The Reach of the State

• Published date: 01 July 2024
• DOI: http://doi.org/10.1177/00104140231194057
• Author: Charles Chang,Yuhua Wang
• Date: 01 July 2024
• Subject Matter: Articles

Article

Comparative Political Studies

2024, Vol. 57(8) 1243–1275

© The Author(s) 2023

Article reuse guidelines:

sagepub.com/journals-permissions

DOI: 10.1177/00104140231194057

journals.sagepub.com/home/cps

The Reach of the State

Charles Chang

1

and

Yuhua Wang

2



Abstract

We conceptualize the reach of the state by examining how the physical

presence of the state helps the state project its power by signaling state

interests and strength. We present a new measurement strategy to capture

the territorial reach of the state using points-of-interest data provided by

location-based service companies. Our measure exhibits several advantages:

(1) it draws on f‌irm-produced or crowd-sourced (rather than government-

produced) data, (2) it includes highly precise, geo-referenced location in-

formation, which can be aggregated to any geographical or administrative

level, (3) it traces temporal changes, and (4) it covers different types of state

agencies. We illustrate its features using original databases that we compiled

on state agencies in China and other countries. We demonstrate how re-

searchers can use our measure by examining the locations and effects of

coercive organizations and provide our data, code, and a tutorial to help

researchers explore new avenues of inquiry.

Keywords

the reach of the state, state capacity, coercion, big data, China

1

Duke Kunshan University, Kunshan, China

2

Harvard University, Cambridge, MA, USA

Corresponding Author:

Yuhua Wang, Department of Government, Harvard University, 1737 Cambridge Street,

Cambridge, MA 02138, USA.

Email: yuhuawang@fas.harvard.edu

Data Availability Statement included at the end of the article

Heaven is high and the emperor is far away.

Chinese proverb

God is high and the czar is far away.

Russian proverb

Since ancient times, the Chinese and Russians have both demonstrated that

they have f‌igured out that the further away the state is, the more likely they are

to get away with ignoring formal rules. This folk wisdom has inspired social

scientists since at least Weber (1946 [1918], 78), who famously declared that

territorial control is an essential component of state power. Mann (1984,

p. 189) likewise argues that a state’s“infrastructural power”allows it to

implement its “political decisions throughout the realm.”According to Shue

(1988), the “reach of the state”determines how far it can penetrate society to

exert control and regulate social relations.

Wereinvigora te thiscon cept byhighlight ing the importance of the physical

presence of the state. We argue that state infrastructure, embodied in gov-

ernment buildings, bureaucratic off‌ices, and police stations, inf‌luences citizen

behavior by signaling state interest and strength. State infrastructure is

therefore different from state capacity. The state decides to establish a physical

presence in a locality not only because it has the capacity but because it also

has an interest in doing so. A sizable political economy literature analyzes

state territorial control as a strategic choice that maximizes f‌iscal extraction or

political support at the lowest possible cost (e.g., Harbers & Steele, 2020;

Herbst, 2014;Steinberg, 2018).

A large literature links a citizen’s physical distance from the state to how

they perceive its inf‌luence on their life. For instance, Stasavage (2010, p. 628)

demonstrates that areas further from the state’s political center have higher

communication and transportation costs associated with political represen-

tation. Likewise, Brinkerhoff et al. (2018, p. 105) show that locations with

fewer state facilities nearby experience lower levels of service utilization and

satisfaction. Scott (2009, p. 167) similarly argues that people living in more

remote locations that are less accessible to the state are culturally distinct from

groups that are more accessible to the state. We build on this previous research

to theorize about how the presence of state institutions helps the state project

its authority.

Empirically, we introduce a new operationalization of the territorial reach of

the state based on geo-referenced points-of-interest (POI) data related to state

agencies provided by location-based service companies. Our approach helps

researchers address four challenges associated with studying the state’sterritorial

reach. First, our measure is based on f‌irm-produced, street-level geo-survey and

crowdsourcing data, which users update in real time. Firm-produced data

1244 Comparative Political Studies 57(8)

overcome some of the biases prevalent in government-released data, which are

often manipulated or not publicly available (Hollyer et al., 2015;Wallace, 2016).

Second, unlike other measures—such as census, taxation, or court cases, which

are tied to certain administrative levels—our data are geo-referenced with highly

precise geographical coordinates and can be aggregated to any geogra phical or

administrative level or combined with other geo-referenced data (e.g., surveys).

Our measure therefore avoids many measurement errors associated with ag-

gregation (e.g., combining local tax accounts) and enablesresearchers to choose a

theory-driven unit of analysis rather than be forced to focus on a parti cular

administrative level due to data limitations (Soifer, 2019,p.93).Third,while

some popular measures of state reach, such as distance to the capital, do not vary

over time, location-based service companies frequently update their d atabases,

which enables researchers to track changes in state infrastructure. Lastly,our data

cover various types of state agencies, including administrative, legal, coercive,

and f‌iscal. Most prior empirical studies have focused on a particular state func tion,

such as taxation. Our disaggregated measure will shed light on how the state

allocates its infrastructure across multiple sectors (Fukuyama, 2013,p.354).

We illustrate the features of our measure using an original database that we

constructed by collecting over 1 million location points of Chinese gov-

ernment agencies from Amap.com—a location-based service company that

has over 100 million daily users.

1

We demonstrate the reliability and validity

of our measure using data from multiple years and different platforms and

sources. We then compare our data with off‌icial statistics and show that

missing information, caused by government off‌icials’failure to release data on

certain indicators, has systematically biased government statistics in China.

We then use two applications to demonstrate how our measure can help

scholars answer interesting, sometimes unsettled, questions. First, we examine

how a state locates its coercive agencies. We hypothesize that public spaces,

such as parks and squares, provide “focal points”(Schelling, 1960, p. 57) for

citizens to coordinate. In response, a government would establish more co-

ercive agencies to monitor these public spaces to either preemptively or

retrospectively deter social unrests. Analyzing f‌ine-grained data at the

township level in China in 2018, we show a strong positive correlation

between the numbers of public spaces and police stations, even controlling for

population and area.

Second, we use our data to shed light on contradictory empirical f‌indings in

the literature about whether state coercion increases or decreases dissent (or

has no effect)—which Davenport (2007, p. 8) refers to as the “punishment

puzzle.”We argue that prior studies have reached different conclusions partly

because many state repressive behaviors are invisible to the public. For

example, harassment, surveillance, spying, bans, arrests, torture, and murder

are focused on the targeted population and may have a deterrent effect on

those who are directly repressed; their impact on the society at large depends

Chang and Wang 1245