [[beta].sub.6]BTSIndex +
[[beta].sub.7]WEFGCIndex + [epsilon] (1)
where [[beta].sub.0] is the Intercept, [[beta].sub.1] is the
Population estimate for Transparency (TICRIndx), [[beta].sub.2] is the
Population estimate for Economic Freedom (HFEFIndex), [[beta].sub.3] is
the Population estimate for Relational Continuity (BIIndex),
[[beta].sub.4] is the Population estimate for Financial Standards and
Compliance (Moody), [[beta].sub.5] is the Population estimate for
Political Rights and Liberties (FHIndex), [[beta].sub.6] is the
Population estimate for Governance Performance (BTSIndex),
[[beta].sub.7] is the population estimate for Economic Growth
(WEFGCIndex), [epsilon] is the Unexplained Error.
The relationship between operational risk and the volumes of a
country's import and export supply chains were similarly specified
by regressing the exports and imports on operational risks as shown
here:
Export = [[beta].sub.8]Operational Risk + [[beta].sub.10] +
[epsilon] (2)
Import = [[beta].sub.9]Operational Risk + [[beta].sub.11] +
[epsilon] (3)
where [[beta].sub.10], [[beta].sub.11] are the Intercepts,
[[beta].sub.8], [[beta].sub.9] are the Population estimates for
operational risk.
The model diagnostics indicate that our specification is accurate.
The validity of the regression model was confirmed by checking the three
assumptions of residuals as defined by the Gauss-Markov theorems: (i)
normally distributed residuals, (ii) errors have zero mean (refer to
Table II) and (iii) (7) errors have constant variance. Figure 3 below
validates that the residuals are evenly distributed between [+ or -] 1
(with one outlier).
[FIGURE 3 OMITTED]
RESULTS
Table III presents the descriptive statistics of the variables of
interest. Table IV presents the estimation results of the three models
from OLS regression, as presented in the equations (1), (2) and (3),
respectively. As Table IV below suggests, all the three models (from
equations 1, 2 and 3, respectively) hold true at a 5 percent level of
significance. (8)
Table V below shows the OLS regression coefficients and their
respective significances, along with the explained variation of the
regression model.
Table VI shows the PLS regression coefficients and their respective
significances, as depicted in Wold's Variable Importance Plot (VIP;
Wold 1985). The statistics summarize the contribution a variable makes
to the model. If a variable has a small coefficient and a small VIP,
then it is a candidate for deletion from the model. Wold (1994)
considers a value of 0.8 to be a small VIP.
We now describe the results for each of our hypotheses and focus
only on the estimation of our hypothesized model in Figure 2. Hypothesis
1 argues that higher transparency reduces operational risk of a country.
The coefficient of transparency is significant and negative in the model
([beta]=-0.408, p<0.05; Table V). Hence, our results support this
hypothesis. Similarly, Hypotheses 3-6 are also supported by our results.
Hypothesis 2 argues that greater economic freedom reduces
operational risk. However, our analysis shows that the coefficient of
economic freedom is not statistically significant even though the
direction of hypothesized causality is maintained ([beta]=- 0.029,
p>0.05; Table V). The fact that our results do not follow the
hypothesis can be attributed to the fact that the Heritage Foundation
Index, widely used as a proxy for economic freedom, is probably not a
sufficient indicator of true economic freedom. The next section
discusses this anomaly in detail.
Hypothesis 7 argues that a higher economic growth reduces
operational risk. However, our results show that while the coefficient
of economic growth follows the same direction of causality, it is
statistically insignificant ([beta]=-0.028, p>0.05; Table V). Thus,
the results do not support this hypothesis.
PLS, by the assumption of orthogonality of its latent factors,
mitigates tire threat of multicollinearity between our explanatory
variables. However, with this added assumption, PLS also loses some
robustness as compared with OLS. With PLS regression, all our
explanatory variables except "Political Rights and Liberties"
turned out to be significant. A possible reason for these results is
explained in the discussion section.
Hypothesis 8 argues that higher operational risk reduces the
volumes of export and import supply chains of a country, respectively
The coefficients of operational risk for both the volume of export
supply chain ([beta]=- 0.057, p=0.003) as well as for the volume of
import supply chain ([beta]=-0.048, p=0.0088) are significant and
negative in the model, thereby supporting the hypothesis.
Finally Hypothesis 9 argues that operational risk of a country
affects its export supply chain more adversely than its import supply
chain. It is important to note here that higher operational risks impact
the volume of export supply chains more than they do to the volume of
import supply chains (note the difference in the coefficients for the
same range in export and import). As noted, we also used export and
import data for the following year to further validate our model. The
validation received confirmation in direction, significance and
magnitude. Hence, our results support this hypothesis. The next section
discusses this finding in greater detail.
Operational Risk as a Mediator Variable
The four steps of Baron and Kenny for testing mediation reveal the
following results as shown in Table VII: (9)
1. For the outbound supply chain (exports) of a country,
operational risk completely mediates the relationship between economic
freedom and exports as well as that between political rights and
liberties and exports; for all the other input variables, operational
risk acts as a partial mediator in their respective relationships with a
country's outbound supply chain.
2. For the inbound supply chain (imports) of a country, operational
risk completely mediates the relationship between economic freedom and
imports, that between political rights and liberties and imports, and
that between financial standards and compliance and imports; for all the
other input variables, operational risk acts as a partial mediator in
their respective relationships with a country's inbound supply
chain.
Implications of these results are discussed next.
DISCUSSION AND CONTRIBUTION
Take calculated risks. That is quite different from, being rash.
(~George S. Patton, US General, WWII)
Supply chains span the globe. For entrepreneurs and mature
businesses competing in a global economy, business survival hinges on a
calculated, rather than a rash, insight on source and market strategies.
Consequently, companies need to rely on third-party assessments of a
country's internal environment, operational risks and trade
dynamics as a prelude and aid to commercial supply chain decision
making. Results support our central argument that third-party indices,
which serve as proxies to various facets of a country's internal
environment, are instrumental in understanding operational risk in a
country. Offering major support for our hypotheses, five of the seven
indices are statistically significant with indices of transparency (a 34
percent impact) and economic-political governance performance (a 27
percent impact) serving as the strongest predictors to operational risk.
The path diagram in Figure 4 illustrates these findings. Moreover,
operational risk in also found to be a key determinant and a mediator of
the volume of a country's import and export supply chains.
[FIGURE 4 OMITTED]
This paper offers important contributions to both research and
practice. First, every third-party index promotes itself as the best
gauge of a country's internal environment. However, one of the
foremost contributions of this study is that no single index can serve
as the quintessential barometer of a country's internal
environment. This study posits that each index is necessary, but not
sufficient. Given that, even in unison, the indices together simply
offer a "partial" view of a country's internal
environment, research and practice should consider these indices as
works in progress.
Using a single index as a measure of a country's internal
environment can be an incomplete gauge. The Heritage Foundation's
Economic Freedom Index is a case in point. Considered a premier source
in this genre of data, the Heritage Foundation's index is popular
as an objective guide of economic success of 183 countries. However, our
research has shown that the data from the Heritage Foundation, while
authentic, is not adequate to understand operational risk in a country
(Table V; [beta]=-0.029, p>0.05). While this index highlights
economic freedom in terms of efficient, strike-free labor forces,
minimum antitrust regulations and no cap in minimum wages, some
significant facts go unmentioned. For example, the index may examine
entrepreneurship as a core indicator in rating a country. However, a
country can have extremely convenient regulations about opening a
business but extremely stringent regulations related to bankruptcy
protection and credit defaults (eg., Dubai, Iceland). Unless the index
covers all facets of entrepreneurial economics, the index remains
incomplete, per se.
While Adam Smith's concept of "laissez faire" is a
central feature in global supply chains, his remark that "when
institutions protect the liberty of individuals; greater prosperity
results for all" (Viner 1927) may not always hold true. Consider
two countries that are classified "free" by the Heritage
Foundation's Economic Freedom Index and the Wall Street Journal:
Singapore and Bahrain. Singapore has had a long history of violation of
civil rights. In Singapore, members of a religious sect spent weeks
behind bars for "peacefully exercising their right to freedom of
expression" (Solomon 2000). Again, while Bahrain wins accolades for
its vibrant and competitive banking market with few government
restrictions, the index overlooks the fact that Bahrain has had a
traditional monarchy and that the government has suppressed dissent for
the past three decades. Amnesty International has in the past noted that
Bahrain's political detainees have included "children as young
as 10." This example supplies fuel to our PLS findings that
illustrate the insignificance of "political rights and
liberties" in determining operational risks in a country.
Second, the article highlights the role of intermediaries in TCE as
a mechanism to reduce uncertainties in commercial supply chain
operations. As there is a natural tendency for global supply chains to
be inefficient, (10) decision makers often seek intermediation.
Commercial supply chains stretch across the globe with little control
over the exact logistics, shipment routes, and local shifts in economy,
operations and regulation. Little control and a lack of end-to-end
coordination reduce supply chain efficiencies and make supply chains
less transparent. Faced with such uncertainties from incomplete
information, parties in supply chains need to rely on independent,
unbiased and reliable sources of information offered by third-party
intermediaries. This paper puts forth the point that third-party index
providers, as intermediaries, play a vital role in reducing transaction
costs (and hence, information asymmetry) while trading with multiple
countries. The upshot of using third-party indices is that, given the
costs and difficulties of direct information acquisition to gauge the
credibility of a country's internal environment, indices provide
useful cues that proxy for direct information about a country's
internal environment.
Third, an effort to find the impact of operational risk on a
country's import and export supply chains leads to interesting
findings. Our results indicate that operational risk has approximately a
20 percent greater adverse effect (coefficient of operational risk in
model II, i.e. exports is--0.057 as opposed to that in model III, i.e.
imports, which is -0.048) on exports than on imports. Higher operational
risks reflect tough production and logistics environments within a
country by dampening the growth of Export Oriented Units (EOUs). (11)
For a country that suffers from production and distributional woes, it
often has to rely more on imports to cater to internal demand while
stagnating trade balance.
Fourth prior research had focused on understanding risk in light of
risk sources and the causality between them. In contrast, we try to
understand a country's operational risk as a function of its
internal environment, using some of the most widely used third-party
global indices as proxies to the latter. Specifically, we observe the
holistic nature of all (or the significant) indices in measuring
operational risk. By investigating 81 member countries of the WTO, and
controlling for country-specific and common method biases, we show that
unlike the business environment (as envisioned in most previous
research), the information environment (transparency) and the
socioeconomic environment (governance performance) of a country impact
its operational risk the most.
Fifth and finally, to the best of our knowledge, no effort has yet
been undertaken to investigate a country's operational risk as a
mediator between the country's internal environment and the volume
of its import and export supply chains. This paper portrays operational
risk in the role of a complete or partial mediator to the relationship
between a country's internal environment and its import and export
supply chain volumes.
So what defines a country's ability to build its import and
export supply chain volume? Is it the freedom of the masses, the
compliance of financial institutions, a market-driven economy, or a
parliamentarian democracy? In contrast with the traditional views that
socioeconomic freedom is the only prerequisite to trade, our study
offers an interesting departure. Our findings suggest that it is not the
internal environment of a country that necessarily drives trade. Rather,
it is how well a country manages perceptions of operational risk. From
our previous discussion and reference to Solomon (2000), Asian countries
like Singapore and China have shown that a country can travel the path
toward a free market while maintaining an authoritarian regime for a
long period of time. China, for example, has a pegged currency for
ensuring stability in long-term contracting with the United States.
China's authoritarian regime ensures certain operational stability
by maintaining uniform codes and capital flows. Note that, over the past
decade, the Chinese government has taken serious steps to facilitate
business operations more than facilitating civil liberties. And most
companies and their host countries welcome China's posture toward
managing operational risk as a salient precondition to building import
and export supply chains.
Limitations and Future Work
Global third-party indices have long been (and still are) used as
proxies to identify internal environments of any country. While the
authenticity of data sources for these indices is beyond debate, some
questions remain about the construct validity of these indices (i.e., do
the indices truly measure what they expect to measure?). Moreover,
because most of these third-party indices originate from world bodies
(like the World Bank, IMF and the WTO) that rely on global data,
precedent factors that make up these indices are sometimes highly
correlated. For example, Moody's Credit Rating index examines
regulatory and banking standards that closely map with the Heritage
Foundation's (EFI) factors of capital infrastructure and
export-import regulations. Similarly the Global Competitive Index's
(GCI) unemployment is closely related to Bertelsmann's
Transformation Index's (TSI) factor on poverty--greater
unemployment may signify increased poverty. Therefore, there remains an
issue regarding redundancy in country risk measurements when a multitude
of such indices are taken into consideration.
Given that multicollinearity with economic data is quite common,
previous researchers have traditionally put a higher cap on the variance
inflation factors to validate their findings. In their analysis of data
pertaining to innovations and acquisitions, Prabhu, Chandy and Ellis
(2005) have defined an acceptable cut-off of 10 while studying the
variance inflation factor statistics for their variables. Nevertheless,
high multicollinearity between economic variables remains an issue with
this type of study.
However, this issue also creates a platform for future research.
First, because each of the global indices used in this study uses a wide
range of factors and concepts to define the respective indices, it is
imperative to choose some of these indices and investigate the
completeness (12) of their composition. Second, building on our argument
that certain constituents for the indices do covary, it would be
operationally prudent to establish a single index that removes
covariance of the correlated constituents.
On a different note, a common problem encountered in choosing or
relying on an index is its credibility to capture short-term shifts in
the environment. We contend that while the nature of the indices may
certainly follow a lagged consideration of internal environments (often
a function of the periodicity of the measurements), we have been
cautious to measure them at similar periodic intervals. Furthermore, it
is useful to note that operational decision-making is often contractual
and considers various windows-of-operation and thus requires a
relatively "temporally stable" assessment of a country's
environment. An "oversensitive" or "nervous" index
may not serve to best inform decision-making over a longer time horizon.
(13)
Finally, while this study examines short-term relationships between
internal governance environments, operational risks and trade, important
questions remain: Can higher economic growth rates be maintained in the
long term? Will an inability to maintain long-term higher economic
growth rates over time adversely affect operational risks? It is
interesting to realize that economic growth rates are a part of a larger
mosaic of factors and it is the promise that a country has in place the
right governance ingredients for mitigating operational risks. Moreover,
there may be some disagreement over the window that constitutes long
term. While it is true that a precipitous economic drop over time will
severely affect operational risk perceptions (e.g., Zimbabwe), it would
be useful to examine the relationship in greater depth using panel data.
Future researchers can conduct a similar study from a time-series lens
on a single country by utilizing all of these indices from multiple time
periods to account for changes in operational risks for that country.
Such a study might also offer key insights into our assumptions of
linearity between operational risk and economic growth over time.
Conclusion
In short, this research examines operational risk of a country in
the light of its internal environment and assesses the mediating role of
operational risk in relating a country's internal environment to
the volume of its import and export supply chains. It turns out that a
country's information environment and socioeconomic environment are
better indicators of the country's operational risk than is the
country's business environment.
Supply chain decisions are core to any business. Information
asymmetry, arising from miscued information in decision-making, can lead
to dire consequences. Firms are advised to have a better understanding
of some of the antecedents of information transparency, like timeliness,
periodicity and coverage of disclosures, corruption levels, and
technical authenticity of reported standards from prospective countries
before negotiating a business proposition. Firms are also advised to
study a region's balance between democracy and market economy,
which is core to understanding operational risks in that region.
REFERENCES
Aron, R., E.K. demons and S. Reddi. "Just Right Outsourcing:
Understanding and Managing Risk," Journal of Management Information
Systems, (22:2), 2005, pp. 37-55.
Baron, R.M. and D.A. Kenny. "The Moderator Mediator Variable
Distinction in Social Psychological Research: Conceptual, Strategic and
Statistical Considerations," Journal of Personality and Social
Psychology, (51:6), 1986, pp. 1173-1182.
Basel Committee on Banking Supervision (BCBS). Consultative
Document. Operational Risk: Bank for International Settlements. Report
2008.
Bauer, A., N. Haltom and W. Peterman. "Decomposing
Inflation," Economic Review, Federal Reserve Bank of Atlanta,
Quarter 1, 2004, pp. 39-51.
Bertelsmann. Transformation Status Index, 2008-2009,
http://www.bertelsmann-transformation-index.de/11.0.html?andL=1, May 22,
2009.
Bo, E.D. and M.A. Rossi. "Corruption and Inefficiency: Theory
and Evidence from Electrical Utilities," Journal of Public
Economics, (91:5-6), June 2007, pp. 939-962.
Business Indicator Index. Financial Standards Foundation @
eStandardsForum, 2008-2009, http://www.estandardsforum.org/jhtml/rm/,
May 22, 2009.
Campbell, D.T. and D.W. Fiske. "Convergent and Discriminant
Validation by the Multitrait-Multi-method Matrix," Psychological
Bulletin, (56:2), March 1959, pp. 81-105.
Cavinato, J.L. "Supply Chain Logistics Risks: From the Back
Room to the Board Room," International Journal of Physical
Distribution and Logistics Management, (34:5), 2004, pp. 383-387.
Clemens, M. and T Moss. "Costs and Causes of Zimbabwe's
Crisis." Report, Centre for Global Development, 2005.
Datta, P. and S. Chatterjee. 'The Economics and Psychology of
Consumer Trust in Intermediaries in Electronic Markets: The EM-Trust
Framework," European Journal of Information Systems, (17), 2008,
pp. 12-28.
Diamantopoulos, A. and H.M. Winklhofer. "Index Construction
with Formative Indicators: An Alternative to Scale Development,"
Journal of Marketing Research, (38:2), 2001, pp. 269-277.
Enyinda, C.I., A. Ogbuehi and C. Briggs. "Global Supply Chain
Risks Management: A New Battleground for Gaining Competitive
Advantage," Proceedings of ASBBS, (15:1), 2008, pp. 278-292.
Freedom House. Freedom House Index, 2008-2009,
http://www.freedomhouse.org/template.cfm?page=395, May 22, 2009.
Gelos, R.G. and S.J. Wei. "Transparency and International
Portfolio Holdings," The Journal of Finance, (60:6), 2006, pp.
2987-3020.
Griswold, D. and J.C. Hidalgo. "A U.S.-Colombia Free Trade
Agreement," Cato Institute, Center for Trade Policy Studies,
February 7, 2008, http://www.freetrade.org/pubs/FTBs/FTB-032.pdf, May
23, 2009.
Heritage Foundation. Heritage Foundation Economic Freedom Index,
http://www.heritage.org/index/Ranking.aspx, 2008-2009, May 22, 2009.
Hoe, S.L. "Issues and Procedures in Adopting Structural
Equation Modeling Technique," Journal of Applied Quantitative
Methods, (3:1), Spring 2008, pp. 76-83.
Hulland, J."Use of Partial Least Squares (PLS) in Strategic
Management Research: A Review of Four Recent Studies," Strategic
Management Journal, (20), 1999, pp. 195-204.
Hsu, S.H., W.H. Chen and M.J. Hsieh. "Robustness Testing of
PLS, LISREL, EQS, and ANN-Based SEM for Measuring Customer
Satisfaction," Total Quality Management, (17:3), 2006, pp. 335-371.
Judd, C.M. and D.A. Kenny. "Process Analysis: Estimating
Mediation in Treatment Evaluations," Evaluation Review, (5), 1981,
pp. 602-619.
Knight, F.H. "Risk, Uncertainty and Profit" Reprint of
the 1981 Edition with a Foreword by George J. Stigler, University of
Chicago Press, Chicago, 1985.
Lamming, R.C., N.D. Caldwell, D.A. Harrison and W. Phillips.
"Transparency in Supply Relationships: Concept and Practice,"
Journal of Supply Chain Management, (37:4), 2006, pp. 4-10.
Landeghema Van, H. and H. Vanmaele. "Robust Planning: A New
Paradigm for Demand Chain Planning," Journal of Operations
Management, (20:6), 2002, pp. 769-783.
Lee, H.L. and C. Billington. "Managing Supply Chain Inventory:
Pitfalls and Opportunities," Sloan Management Review, (33:3),
Spring 1992, pp. 65-73.
Madslien, J. "No Quick Fix for Zimbabwe's Economy,"
BBC News, 14 April 2008,
http://news.bbc.co.uk/2/hi/business/7346042.stm, May 15, 2009.
March, J.G. and Z. Shapira. "Managerial Perspectives on Risk
and Risk Taking," Management Science, (33), 1987, pp. 1404-1418.
Moody's Credit Rating. 2008-2009,
http://www.countryrisk.com/guide/archives/cat_country_risk_ratings.html,
May 24, 2009.
Neter, J., W. Wasserman and H.J. Kutner Applied Linear Statistical
Models, 3rd ed., Irwin, Homewood. TL, 1990.
Oetzel, J.M., R.A. Bettis and M. Zenner. "Country Risk
Measures: How Risky are they?," Journal of World Business, (36:2),
2001, pp. 128-145.
Porter, M.E. "Location, Competition, and Economic Development:
Local Clusters in a Global Economy," Economic Development
Quarterly, (14:1), 2000, pp. 15-34.
Prabhu, J.C., R.K. Chandy and M.E. Ellis. "The Impact of
Acquisitions on Innovation: Poison Pill, Placebo, or Tonic?,"
Journal of Marketing, (69:)January), 2005, pp. 114-130.
Rindfleisch, A. and K.B. Heide. "Transaction Cost Analysis:
Past, Present, and Future Applications," Journal of Marketing,
(61:October), 1997, pp. 30-54.
Roberts, J.M. The Best Protection for both Colombian and American
Workers: Stronger, Market-Based Democratic Institutions in Colombia,
Heritage Foundation Report, May 7, 2009,
http://www.heritage.org/research/tradeandeconomicfreedom/tst050709a.cfm,
May 15, 2009.
Rose-Ackerman, S. International Handbook on the Economics of
Corruption, Edward Elgar, Cheltenham, UK, 2006.
Roth, A.V., A.A. Tsay, M.E. Pullman and J.V. Gray. "Unraveling
the Food Supply Chain: Strategic Insights from China and the 2007
Recalls," Journal of Supply Chain Management, (44:1), 2008, pp.
22-39.
Shah, M. Is Your Supply Chain Catching Flu? A Blog Report,
http://www.infosysblogs.com/supplychain/2009/05/is_your_supply_chain_catching.html, May 24, 2009.
Solomon, C. Cracks in the Glass Ceiling, Business Services
Industry, September 2000,
http://findarticles.com/p/articles/mi_m0FXS/is_9_79/ai_65650779/, May
22, 2009.
Spence, M. "Signaling in Retrospect and the Informational
Structure of Markets," American Economic Review, (92:3), 2002, pp.
434-459.
Supply Chain Digest. "Supply Chain News and Views: Where Are
the Risks to the Global Supply Chain? New Tool Assesses Risk by
Country," Supply Chain Digest, 2006,
http://www.scdigest.com/assets/NewsViews/06-01-30-1.cfm?cid=127andctype=content, May 22, 2009.
The Economist. "Argentina Jiggles Its Inflation Index,"
The Economist, June 12, 2008.
The Economist. "Iceland: Cracks in the Crust," The
Economist, November 11, 2008, Print Edition, May 30, 2009.
The Economist. "A Second Life," The Economist, December
17, 2009.
Transparency International's Corruption Perception Index.
2008-2009, http://transparency.org/news_room/in_focus/2008/cpi2008/cpi_2008_table, May 22, 2009.
Van Oosterhout, J.V., P. Heugens and M. Kapstein. "The
Internal Morality of Contracting: Advancing the Contractualist Endeavor
in Business Ethics," Academy of Management Review, (31:3), 2006,
pp. 521-539.
Viner, J. "Adam Smith and Laissez Faire," The Journal of
Political Economy, (35:2), April 1927, pp. 198-232.
Williamson, O. The Economic Institution of Capitalism, Free Press,
New York, 1985.
Williamson, O.E. "Outsourcing: Transaction Cost Economics and
Supply Chain Management," Journal of Supply Chain Management,
(44:2), 2008, pp. 5-16.
Wold, H. "Partial Least Squares." In S. Kotz and N.L.
Johnson (Eds.), Encyclopedia of Statistical Sciences, Vol. 6, Wiley, New
York, 1985, pp. 581-591.
Wold, H. "Exponentially Weighted Moving Principal Components
Analysis and Projections to Latent Structures," Chemometrics and
Intelligent laboratory systems, (23), 1994, pp. 149-161.
World Economic Forum. Global Competitive Index, 2008-2009,
http://www.weforum.org/documents/GCR0809/index.html, May 22, 2009.
* This research was funded by a grant to Pratim Datta from the
Farris Family Research Innovation Fellowship.
(1) The authors will make their dataset publicly available, on
request.
(2) It is worthwhile to mention here that the entire notion of TCE
is built on asymmetric information between two transacting parties. It
thereby comes as no surprise that one of the salient features of a
governance structure, in order to mitigate asymmetry, would be to
institute transparency in the transactional process (Lamming et al.
2006). In fact, transparency has gained a lot of momentum with research
in all supply chains that aim for a global reach. As an example, Roth et
al. (2008) sketch the vulnerability of global food supply chains with
poor visibility.
(3) For the purposes of this paper, we use the term "internal
environment" synonymously with concepts of "governance
environment" and "institutional environment."
(4) The Business Indicator Index (Bllndex) is courtesy of Financial
Standards Foundation.
(5) The heuristics follow a four-step process (Baron and Kenny
1986): step 1 establishes an effect that may be mediated; step 2 treats
the mediator as an outcome variable; step 3 establishes the coexistence
of the mediator and the input variables in explaining the outcome;
finally, step 4 explains the extent of mediation.
(6) The assumption of linearity is underpinned by the choice of a
linear regression as a robust technique. While it is difficult to
accurately establish linearity (somewhat mitigated with bivariate
plots), multiple regression analysis is not largely affected by minor
deviations in the linearity assumption (Neter et al. 1990).
(7) The fourth assumption regarding autocorrelation was relaxed
since the data was not time series.
(8) As an additional validation step, we tested the relationship
between operational risk and the volumes of a country's export and
import supply chains by looking at export and import data of the 81
countries in our dataset for two consecutive years. The results mimic
our current findings. The results are omitted for purposes of brevity
and are available from the authors upon request.
(9) Table VII shows steps 1-4 as proposed by Baron and Kenny
(1986); total observations = 81; * significance at the [alpha]=0.05
level.
(10) As noted earlier, inefficiencies arise from the distributed
nature of commercial supply chains linking multiple entities that create
severe coordination, communication, and contracting hurdles.
(11) FOUs are firms whose main focus is generating goods and
services for explicit purposes of export. Such EOUa are granted certain
tax havens and safeguards by the government to promote exports.
(12) As each global index used in this study comes from an
independent research body, the indices, taken together, may provide
complete information. However, certain indices may be more complete than
others. If an index is more complete, it would require less
complementary (and supplementary) information, and research and practice
can focus on extending that particular index to make it more complete.
(l3) We thank an anonymous reviewer for his/her direction and
helpful comments here.
Kuntal Bhattacharyya (MBA, The University of Akron) is a Ph.D.
candidate in the Department of Management and Information Systems in the
College of Business, Kent State University in Kent, OH. In addition to
the MBA, Bhattacharyya also holds a Master of Science degree in
electrical engineering. Before the beginning the doctoral program at
Kent State University, he worked for The Timken Company as a sourcing
analyst and then in the company's industrial Operations.
Bhattacharyya's research interests focus on strategic sourcing,
supply chain risk management, buyer-seller relationships and supply
chain resilience. His research has been presented at numerous
conferences and appears in the Mid-West Decision Sciences Proceedings
and America's Conference on information Systems.
Pratim Datta (Ph.D., Louisiana State University) is an assistant
professor of information systems in the Department of Management and
Information Systems in the College of Business at Kent State University
in Kent, OH. He is also a Farris Research Innovation Fellow and a
Teaching Scholar. Dr. Datta's research interests include the
design, security and innovation of information systems, as well as
supply chain process and reengineering. He currently is developing a
green supply chain index.
O. Felix Offodile (Ph.D., Texas Tech University) is a professor in
the Department of Management and Information Systems in the College of
Business at Kent State University in Kent, OH. His teaching and research
activities focus on operations management with special emphasis on
robotics and technology assessment, cellular manufacturing, quality
control, inventory management and supply chain management. Dr. Offodile
has published the results of his research in the International journal
of Production Research, the international Journal of Production
Economics, Technometrics, the Journal of Manufacturing Systems and the
Journal of Operations Research, among other publications.
TABLE IA
Constructs and Indices
Internal Factor Construct Function of index [Variable
Environment ... Definition in
Parentheses]
Information X1 Transparency Timeliness of Transparency
environment disclosure International's
Periodicity and Corruption
coverage Perception
Index
[TICRIndx]
Corruption
Quality of
statistics to
support
standards
Business X2 Economic Import-export Heritage
environment freedom restrictions Foundation
Labor Economic
flexibility Freedom Index
[HFEFIndex]
Capital
infrastructure
X3 Relational Ease and Business
continuity turnover of Indicator Index
Business cycle [BIIndex]
X4 Financial Reporting Moody's Credit
standards and standards Rating [Moody]
compliance
Banking system
Regulatory
system
Socio-economic X5 Political Democracy Freedom House
environment rights and Index [FHIndex]
liberties
Legal system
Gender
equality
Freedom and
respect
X6 Governance Market Economy Bertelsmann
performance Transformation
Poverty Status Index
[BTSIndex]
Population
X7 Economic Real GDP World Economic
growth forecast Forum
Unemployment Global
Competitive
Index
[WEFGCIndex]
Education
HealthTABLE IB
Explanation of Indices in the Study
Index name Criterion Type of Data Data Source
(Criterion) Attributes
Transparency Timeliness of Ranked, frequency Country experts
International's disclosures data (from low to (7 sources, e.g.,
Corruption Periodicity and high EIU) and business
Perception Index coverage transparency) leaders (3
(CPI) Corruption sources, e.g.,
Quality of IMD)
statistics to
support
standards
Heritage Import-export Continuous (Ratio World Bank, EIU,
Foundation's restrictions type, rate; from U.S. Department
Economic Freedom Labor low to high of Commerce,
Index (EFI) flexibility freedom) Government
Capital Publication from
infrastructure each country
Financial Ease of business Ranked data by 12 business
Standards Turnover of degree of indicators (e.g.,
Foundation's business cycle compliance (from FOREX, US foreign
Business low to high "ease investment Law,
Indicator Index of business") US trade
(Bll) regulations)
Moody's Credit Reporting Ranked data (from Sovereign Default
Rating (Moody) standards low to high and Recovery
Banking system financial rates, Moody's
Regulatory standards) financial metrics
system key ratios
Freedom House Democracy Numerical rating, Analytical
Index (FHI) Legal system ranked by reports from 193
Gender equality countries (from countries in 16
Freedom and low to high territories
respect to rights and
minorities liberties)
Bertelsmann's Market Economy Ranked data, by Analytical
Transformation Poverty market economy assessment on 17
Status Index Population and management criteria
(TSI) performance (from pertaining to
low to high market economy
governance and performance,
performance) across 128
countries
World Economic Real GDP Ranked data, Publicly
Forum's Global forecast frequency (from available data
Competitive Unemployment low to high for 134 countries
Index (GCI) health economic growth) and the
"Executive
Opinion Survey"
conducted by
World Economic
Forum annually in
these countries
Index name Method of Collection Frequency Common Usage
(Criterion) of Update
Transparency Meta survey of 13 Annual Global
International's surveys from 10 corruption
Corruption sources Bribery watch
Perception Index Promoting
(CPI) revenue
Transparent
reporting
Heritage 10 components of Annual Economic
Foundation's economic freedom are performance
Economic Freedom equally weighed and Economic freedom
Index (EFI) averaged
Financial Sum scores from 12 Annual Adherence to
Standards business indicators to global standards
Foundation's rank against a Country risk
Business standard profile
Indicator Index
(Bll)
Moody's Credit The risk is assessed Annual Political Risk
Rating (Moody) based on Moody's key Risk of investing
ratios and comparison in a foreign
to rates; a rating is land
then applied to the
assessed risk
Freedom House Survey; numerical Annual Democratic
Index (FHI) ratings for both values
political rights and Political rights
human liberties are Civil liberties
assigned for ranking
Bertelsmann's The 17 criteria are Annual Best practices in
Transformation subdivided into 52 constitutional
Status Index questions; mean value democracy
(TSI) of the answers are Practices in
ranked socially
responsible
market economy
Efficient
political
management for
boosting
international
trade
World Economic A polling of over Annual Impediments to
Forum's Global 12,000 business economic growth
Competitive executives worldwide Strategies to
Index (GCI) (courtesy, Executive achieve sustained
Opinion Survey) economic
progressTABLE II
Normality and Zero Mean
Regression error average 0.00
Standard deviation of errors 0.52
D statistic 0.0836
D critical at 1% 0.0894
D critical at 5% 0.0962
D critical at 10% 0.1146
Null hypothesis: The errors are normally distributed. Conclusion: The
errors are normally distributed at the 1% [alpha] level.
TABLE III
Descriptive Statistics
Mean Standard Deviation
Transparency 4.98 2.3
Economic freedom 6.42 1.02
Relational continuity 6.76 2.05
Financial standards and compliance 5.86 2.97
Political rights and liberties 3.81 2.57
Governance performance 7.3 1.73
Economic growth 6.35 0.97
TABLE IV
Test of Models
Model Criterion Predictor(s) F Ratio Probability
Model I Operational risk Third-party indices 105.59 < 0.0001 *
Model II Export Operational risk 9.375 0.003 *
Model III Import Operational risk 7.222 0.0088 *
Number of observations: 81;
* Significance at the [alpha] = 0.05 level [R.sup.2] for Model I: 90.1%.TABLE V
Estimation of Coefficients (OLS)
Model Factors Coefficient p Value
Model 1 Transparency - 0.408 < 0.0001 *
Economic freedom - 0.029 0.8378
Relational continuity - 0.146 0.0135 *
Financial standards and compliance - 0.107 0.0235 *
Political rights and liberties - 0.161 0.0194 *
Governance performance - 0.326 0.0053 *
Economic growth - 0.028 0.8421
Model II Operational risk - 0.057 0.003 *
Model III Operational risk - 0.048 0.0088 *
# of observations: 81.
* Significance at [alpha] = 0.05 level.TABLE VI
Estimation of Coefficients (PLS)
Variable Importance Plot
Predictor VIP
TICRIndx 1.143647
HFEFIndex 1.0527949
BlIndex 0.9964911
Moody 1.0370518
FHIndex 0.7890031
BTSIndex 0.9214549
WEFGCIndex 1.0589557
TABLE VII
Operational Risk as a Mediator Variable
Baron-Kenny Model Criterion Predictor F ratio Probability
steps
Step I Model IV Export Third-party 4.57 0.0003 *
indices
Model V Import Third-party 4.00 0.0009 *
indices
Step II Model II Export Operational 9.375 0.003 *
risk
Model III Import Operational 7.222 0.0088 *
risk
Step III Model VI Export Third-party 4.05 0.0005 *
indices
Operational
risk
Model VII Import Third-party 3.91 0.0007 *
indices
Operational
risk
p Value
Third-Party Indices Model IV Model VII
Step IV Transparency 0.6295 0.2343
Economic freedom 0.0263 * 0.0227 *
Relational continuity 0.8505 0.3982
Financial standards and 0.4140 0.0259 *
compliance
Political rights and liberties 0.0117 * 0.0485 *
Governance performance 0.2220 0.3254
Economic growth 0.6855 0.6097
* Significance at [alpha] = 0.05 level.APPENDIX
Country List Used in the Study
Country Region Country Region
Algeria Africa Canada N. America
Cameroon Africa Dominican Republic N. America
Egypt Africa United States N. America
Ghana Africa Australia Oceania
Kenya Africa New Zealand Oceania
Morocco Africa Argentina S. America
Nigeria Africa Bolivia S. America
South Africa Africa Brazil S. America
Tanzania Africa Chile S. America
Tunisia Africa Colombia S. America
Bangladesh Asia Ecuador S. America
China Asia Guatemala S. America
Hong Kong Asia Honduras S. America
India Asia Mexico S. America
Indonesia Asia Peru S. America
Japan Asia Uruguay S. America
Kazakhstan Asia Venezuela S. America
Malaysia Asia Austria W. Europe
Pakistan Asia Belgium W. Europe
Philippines Asia Bulgaria W. Europe
Singapore Asia Croatia W. Europe
South Korea Asia Czech Republic W. Europe
Sri Lanka Asia Denmark W. Europe
Taiwan Asia Finland W. Europe
Thailand Asia France W. Europe
Vietnam Asia Germany W. Europe
Estonia E. Europe Greece W. Europe
Latvia E. Europe Hungary W. Europe
Lithuania E. Europe Ireland W. Europe
Poland E. Europe Italy W. Europe
Romania E. Europe Luxembourg W. Europe
Slovakia E. Europe Netherlands W. Europe
Slovenia E. Europe Norway W. Europe
Turkey E. Europe Portugal W. Europe
Ukraine E. Europe Russia W. Europe
Iran Middle East Spain W. Europe
Israel Middle East Sweden W. Europe
Jordan Middle East Switzerland W. Europe
Lebanon Middle East United Kingdom W. Europe
Saudi Arabia Middle East
Syria Middle East
UAE Middle East