Subnational Government Performance: Testing Multiple Dimensions of Governors' Experience

• DOI: http://doi.org/10.1111/puar.13035
• Published date: 01 May 2019
• Date: 01 May 2019

Subnational Government Performance: Testing Multiple Dimensions of Governors’ Experience 383

Abstract: This article tests whether governors’ work experience explains their state performance. Although the

experience-performance relationship has been researched, empirical studies report inconsistent results. This

inconsistency might be due to the dissimilar conceptualization and operationalization of the work experience construct,

for it can be conceptualized in qualitative or quantitative components. To address this disparity, the authors follow

Tesluk and Jacob’s conceptualization by operationalizing gubernatorial experience across three dimensions: type

(qualitative component), number of jobs, and years of work experience (quantitative component). The study tests

the moderating effects of state context (level of violence) on the experience-performance relationship, relying on data

derived from 32 Mexican states over 16 years. Education outputs (high school enrollment) and health outcomes (infant

mortality rate) are used to assess state performance. Results indicate that only the qualitative dimension of governors’

work experience boosts state performance by reducing infant mortality and increasing high school enrollment.

However, this positive effect is undermined by state violence. The number of jobs decreases gubernatorial performance,

magnified by the degree of state violence.

Evidence for Practice

• A chief executive’s work experience can be conceptualized and operationalized across three dimensions:

tenure, type of work, and number of jobs.

• The impact of chief executives’ work experience on performance is contingent on the dimension of work

experience.

• While a chief executive’s number of jobs fails to boost performance, type of experience (local and state)

increases performance.

• Challenging context, such as violence, undermines the positive effect of experience on performance.

Johabed G. Olvera

Indiana University Bloomington

Claudia N. Avellaneda

Indiana University Bloomington

Subnational Government Performance: Testing Multiple

Dimensions of Governors’ Experience

Research Article

With the adoption of decentralization,

subnational governments have become

responsible for planning, funding, and

implementing social programs. Despite having

common responsibilities and comparable funding

transfers from central governments, subnational

governments vary in their performance across several

indicators. For example, the states of Morelos,

Zacatecas, and Puebla in Mexico have relatively

higher levels of education enrollment (see figure 1),

while Zacatecas, Baja California, and Yucatán have

the lowest levels of infant mortality among states

(see figure 2). We ask, What explains subnational

government performance?

Considerable scholarship has addressed the

determinants of government performance (Boyne

2003). Scholars have identified several drivers, including

political competition (Boyne 1998; Hilke 1993),

human and material resources (Meier and Keiser 1996;

Sharkansky 1967), organizational structure (Bohte and

Meier 2000; Lan and Rainey 1992; Whetten 1978),

monitoring mechanisms (Wolf 1993), networking and

intergovernmental cooperation (Agranoff and McGuire

2003; O’Toole 2015), and managerial variables

such as strategy, leadership, and human resource

management (Andrews and Boyne 2010; Boyne and

Walker 2005; Brewer and Selden 2000; Weisman and

Nathanson 1985). Managerial quality contributes to

government performance because qualified managers

are better equipped to take advantage of opportunities,

address challenges, maneuver difficult situations, and

exploit networks (Avellaneda 2009, 2012). These

qualification differences among chief executives explain

organizational actions (Boyne 2004; Hambrick 2007;

Hambrick and Mason 1984; Lynn 2003).1 Among the

background qualifications that contribute to managerial

quality, experience has received considerable attention

(Avellaneda 2009, 2012; Carmeli 2006; Meier and

O’Toole 2002; Petrovsky, James, and Boyne 2015).

Nevertheless, studies report inconsistent results for the

experience-performance relationship.

Public Administration Review,

Vol. 79, Iss. 3, pp. 383–398. © 2019 by

The American Society for Public Administration.

DOI: 10.1111/puar.13035.

Claudia N. Avellaneda is associate

professor in the School of Public and

Environmental Affairs at Indiana University

Bloomington. Her research interests include

public management and governance

in developing countries as well as local

governments, mayoral leadership, public

policy, and comparative politics with a

regional focus on Latin America. Her current

research examines the determinants of local

and subnational government performance

in Latin America, highlighting the role

of mayors’ and governors’ background,

innovation, networking, and decision making.

E-mail: cavellan@indiana.edu

Johabed G. Olvera is a PhD candidate

in the School of Public and Environmental

Affairs at Indiana University Bloomington.

Her research interests include public policy

analysis, implementation, and management

in Latin America. She specializes in health

programs implemented by subnational and

local governments targeting vulnerable

population. Her current research looks

at the impact of healthcare interventions

aimed at improving maternal and infant

health and the role of managers and street-

level bureaucrats in program performance.

E-mail: jolverae@indiana.edu

384 Public Administration Review • May | J une 2 019

This may be due to the conceptualization and operationalization of

experience. Most studies assess managerial experience in quantitative

terms, equating it with tenure and operationalizing it with “years

of experience.” But few works test the qualitative dimension of

experience, and fewer still test the performance effects of both

qualitative and quantitative experience measures (Quiñones, Ford,

and Teachout 1995; Tesluk and Jacobs 1998). This scarcity is more

evident in the public sector. Therefore, this study’s first contribution

is to test the effect of three dimensions of experience on performance,

using qualitative and quantitative measures of experience.

Inconsistent results also may be due to few tests of the moderating

effects of context on the experience-performance relationship (Meier,

Rutherford, and Avellaneda 2017). According to Tesluk and Jacobs,

“contextual factors can moderate the influence of experience because

specific situations influence what individuals can extract from work

experiences” (1998, 332–33). Consequently, our second contribution

is to test the indirect effects of experience by assessing the moderating

effect of context on the experience-performance relationship.

We assess government performance in terms of education outputs

(high school enrollment) and health outcomes (infant mortality

rate, or IMR). We rely on data derived from 32 Mexican states

over 16 years. We selected the Mexican case for several reasons.

First, the Mexican decentralization process in the 1980s and 1990s

changed the balance of political power toward subnational and local

governments (Falleti 2010), extending states’ fiscal and political

autonomy. States are responsible for delivering key public services

(Hernández Rodríguez 2008; Langston 2010). This increased

gubernatorial autonomy and power decentralization make Mexico

an appropriate setting in which to investigate whether governors’

backgrounds influence state performance. Although a considerable

literature addresses the role of Mexican governors (Hernández

Rodríguez 2008), no empirical studies exist, to our knowledge, on

the drivers of gubernatorial performance in Mexican states. Our

third contribution is to test the experience-performance relationship

at the subnational level in a developing setting with a democracy

characterized by a long-lasting hegemonic party system.

Our results demonstrate, after controlling for time and state- and

governor-specific factors, that governors’ experience is positively

correlated with performance. The effects of gubernatorial experience

are contingent on its operationalization. Only the qualitative

dimension of work experience boosts state performance by reducing

IMR and increasing high school enrollment. However, this positive

effect is undermined by state violence. The number of jobs decreases

gubernatorial performance, magnified by the degree of state violence.

First, we define “experience.” Next, we elaborate on the

experience-performance relationship by presenting the

theoretical underpinnings supporting our hypotheses. Then, we

describe our case selection and outcome variables, as well as the

operationalization and justification of control variables. Finally, we

present the results, discussion, limitations, and conclusions.

Figure 1 High School Enrollment for Mexican States (1995-2010)