The Factional Logic of Political Protection in Authoritarian Regimes

• Published date: 01 January 2025
• DOI: http://doi.org/10.1177/00104140241237454
• Author: Duy Trinh
• Date: 01 January 2025

Article

Comparative Political Studies

2025, Vol. 58(1) 155–189

© The Author(s) 2024

Article reuse guidelines:

sagepub.com/journals-permissions

DOI: 10.1177/00104140241237454

journals.sagepub.com/home/cps

The Factional Logic of

Political Protection in

Authoritarian Regimes

Duy Trinh

1



Abstract

Existing literature highlights an authoritarian sanction dilemma: Dictators

must deter rent-seeking, yet in doing so they risk antagonizing factional allies.

Using a new dataset of disciplinary investigations within the Chinese and

Vietnamese Communist Party, I show that some dictators navigate this di-

lemma by tailoring the political protection they provide to their followers.

Factional malleability, the extent to which a regime’s factions are formed

around mutable personal connections, moderates the choice of protection

method. In China, where factions are rigid, factional allies’defection threat is

non-credible. Thus, the dictator offers ex post protection, which is more

desirable to him than to his subordinates, by giving delayed, lenient pun-

ishments to investigated off‌icials in factionally-connected provinces. In con-

trast, under Vietnam’s malleable factions, the dictator provides ex ante

protection by excluding the same off‌icials from investigations. The f‌indings

illuminate how authoritarian regimes with similar formal institutions produce

divergent anti-corruption outcomes.

Keywords

China, non-democratic regimes, Vietnam, factionalism, corruption and

patronage

1

Princeton University, Princeton, NJ, USA

Corresponding Author:

Duy Trinh, Niehaus Center for Globalization and Governance, School of Public and International

Affairs, Princeton University, Robertson Hall, Princeton, NJ 08544, USA.

Email: ddtrinh@princeton.edu

Data Availability Statement included at the end of the article

To ward off elite challengers, a dictator needs to cultivate their own faction or

ruling coalition (Svolik, 2012). One way to do so is to reward members of their

faction with rent-seeking opportunities in exchange for loyalty (Bueno De

Mesquita, 2003). However, unpunished rent-seeking by factional followers

leads to high corruption, which in turn has adverse consequences on the

dictator’s survival. This trade-off between punishing and protecting rent-

seekers is the dictator’s sanction dilemma.

The literature argues that dictators ultimately have to make do with a lesser

evil by either protecting or sanctioning their followers (Hollyer &

Wantchekon, 2014). Contrary to this view, I argue that a dictator has yet

another solution to the sanction dilemma at their disposal: They can strate-

gically provide one of two forms of political protection to his allies, namely ex

ante and ex post protection. With ex ante protection, the dictator shields an

off‌icial from being investigated. Ex post protection, on the other hand, occurs

after an investigation has taken place: It lessens the severity of disciplinary

actions but does not prevent them from happening. The choice of protection

method hinges on the malleability of factional membership in the regime.

Holding everything else constant, the dictator prefers to keep an allied bu-

reaucrat in the faction and in the regime. At the same time, they also seek to

minimize the cost of maintaining the bureaucrat’s allegiance. Where factions

are rigid, the high cost of faction-switching preventsallied bureaucrats from

credibly threatening to defect from the dictator’s faction. In consequence, the

dictator engages more in ex post protection, which is comparatively more

benef‌icial for the dictator but less desirable for their allies. In contrast, where

factions are malleable, allied bureaucrats can leverage their credible defection

threat to obtain ex ante protection from the dictator.

I provide evidence for my theory with a comparative analysis of political

protection in the Chinese and Vietnamese Communist Party (CCP and VCP).

The two parties share similar political institutions, socioeconomic conditions,

levels and forms of corruption, survival threats, as well as an underlying

cultural emphasis on informal ties as building blocks of patron-client rela-

tionships. However, on top of these, elite factions in Vietnam and China

exogenously diverge in terms of malleability. Due to unique experiences with

external military threats during the two regimes’revolutionary era, highly

rigid geographic and professional ties become the exclusive form of con-

nections around which elite factions in the CCP solidif‌ied, whereas in

Vietnam’s, factions are more f‌lexible (Trinh, 2020). I collect two original

datasets on Vietnamese politics: a dataset of disciplinary activities conducted

by the VCP’s Central Committees for Discipline Inspection, and a bio-

graphical dataset of the Party’s Politburo members, provincial party secre-

taries, and Central Committee members. Combining these data with existing

biographical data of the CCP elites as well as a recent dataset of anti-

corruption investigations in China, I f‌ind a divergence in investigation

156 Comparative Political Studies 58(1)

outcomes across the two regimes. In China, disciplinary investigations show

evidence of ex post but not ex ante protection. Investigations are no less likely

to take place in provinces whose party secretaries share hometown, work-

place, and education ties with the incumbent CCP General Secretary than in

provinces without such ties; however, once occurred, investigations in the

former set of provinces are more drawn out and result in more lenient

sanctions. In contrast, I observe ex ante but not ex post protection in Vietnam.

Investigations are less likely to occur in provinces with ties to the incumbent

VCP General Secretary; yet once an investigation has taken place, shared ties

no longer explain variations in investigation outcomes. These results remain

statistically and substantively signif‌icant after controlling for provincial-level

variations in socioeconomic conditions, governance capacity and perceived

corruption; individual-level variations across investigated off‌icials; and

investigation-level variations such as case seriousness or authority of in-

vestigation agency.

The paper adds nuances to the debate on the sincerity of anti-corruption

campaigns both specif‌ically in China and Vietnam and more generally across

authoritarian regimes. On one hand, it demonstrates that punishments given

for disciplinary violations are a function of not only the characteristics of the

violations themselves (Cai, 2014) but also those of the violators and their

superiors, including their factional connections. This f‌inding lends support to

the view that authoritarian anti-corruption campaigns are politically motivated

(Lorentzen & Lu, 2018). On the other hand, my theoretical framework also

suggests that political motivations and meritocratic governance goals can

coexist. In fact, through the f‌lexible use of ex ante and ex post protection,

dictators can simultaneously deter rent-seeking activities while maintaining if

not strengthening their faction. More broadly, my novel framework of se-

lective protection enriches the literature on authoritarian resilience by

shedding light on how a dictator maximizes available resources to maintain

their rule. Specif‌ically, selective protection offers a means through which the

dictator tailors the cost of maintaining a winning coalition to their allies’exit

options. Finally, the paper highlights the importance of a regime’s structural

context, in particular the characteristics of elite networks, in mediating intra-

elite conf‌lict outcomes.

Theoretical Argument

The Dictator’s Sanction Dilemma

Dictators cannot rule alone. Their survival is conditional on the support of a

winning coalition of regime elites and their supporters (Bueno De Mesquita,

2003), which altogether constitutes the dictator’s faction (Nathan & Tsai,

1995;Shih, 2008a). In exchange for allegiance, the dictator often needs to

Trinh 157