Strategic Purity and Efficiency in the Motor Carrier Industry: A Multiyear Panel Investigation

• Author: Matthew A. Schwieterman,William A. Muir,Stanley E. Griffis,Yemisi A. Bolumole,Jason W. Miller
• DOI: http://doi.org/10.1111/jbl.12204
• Published date: 01 September 2019
• Date: 01 September 2019

Strategic Purity and Efficiency in the Motor Carrier Industry:

A Multiyear Panel Investigation

William A. Muir

1

, Jason W. Miller

2

, Stanley E. Griffis

2

, Yemisi A. Bolumole

2

, and

Matthew A. Schwieterman

2

1

Naval Postgraduate School

2

Michigan State University

Two questions facing motor carrier managers are (1) whether carriers should specialize in providing full truckload (TL) or less-than-truck-

load (LTL) services vis-

a-vis offering mix of both and (2) whether this decision is contingent on carrier size. Yet, the literature provides

little guidance because research to date has offered contradictory theoretical predictions and inconsistent empirical findings. Drawing on the the-

ory of strategic purity and information processing theory, we explain why service specialization is likely to increase carriers’technical efficiency

and why size will have a more pronounced effect on technical efficiency for carriers specializing in LTL services versus TL services. To test

our theory, we assemble a panel data set from archival government sources regarding general freight motor carriers’provision of LTL and TL

services. We measure carriers’technical efficiency using data envelopment analysis and test our hypotheses by fitting a series of panel data

mixed-effects models. Our results indicate that carriers are most technically efficient when they specialize in one service type. We also find that

size positively affects technical efficiency but only for carriers specializing in LTL services; no returns to scale with regard to technical effi-

ciency exist for carriers specializing in TL services.

Keywords: motor carrier; productivity; efficiency; data envelopment analysis; panel data

INTRODUCTION

A goal of logistics and supply chain management research is to

understand how firms can better leverage resources to improve

their productivity and efficiency (Caplice and Sheffi1994; Froh-

lich and Westbrook 2001; Buttermann et al. 2008). One industry

where productivity and efficiency are of central importance is

motor carriage (Belzer 2000), given that carriers’efficiency has

widespread implications for carriers and shippers. For carriers,

understanding how different strategic decisions affect efficiency

is essential to ensuring financial viability given the industry’s

intense competitive pressures (Nebesky et al. 1995; Fawcett

et al. 2016; Jin et al. 2017). For shippers, understanding factors

that affect carriers’efficiency is important because, as noted by

Matt Harding (a principal at Chainalytics), efficient carriers can

more easily respond to demand fluctuations and changing market

conditions (Bowman 2016).

In spite of this, two important questions have eluded clear

answer. The first concerns whether carriers are more efficient if

they specialize in either full truckload (TL) or less-than-truckload

(LTL) operations versus serving both segments. To date, scholars

have put forward arguments both for and against specialization.

On the one hand, Harmatuck (1981) provide empirical evidence

for the existence of economies of scope (Panzar and Willig

1981) whereby carriers can lower costs by serving both seg-

ments, which is theorized to occur due to freight flow balancing.

On the other hand, the highly distinct nature of TL and LTL

operations (McMullen 2005), suggest a counter-argument that

specialization will increase efficiency. Specialization in TL ver-

sus LTL is highly relevant today given the rise of e-commerce

has resulted in the LTL market growing at a more rapid rate than

the TL market, consequently resulting in many TL carriers enter-

ing the LTL segment or giving serious consideration to entry

(Solomon 2016; Jaillet 2017; Lockridge 2017). The second unan-

swered question concerns whether returns to scale with regard to

output efficiency are greater for LTL-predominant carriers rela-

tive to TL-predominant carriers. Some scholars have found no

evidence of returns to scale in either sector with regard to costs

(Grimm et al. 1989; McMullen and Tanaka 1995), whereas

others studies using LTL samples have found evidence that lar-

ger LTL carriers have lower costs per unit of output than smaller

LTL carriers (Xu et al. 1994; Allen and Liu 1995). Answering

this question is important because it provides managers important

information regarding the expected efficiency benefits from

expanding the scale of their operations.

In this manuscript, we devise and test theory predicting that

efficiency benefits exist from specialization such that the most

efficient carriers will be those that focus on either the TL or LTL

sector. Our theory further explains why “efficiencies of special-

ization”will be contingent on carriers’size for LTL-predominant

carriers. In contrast, we theorize that no such contingency will

exist for TL-predominant carriers. We devise this middle range

theory (Merton 1968; Craighead et al. 2016; Stank et al. 2017)

by drawing on core principles from the theory of strategic purity

(Thornhill and White 2007) and information processing theory

(Galbraith 1977; Tushman and Nadler 1978) by explaining how

these principles apply to the motor carrier industry. To test our

theory, we assemble a five-year balanced panel of N=120 gen-

eral freight motor carriers using archival data made available by

the U.S. Department of Transportation (DOT). We first conduct

Corresponding author:

Jason W. Miller, Department of Supply Chain Management, Eli Broad

College of Business, Michigan State University, 632 Bogue Street

N370, East Lansing, MI 48824, USA; E-mail: mill2831@msu.edu

Disclaimer: The views presented are those of the authors and do not

necessarily represent the views of the U.S. Department of Defense

or its components.

Journal of Business Logistics, 2019, 40(3): 204–228 doi: 10.1111/jbl.12204

© 2019 Council of Supply Chain Management Professionals

a multi-input, multi-output data envelopment analysis (DEA) to

construct carrier-year efficiency scores that capture how well car-

riers transform inputs (labor and capital) into outputs (TL and

LTL miles). We utilize these efficiency scores as the dependent

variable for a series of a mixed-effects panel models (Curran and

Bauer 2011) structured to estimate between-carrier efficiency dif-

ferences as a function of service specialization, with the appro-

priate interaction to test the theorized moderating effect of carrier

size. Results from these models corroborate each of our theoreti-

cal predictions and remain after extensive robustness testing.

The remainder of this manuscript is structured into five sec-

tions. The next section summarizes the relevant literature, fol-

lowed by a section sketching underlying explanations for the

hypothesized predictions. The third section describes the research

design, details data sources, and explains the various measures.

The penultimate section explains our two-stage analysis strategy

and presents results. The final section describes theoretical contri-

butions, describes implications for practitioners, notes limitations,

and suggests future research avenues.

LITERATURE REVIEW

Before we discuss the literature, we wish to clarify several terms

from the economics literature that are relevant to our research.

An efficiency frontier refers to the maximum level of output pos-

sible for a fixed level of input; efficiency frontiers can be esti-

mated when firms (or, more broadly, decision-making units)

produce single or multiple outputs using single or multiple inputs

(Fried et al. 2008). Technical inefficiency refers to the distance

that a firm resides from its industry’sefficiency frontier—the far-

ther a firm is from an efficiency frontier, the more inefficient it is

(Greene 2008). Firms are said to be technically efficient when

they reside on the efficiency frontier (Chen et al. 2015). Produc-

tivity, in contrast, refers to the ratio of outputs to inputs. Impor-

tantly, two firms can be technically efficient but differ in

productivity if there are returns to scale

1

(Fried et al. 2008). One

important way econometric approaches for productivity and effi-

ciency analysis differ is that some techniques assume all firms

are technically efficient (e.g., OLS regression models using para-

metric production or cost functions) while others (e.g., data

envelopment analysis [DEA] and stochastic frontier analysis

[SFA]) allow firms to be technically inefficient (Greene 2008).

As we will expand upon, this difference has important implica-

tions for evaluating questions regarding returns to scale and

specialization.

With these terms defined, we now turn to the extant literature.

Extensive work considers productivity and efficiency in the

motor carrier industry, both before (Roberts 1956; Emery 1965;

Ladenson and Stoga 1974; Koenker 1977; Rakowski 1978; Har-

matuck 1981; Sugrue et al. 1982; Friedlaender and Wang Chiang

1983) and after (McMullen and Stanley 1988; Grimm et al.

1989; Xu et al. 1994; McMullen 2004; Scheraga 2011) industry

deregulation. As such, we do not conduct an all-encompassing

literature review. Rather, we direct attention toward studies con-

ducted after the Motor Carrier Act of 1980 for general freight

motor carriers

2

that mainly (1) examine TL or LTL specialization

or (2) use samples that are well represented by LTL and TL

firms.

3

Studies meeting the first criteria are summarized in

Table 1, whereas studies meeting the second criteria are summa-

rized in Table 2.

Examining Tables 1 and 2, we observe that the extant litera-

ture offers few, if any, consistent conclusions regarding the rela-

tionships between service specialization, size, and productivity

and/or efficiency. Below, we offer several possible explanations

—often methodological in nature—for these inconsistencies. We

begin by calling attention to three features of Table 1. First, two

studies (Harmatuck 1991, 1992) focused on within-firm time-ser-

ies analyses for individual LTL carriers. Thus, these studies can-

not answer our question of interest, which concerns

specialization in LTL or TL services. Second, two other studies

(Ying 1990; Bruning 1992) use only one year of data. While this

allows examination of between-carrier effects, using one year of

data may not be representative of a carriers’specialization

4

because carriers’TL versus LTL emphasis can vary over time.

Furthermore, though McMullen and Lee (1999) use panel data,

their approach pools between-firm and within-firm effects, thus

complicating interpretation of their findings (Certo et al. 2017).

Third, the three studies that include both TL and LTL carriers

entered linear terms in their models for a variable representing

LTL focus (i.e., percentage of total output (TL plus LTL) that

LTL represents). Consequently, these studies do not test for the

existence of economies of specialization,

5

which posit that effi-

ciency would be greatest for carriers primarily offering one ser-

vice.

Turning to Table 2, we call attention to two features. First,

even though these studies feature samples that combine TL and

LTL carriers, only Grimm et al. (1989) and McMullen and

1

Stated differently, technical efficiency concerns whether the

firm is optimally transforming the services of productive

resources into outputs at its current size. If larger size allows one

firm to have greater ability to transform their larger amounts of

inputs into outputs for reasons described by Pratten (1971) and

Penrose (2009), then larger firms that are equally efficient at

transforming inputs into outputs will be more productive than

smaller firms.

2

This results in the exclusion of studies that focused on pro-

ductivity or efficiency of specialized segments such as household

goods (Thomas and Callan 1989; Callan and Thomas 1992).

3

As such, if authors state that samples only contain LTL-domi-

nant firms, they are removed (Xu et al. 1994; Allen and Liu

1995; Nebesky et al. 1995; Giordano 1997, 2008). Similarly, we

do not focus on studies that only look at TL-predominant carriers

(McMullen and Stanley 1988; Soirinsuo and M€

akinen 2011).

4

This parallels argument that a single year of financial perfor-

mance may not be representative of a carriers’underlying finan-

cial performance (Beard 1992; Miller and Saldanha 2016).

5

It should be noted that these studies estimated cost functions.

As such, it is more reasonable to assume that costs will be higher

as carriers focus more on LTL operations due to smaller ship-

ment sizes and the need for greater handling. However, using

only the linear term does not allow for the possibility that costs

would be higher for carriers offering a mix of TL and LTL ser-

vices vis-

a-vis those offering only LTL services.

Strategic Purity and Efficiency 205