ArticlePDF Available

Abstract and Figures

Less than 30 years after Fukuyama and others declared liberal democracy’s eternal dominance, a third wave of autocratization is manifest. Gradual declines of democratic regime attributes characterize contemporary autocratization. Yet, we lack the appropriate conceptual and empirical tools to diagnose and compare such elusive processes. Addressing that gap, this article provides the first comprehensive empirical overview of all autocratization episodes from 1900 to today based on data from the Varieties of Democracy Project (V-Dem). We demonstrate that a third wave of autocratization is indeed unfolding. It mainly affects democracies with gradual setbacks under a legal façade. While this is a cause for concern, the historical perspective presented in this article shows that panic is not warranted: the current declines are relatively mild and the global share of democratic countries remains close to its all-time high. As it was premature to announce the “end of history” in 1992, it is premature to proclaim the “end of democracy” now.
Content may be subject to copyright.
Full Terms & Conditions of access and use can be found at
https://www.tandfonline.com/action/journalInformation?journalCode=fdem20
Democratization
ISSN: 1351-0347 (Print) 1743-890X (Online) Journal homepage: https://www.tandfonline.com/loi/fdem20
A third wave of autocratization is here: what is
new about it?
Anna Lührmann & Staffan I. Lindberg
To cite this article: Anna Lührmann & Staffan I. Lindberg (2019): A third wave of autocratization is
here: what is new about it?, Democratization
To link to this article: https://doi.org/10.1080/13510347.2019.1582029
© 2019 The Author(s). Published by Informa
UK Limited, trading as Taylor & Francis
Group
Published online: 01 Mar 2019.
Submit your article to this journal
View Crossmark data
A third wave of autocratization is here: what is new about
it?
Anna Lührmann and Staan I. Lindberg
V-Dem Institute/Department of Political Sciences, University of Gothenburg, Gothenburg, Sweden
ABSTRACT
Less than 30 years after Fukuyama and others declared liberal democracys eternal
dominance, a third wave of autocratization is manifest. Gradual declines of
democratic regime attributes characterize contemporary autocratization. Yet, we lack
the appropriate conceptual and empirical tools to diagnose and compare such
elusive processes. Addressing that gap, this article provides the rst comprehensive
empirical overview of all autocratization episodes from 1900 to today based on data
from the Varieties of Democracy Project (V-Dem). We demonstrate that a third wave
of autocratization is indeed unfolding. It mainly aects democracies with gradual
setbacks under a legal façade. While this is a cause for concern, the historical
perspective presented in this article shows that panic is not warranted: the current
declines are relatively mild and the global share of democratic countries remains
close to its all-time high. As it was premature to announce the end of historyin
1992, it is premature to proclaim the end of democracynow.
ARTICLE HISTORY Received 13 September 2018; Accepted 30 January 2019
KEYWORDS Autocratization; democratic backsliding; waves of democratization; democratic erosion;
comparative democratization; democracy; quantitative methods; autocracy; autogolpes
Introduction
The decline of democratic regime attributes autocratization has emerged as a con-
spicuous global challenge. Democratic setbacks in countries as diverse as Brazil,
Burundi, Hungary, Russia, Serbia, and Turkey have sparked a new generation of
studies on autocratization.
1
Two key issues are not yet settled in this reinvigorated eld. First, scholars agree that
contemporary democracies tend to erode gradually and under legal disguise.
2
Demo-
cratic breakdowns used to be rather sudden events for instance military coups
and relatively easy to identify empirically.
3
Now, multi-party regimes slowly become
less meaningful in practice
4
making it increasingly dicult to pinpoint the end of
democracy. Yet, in face of this emerging consensus we lack the appropriate conceptual
and empirical tools to systematically analyse such obscure processes.
© 2019 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.
org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is
properly cited.
CONTACT Anna Lührmann anna.luehrmann@v-dem.net
Supplemental data for this article can be accessed https://doi.org/10.1080/13510347.2019.1582029.
DEMOCRATIZATION
https://doi.org/10.1080/13510347.2019.1582029
The second key issue, partly a product of the rst, is that analysts disagree about how
momentous the current wave of autocratization is. Some draw parallels to the break-
down of democracies in the 1930s and the rise of anti-democratic demagogues.
5
Others maintain that the world is still more democratic,
6
developed
7
and emancipated
8
than ever during the twentieth century. How wide and deep does the current autocra-
tization trend cut?
This article addresses these gaps with a three-pronged strategy. First, it provides a
denition of autocratization as substantial de-facto decline of core institutional require-
ments for electoral democracy. This notion is more encompassing than the frequently
used term democratic backsliding, which suggests an involuntary reversal back to his-
torical precedents. Our notion of democracy is based on Dahls famous conceptualiz-
ation of electoral democracy as polyarchy, namely clean elections, freedom of
association, universal surage, an elected executive, as well as freedom of expression
and alternative sources of information.
9
Second, this article oers a new type of operationalization that in a systematic fashion
captures the conceptual meaning of autocratization as episodes of substantial change
based on data from the Varieties of Democracy Project (V-Dem). This new measure
has four major advantages: It measures what we actually want to study; it is sensitive
to changes in the de-facto implementation of democratic rules; and it is nuanced
enough to also capture gradual autocratization processes and thus avoid biasing the
sample towards fast-moving changes. Finally, it allows us to pinpoint the year of the
onset of autocratization processes, which opens new avenues for empirical studies.
Third, this article employs the new measure in a systematic study that adds a histori-
cal perspective on contemporary autocratization. The resultant ndings are mixed. On
the one hand, we are the rst to show that a third wave of autocratizationaecting an
unprecedented high number of democracies is under way. This wave unfolds slow and
piecemeal making it hard to evidence. Ruling elites shy away from sudden, drastic
moves to autocracy and instead mimic democratic institutions while gradually
eroding their functions. This suggests we should heed the call of alarm issued by
some scholars.
On the other hand, the evidence here also shows that we still live in a democratic era
with more than half of all countries qualifying as democratic. In addition, most episodes
of contemporary autocratization are not only slower, but also feebler than their histori-
cal cousins, at least as of yet. Thus, the aected countries remain more democratic than
their equivalents hit by earlier waves of autocratization.
Below, we rst pursue a review of the literature followed by a reconceptualization of
autocratization with accompanying operationalization, description of data, and coding
procedures. The following section presents a series of descriptive analyses of the three
waves of autocratization, followed by a closer examination of the autocratization of
democracies. The nal section introduces a new metric the rate of autocratization
as an indicator for the pace of such processes. We conclude with a summary of the
ndings and avenues for future research.
State of the art at present
Many have noted that the optimism spurred by the force of the third wave of democra-
tization
10
was premature, including Fukuyamas
11
relegation of the reverse process
autocratization to the history books. A plethora of autocracies deed the trend
12
or
2A. LÜHRMANN AND S. I. LINDBERG
made some half-hearted reforms while remaining in the grey zone between democracy
and autocracy.
13
Yet, when assessments about freedom in retreat
14
or democratic rollback
15
emerged, they were frequently challenged. At the time, global measures of democracy
had merely plateaued and established democracies did not appear to be in distress.
16
Now evidence is mounting that a global reversal is challenging a series of established
democracies, including the United States who were downgraded by both Freedom
House and V-Dem in 2018.
17
Substantial autocratization has been recorded over the
last 10 years in countries as diverse as Hungary, India, Russia, Turkey, and Venezuela.
18
An increasingly bleak picture is emerging on the global state of democracy,
19
even if
some maintain that the achievements of the third wave of democratization are still
noticeable.
20
Waldner and Lust recently concluded that [t]he study of [democratic] backsliding is
an important new research frontier.
21
A series of new studies on autocratization seems
to have generated an emerging consensus on one important insight: the process of auto-
cratization seems to have changed. Bermeo for example suggests a decline of the most
blatant forms of backsliding”–such as military coups and election day vote fraud.
22
Conversely, more clandestine ways of autocratization harassment of the opposition,
subversion of horizontal accountability are on the rise.
23
Svolik similarly argues
that the risk of military coups has declined over time in new democracies, while the
risk of self-coups remains.
24
Mechkova et al. demonstrate that between 2006 and
2016 autocratization mainly maimed aspects such as media freedom and the space
for civil society leaving the institutions of multiparty elections in place.
25
Coppedge
singled out the gradual concentration of power in the executive as a key contemporary
pattern of autocratization next to what he calls the more classicalpath of intensied
repression.
26
Executive aggrandizementis the term Bermeo uses for this process when
elected executives weaken checks on executive power one by one, undertaking a series
of institutional changes that hamper the power of opposition forces to challenge execu-
tive preferences.
27
While the literature thus agrees that the process of autocratization has changed, it does
not yet oer a systematic way of measuring the new mode of autocratization. The contri-
butions build on case examples,
28
statistics on selected indicators of gradual autocratiza-
tion that is, military coups and electoral fraud,
29
opinion polls,
30
or on changes in
quantitative measures over a set time period.
31
Most existing studies on the causes of
autocratization
32
as well as descriptive overviews
33
are also biased in that they include
only cases of complete breakdown of democracies. Such binary approaches not only
fail to capture the often protracted, gradual and opaque processes of contemporary
regime change,
34
but also exclude important variations: autocratization in democracies
that have not (yet) lead to complete breakdown (for example Hungary) and reversals
in electoral autocracies that never became democracies (for example Sudan).
This is important because the archetype of dramatic reversals to closed autocracy is
becoming rare as are closed autocracies. About half of all countries were closed auto-
cracies in 1980, but by 2017 they only make up 12% of regimes in the world.
35
Contem-
porary autocrats have mastered the art of subverting electoral standards without
breaking their democratic façade completely.
36
Some have labelled this phenomenon
illiberal democracy.
37
Hence, as of 2017 a majority of countries still qualify as democ-
racies (56%) and the most common form of dictatorship (32%) are the electoral
autocracies.
38
DEMOCRATIZATION 3
This dominance of multi-party electoral regimes made other analysists posit that
democracy as a global norm after the end of the Cold War
39
continues to shape expec-
tations and behaviour even of autocrats.
40
If that is true, it does not come as a surprise
that sudden reversals to authoritarianism have grown out of fashion since they involve
the abolishment of multi-party elections in a coup. Such evident violations of demo-
cratic norms carry with them high legitimacy costs.
41
Obviously stolenelections
have triggered mass protests leading up to the colour revolutions.
42
Likewise, the inter-
national community tends to sanction political leaders who explicitly disrespect elec-
toral results, and international aid is often conditioned on a country holding multi-
party elections.
43
For instance, after the Gambian elections in 2016, president
Jammehs refusal to accept defeat was quickly met with a military intervention from
neighbouring countries forcing him into exile.
44
The same seems to apply for military
coups which might explain the sharp drop of coups in recent decades.
45
A gradual transition into electoral authoritarianism is more dicult to pinpoint than
a clear violation of democratic standards, and provides fewer opportunities for domestic
and international opposition. Electoral autocrats secure their competitive advantage
through subtler tactics such as censoring and harassing the media, restricting civil
society and political parties and the undermining the autonomy of election manage-
ment bodies. Aspiring autocrats learn from each other
46
and are seemingly borrowing
tactics perceived to be less risky than abolishing multi-party elections altogether.
Thus, the literatures on autocratization as well as on the global rise of multiparty
elections suggest that the current wave of autocratization unfolds in a more clandestine
and gradual fashion than its historical precedents.
This leads to the next question: If autocratization occurs more gradually does this
also reduce the magnitude of change? Bermeo suggests it does;
47
others entertain
more pessimism in books titled for instance How democracies dieand How democ-
racy ends.
48
Yet, the recent literature on autocratization does not oer ne-grained,
systematic empirical comparisons on this issue either.
Thus, we nd important contributions and emerging propositions in the extant lit-
erature on contemporary autocratization. This article seeks to ll two main gaps. First, it
provides a comprehensive conceptualization of autocratization with an accompanying
operationalization with high validity, which is clearly needed to make future ndings
comparable. Second, we lack a comprehensive empirical analysis diagnosing contem-
porary autocratization in historical perspective: (1) its extent and which types of
regimes are mostly aected compared to previous waves; (2) the nature of how it is
enacted by rulers in comparative perspective; and (3) its pace and magnitude of change.
What is, and is not, autocratization?
Just like with the debate about whether democratization should be understood as a
dierence in kind (countries moving across a qualitative threshold
49
), or in degree
(gradual moves away from pure dictatorship
50
), there are seemingly opposed under-
standings of autocratization. Three dierent terms are commonly used for moves
away from democracy: backsliding, breakdown of democracy, and autocratization.
51
We suggest that it is preferable to conceptualize autocratization the antipode of
democratization as a matter of degree that can occur both in democracies and auto-
cracies. Democracies can lose democratic traits to varying degrees without fully, and
long before breaking down. For instance, it is still an open question if Orbáns model
4A. LÜHRMANN AND S. I. LINDBERG
of illiberal democracyin Hungary will transmute into authoritarianism, and non-
democratic regimes can be placed on a long spectrum ranging from closed autocracies
such as North Korea or Eritrea to electoral autocracies with varying degrees of clo-
seness to democracy such as Nigeria before the 2015 elections. Thus, even most auto-
cracies harbour some democratic regime traits to dierent degrees (for example
somewhat competitive, but far from fully free and fair elections) and can lose them,
such as the 1989 military coup in Sudan when Omar Al-Bashir replaced an electoral
autocracy with one of Africas worst closed dictatorships.
The classic literature focuses on the breakdown of democracies
52
even if some also
identied gradual erosion of democracy in this earlier period.
53
Sudden transitions domi-
nated the moves away from democracy in the 1960s and 1970s making it a proper label
for moves away from democracy at the time. However, the concept of breakdownis
useful only for a subset of possible episodes of autocratization. First, it requires a crisp
approach to the dierence between democracy and dictatorship to enable the identi-
cation of the point of breakdown. That excludes studies of the protracted undermining
of democratic institutions encapsulated by autogolpe and unnished degeneration of
qualities in democracies, as well as the waning away of partial democratic qualities in elec-
toral authoritarian regimes. This is particularly problematic for the contemporary period
when instances of sudden autocratization coups détat for instance are rare.
Some scholars have suggested democratic backsliding to denote the diminishing of
democratic traits. For example, Bermeo denes backsliding as state-led debilitation
or elimination of any of the political institutions that sustain an existing democracy.
54
Waldner and Lust understand backsliding as a deterioration of entails a deterioration
of qualities associated with democratic governance, within any regime(emphasis
added).
55
While we are sympathetic to Waldner and Lusts move away from an exclu-
sive focus on democracies, we nd the use of term backsliding problematic for three
reasons: First, democratic backsliding implies a decline in terms ofdemocracy and
thus a conceptual extension beyond the democratic regime spectrum would border
to conceptual stretching.
56
From our point of view, an already autocratic country
cannot undergo democraticbacksliding into a deeper dictatorship. Second, the
term suggests that regimes slide backto where they were before whereas in reality
they may develop in a new direction, to a dierent form of authoritarianism for
example.
57
Finally, slidingmakes it sound like an involuntary, unconscious process,
which does not do justice to conscious actions political actors take in order to
change a regime. It simply invokes the wrong kind of notion about the process.
Third, we suggest that the overarching, or superior in Sartoris terms, concept is
autocratization.
58
Semantically, it signals that we study the opposite of democratization,
thus describing any move away from [full] democracy.
59
As an overarching concept
autocratization covers both sudden breakdowns of democracy álaLinz and gradual
processes within and outside of democratic regimes where democratic traits decline
resulting in less democratic, or more autocratic, situations (Figure 1). This conceptual-
ization enables us to study both the pace and the methods of bringing a regime closer to
a closed dictatorship, while keeping the distinction between democratic recessions start-
ing in democracies, democratic breakdowns, and further consolidation of already
authoritarian regimes.
To provide a comprehensive denition of autocratization processes, we use the term
democratic recession to denote autocratization processes taking place within democra-
cies, democratic breakdown to capture when a democracy turns into an autocracy, and
DEMOCRATIZATION 5
autocratic consolidation as designation for gradual declines of democratic traits in
already authoritarian situations.
Operationalization and data
Contemporary political science puts a heavy emphasis on identication of causal factors
in experimental research designs. However, we cannot randomly assign either autocra-
tization nor its potential causes to countries. Whether we like it or not, we must rely on
observational data to depict, understand, and explain a phenomenon like autocratiza-
tion. Taking one step back, any causal analysis is predicated on an accurate description
of the outcome: how do we know an autocratization process when we see it? What are
the more useful ways to decipher the dynamics and depict patterns, so as to facilitate
descriptive inferences?
While there is relatively satisfactory data on sudden breakdowns for instance on
military coups
60
and dichotomous measures focusing on transitions from democracy
to autocracy recorded in extant datasets
61
we have lacked suciently nuanced yet sys-
tematic cross-national, times-series data on various aspects of regimes to detail incre-
mental autocratization processes.
This article presents a novel approach identifying autocratization episodes con-
nected periods of time with a substantial decline in democratic regime traits. We use
V-Dems data
62
on 182 countries from 1900 to the end of 2017, or 18,031 country-
years.
63
To identify autocratization episodes, we rely on the Electoral Democracy
Index (EDI, v2x_polyarchy). The EDI captures to what extend regimes achieve the
core institutional requirements in Dahls famous conceptualization of electoral democ-
racy as polyarchy: universal surage, ocials elected in free and fair elections, alterna-
tive sources of information and freedom of speech as well as freedom of association.
64
For present purposes, V-Dems EDI has four key advantages. First, V-Dem data provide
vast temporal and geographical coverage with data reaching back to 1900. Second, the
EDI reects how democratic a political regime is de-facto beyond the mere de-jure pres-
ence of political institutions. Additionally, it has a strong theoretical foundation in
regime attributes that Dahl has identied as core requirements for an electoral democ-
racy.
65
Finally, as a continuous index of de-facto levels of democracy it is sensitive to
gradual and slow-moving autocratization processes.
The EDI runs on a continuous scale from 0 to 1, with higher values indicating a more
democratic dispensation. We operationalize autocratization as a substantial decline on
the EDI (within one year or over a connected time period). A decline is substantial if it
amounts to drop of 0.1 or more on the EDI. The choice of cut-opoint on a continuous
Figure 1. Autocratization as democratization in reverse.
6A. LÜHRMANN AND S. I. LINDBERG
index is naturally arbitrary but a change of 10% seems a reasonable and intuitive choice
for the following reasons. This relatively demanding cut-opoint of 0.1 minimizes the
risk of measurement error driving the results since it requires more of an agreement
among V-Dem coders that declines occurred among the 40 components of the EDI
to achieve this magnitude of dierence on the EDI scale.
66
The cut-opoint should
also be high enough to rule out inconsequential changes but low enough to capture sub-
stantial yet incremental changes that do not amount to a full breakdown. A typical
example would be the series of declines in democratic qualities in Hungary from
2006 to 2017 adding up to drop of the EDI of 0.11. In appendix A4, we demonstrate
the robustness of our main ndings to a higher cut-opoint.
Episodes of autocratization have a start and an end. We proceed in two steps to
identify such episodes. First, we identify potential autocratization episodes, which are
adverse regime change of any magnitude. Second, we exclude all cases that involve
only minor overall change, hence are not really cases of autocratization.
First, a potential autocratization episode starts with a decline on the EDI of 0.01
points or more, from one year to the next. We chose this relatively low threshold in
order to spot the very beginning of incremental autocratization episodes.
67
Second,
we follow the potential episode as long as there is a continued decline, while allowing
up to four years of temporary stagnation (no further decline of 0.01 points on the
EDI) in order to reect the concept of slow-moving processes that can move in ts
and starts with a careful autocrat at the helm. The potential autocratization period
ends when there are no further declines on the EDI of 0.01 or more over four years,
or if the EDI increases by 0.02 points or more during one of those years since the
latter would indicate a potential democratization episode.
68
Second, we calculate the total magnitude of change from the year before the start of
an episode to the end, and record as manifest autocratization episodes only those which
add up to a change of at least 0.1 (10% of the total 01 scale) on the EDI.
69
These coding rules ensure that periods of some ts and starts in what is often a pro-
tracted and messy process, are counted as one episode while at the same time minimiz-
ing the risk that measurement error plays a role in determining when an episode starts
or nishes. Appendix A.E also demonstrates that the main ndings of this article are
robust to modications of these coding rules.
For some analyses, one obviously needs a clear-cut distinction between democracies
and autocracies. Following Lührmann et al.,
70
we dene countries as democracies if they
hold free and fair and de-facto multiparty elections, and achieve at least a minimal level
of institutional guarantees captured by the EDI (universal surage, ocials elected in
multiparty elections, freedom of association and alternative sources of information).
Diagnosing autocratization from 1900 to 2017
Here we present the rst ever comprehensive identication of the 217 autocratization
episodes taking place in 109 countries from 1900 to 2017 (Table A1 in the Appendix)
leaving only 69 states unaected (Table A2 in the Appendix).
71
This count includes 33
countries classied as autocracies in 2017 such as North Korea and Angola who seem to
be caught in an autocracy trapand due to the oor eectnever had much possibility
to become worse. The remaining 36 non-autocratizersare classied as democracies in
2017. This group consists mainly of countries with a long democratic history, such as
Sweden and Switzerland, or that democratized recently, such as Bhutan and
DEMOCRATIZATION 7
Namibia. Additionally, seven countries experienced autocratization solely due to
foreign invasion during the two World Wars.
72
Roughly two-thirds the autocratization episodes (N= 142, 65%) took place in already
authoritarian states. Noteworthy are the many (60) episodes of autocratization in
Africa, most of which occurred in electoral autocracies where autocratization dissipated
initial democratic gains. For instance, three autocratization episodes in Sudan (1958
1959; 1969; 19891990) followed military coups disposing presidents elected in less-
than perfect elections.
About a third of all autocratization episodes (N= 75) episodes started under a demo-
cratic dispensation. Almost all of the latter (N= 60, 80%) led to the country turning into
an autocracy. This should give us great pause about spectre of the current third wave of
autocratization. Very few episodes of autocratization starting in democracies have ever
been stopped before countries become autocracies.
The third wave of autocratization is real and endangers democracies
Huntington conspicuously identied three waves of democratization and two waves of
reversals.
73
Our new measure of autocratization episodes picks up these two reverse
waves and demonstrates that a third wave of autocratization is now unfolding.
For the precise delineation of the reverse waves or waves of autocratization we
deviate slightly from Huntingtons original approach in order to reect our conceptual
and methodological innovations. First, we take as our point of departure democracy in
Dahls understanding as polyarchy.
74
With its seven (later collapsed to six) insti-
tutional requirements it is much more ambitious, and demanding, than Huntingtons
Schumpeterian measure focusing on competition.
75
Second, we are concerned here
with gradual moves away from democracy. Huntington focused in his 1991 book on
the crisp distinctions of democratic transitions and breakdowns. He speaks of a demo-
cratization wave when the transitions to democracy as events outnumber the demo-
cratic breakdowns.
76
Our approach better captures the empirical realities in
particular during recent decades that regime change is typically gradual and slowly
leading to hybridization into electoral authoritarianism instead of sudden, dramatic
transitions. The more sensitive and ne-grained measures we have at our disposal com-
pared to what was available to Huntington, also make it possible to pick up such
dynamic processes in a greater number of countries than Huntington could capture
with binary transitions. Therefore, we use the direction of these changes to delineate
waves of autocratization. We dene as an autocratization wave the time period
during which the number of countries undergoing democratization declines while at
the same time autocratization aects more and more countries.
77
In Figure 2 the dashed grey line represents the number of countries that were aected
by democratization each year.
78
The black, thick line represents the number of countries
that underwent autocratization each year. The black, thin line indicates how many of
the latter started in democracies. Thus, Figure 2 delineates the three waves of autocra-
tization: the rst wave of autocratization occurred roughly from 1926 to 1942 and the
second from 1961 to 1977. The post-cold war democratization surge already slowed
down in the early 1990s and gradually reverse processes began to spread, beginning
in Russia, Armenia and Belarus. Therefore, we can for the rst time show that the
third wave of autocratization already began in earnest in 1994. Notably, his undercur-
rent remained under the radar of most political scientists until Carothers declared The
8A. LÜHRMANN AND S. I. LINDBERG
End of the Transition Paradigmin his seminal 2002 article.
79
By 2017, the third wave
of autocratization dominated with the reversals outnumbering the countries making
progress. This had not occurred since 1940.
The dates for the rst two reverse waves presented here are very similar to Hunting-
tons despite the conceptual and measurement dierences (rst reverse wave 1922
1942; second reverse wave to 19601975). There were 32 autocratization episodes in
the rst wave; 62 episodes during the second reversed wave; and 75 episodes occurring
since the start of the third wave.
80
A list of these episodes is found in Appendix B.
One observation immediately stands out from Figure 2. Whereas the rst reversed
wave aected both democracies and autocracies and the second reversal period
almost only worsened electoral autocracies, almost all contemporary autocratization
episodes aect democracies. We are the rst to show also this systematic dierence.
It is a source of concern especially given the nding reported above that few such epi-
sodes stop short of decent into authoritarianism. At the same time, fewer autocracies are
aected by autocratization, that is, transition from electoral to closed autocracy. This
reects the trend that even in the authoritarian regime spectrum multi-party elections
have become the norm.
81
Post-communist East European countries account for 16 mainly protracted, autocra-
tization episodes in the third wave for example the gradual autocratization processes in
Russia, Hungary, and Poland. The third wave of reversals may still be mounting
aecting as many as 22 countries in 2017. At the same time, the share of countries in
the world that are democratic remains close to its highest ever 53%. To some
extent, the latter explains the former. The more democratic countries there are, the
greater the likelihood that democracies suer setbacks.
In sum, an important characteristic of the third wave of autocratization is unprece-
dented: It mainly aects democracies and not electoral autocracies as the earlier
1. Wave 2. Wave 3. Wave
0
5
10
15
20
25
30
35
40
45
50
55
60
65
70
75
80
Number of countries
1900 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000 2010 2017
Democratization ongoing Autocratization ongoing
Autocratization in (former) democracies
Figure 2. The three waves of autocratization.
DEMOCRATIZATION 9
period and this occurs while the global level of democracy is close to an all-time high.
Hence, for now at least, the trend is manifest, but less dramatic than some claim. This
observation reiterates Brunkert, Kruse and Welzelsnding that while the centennial
democratic trend has passed its climax, the ongoing reverse process remains relatively
mild.
82
In democracies: the third wave of autocratization has a legal facade
Arguably, the loss of democratic traits in regimes that were democratic when an auto-
cratization episode started matters more for the state of democracy in the world than
further deterioration in already autocratic regimes. In this and the next section, we
analyse these 75 episodes of autocratization of democracies in more depth.
The case-based literature suggests that incumbents behind the current processes of
autocratization are using mostly legal means and that illegal power grabs have become
less frequent. We test this proposition by distinguishing between three dierent types of
autocratization strategies based on how they abolish or undermine democratic insti-
tutions. The results are reported in Figure 3. The analysis uses original data covering
all autocratization episodes aecting democracies from 1900 to 2017.
83
The rst and second waves of reversals were almost completely dominated by the
classicform of autocratization tactics of illegal access to power, such as a military
coup (39% of episodes) or foreign invasion (29%), and by autogolpes, where the chief
executive comes to power by legal means but then suddenly abolishes key democratic
Figure 3. Types of autocratization of democracies.
Note: 28 episodes are included in the pre-third wave period, and 47 in the third wave.
10 A. LÜHRMANN AND S. I. LINDBERG
institutions such as elections or parliaments (32%). The paradigmatic example of an
autogolpe is president Fujimoris suspension of the Peruvian constitution and parlia-
ment in 1992.
84
Even Hitler came to power by legal means and then disposed demo-
cratic institutions with the Ermächtigungsgesetz(Enabling Act) in 1933.
Democratic erosion became the modal tactic during the third wave of autocratization.
Here, incumbents legally access power and then gradually, but substantially, undermine
democratic norms without abolishing key democratic institutions. Such processes
account for 70% in the third reversal wave with prominent cases of such gradual
deterioration in Hungary and Poland. Aspiring autocrats have clearly found a new
set of tools to stay in power, and that news has spread.
In democracies: the third wave of autocratization is gradual
We have developed another new metric to measure the rate of autocratization in an
informative way: maximum annual depletion rate. This metric captures how fast
democratic traits decline during an autocratization episode in terms of changes
from one year to the other on the V-Dem EDI. Using the maximum allows us to
distinguish between episodes where a period of gradual declines combines with a
sudden decline in democratic traits; and stretches that consist of gradual declines
only. The advantage of the maximum depletion rate is that a high value indicates
that the episode encompassed a sudden and radical change whereas a low value indi-
cates an autocratization process that was incremental throughout. For ease of
interpretation, we report maximum depletion rate values as a percentage of 1 (the
highest possible score on EDI). Thus, if the maximum change in the EDI from
one year to the next during an autocratization episode was 0.1, the corresponding
autocratization rate is 10%.
For instance, the autocratization episode in Germany from 1930 to 1935 started with
three years of gradual declines during the Weimar Republic. Yet, the main characteristic
of this episode was Hitlers accession to power in 1933 and the subsequent sudden
breakdown of the democratic system. This is reected by a high maximum depletion
rate of 26%. Conversely, chapters such as Turkeys from 2008 to 2017 and Russias
from 1993 to 2017, involve only gradual changes reected by relatively low depletion
rates of 7% (Turkey) and 5% (Russia). Alternative measures of pace such as the average
depletion rate, the annual depletion rate and the decay rate, do not fully capture the
dierence between these two patterns. However, we include those as robustness tests
to the subsequent empirical analysis (see more detailed discussions in Appendix C
and E).
Figure D.1 in Appendix D shows a box plot comparing autocratization during the
three reversal waves using this new metric. The median autocratization rate during
the rst and second waves was 31% and it dropped to 8% in the third wave. At the
bottom end of the scale with a 3.8% maximum depletion rate we nd with the extremely
gradual autocratization process in Philippines from 2001 to 2005, followed by Vanua-
tus spell from 1988 to 1996 at 4.3%. The most sudden breakdowns occurred after the
German invasion in the Czech Republic (55%) and in the Netherlands (52%) during
World War II.
The rate of autocratization of democracies has dropped signicantly (r=0.66,
dashed line on Figure 4) over time. At the same time, the global share of democracies
increased remarkably to hover well-above 50% after the turn of the century (black
DEMOCRATIZATION 11
line). The global share of democracies is negatively and statistically signicantly corre-
lated with the autocratization rate. This relationship holds even when controlling for
important confounders such as GDP, time since transition, level of democracy and
foreign occupation as well as the types of autocratization reported in the prior
section. Based on these regression analyses (results omitted here, see Appendix C),
the rate of autocratization among democracies is predicted to drop from 35% when
few countries were democratic (15%; for example, in the early 1930s) to 10% in 2017
when more than half of the worlds countries were democratic. This nding is robust
to alternative specications of the autocratization rate (Appendix C) as well as of the
autocratization episodes (Appendix E).
However, since we have to rely on observational data and a relatively small number
of cases (75), we need to acknowledge these empirical tests as tentative ndings. Never-
theless as discussed in the literature review there are reasonable intuitions for why a
global rise of democracy should be expected to have a dampening eect on the rate of
autocratization.
This development results in opposite expectations for the further prospects of
democracy. On the one hand, autocratization has become more obscure and therefore
one can suspect less likely to produce triggers for mobilization of pro-democratic forces.
On the other hand, autocratization has also become less severe at least on average in
(former) democracies. Figure 5 illustrates how the eect of autocratization on the level
of democracy has changed over time. The y-axis shows the total EDI drop during an
autocratization episode and the x-axis the EDI score at the nal year of autocratization.
Before 1994, autocratization typically resulted in the dramatic transitions to closed
autocracy with a median EDI score of 0.13 at the end of the episode. During the
Figure 4. Global trend rate of autocratization in democracies and share of democracies.
Note: The autocratization rate captures how fast the V-Dem Electoral Democracy Index declines at the peak of the autocratization
episode in terms of changes from one year to the other. High values indicate sudden autocratization and low values more gradual.
The x-axis of the gure shows the year where the peak of the autocratization rate occurred during the episode.
12 A. LÜHRMANN AND S. I. LINDBERG
third wave of autocratization, the median democracy level at the end of autocratization
episodes remains much higher with a score of 0.45 on the EDI. Also, the median total
decline of democratic attributes during the third wave (0.19) is less than half of the
decline during the pre-third wave period (0.50). This is mainly due to the emergence
of the phenomenon of democratic erosion (33 out of 47 cases, or 70%) in the third wave,
which was not discernible before.
The sudden forms of autocratization invasions, military coups, autogolpes always
result in a democratic breakdown. Even democratic erosion processes are more often
than not lethal for democracy: 18 (55%) of them have resulted in democratic break-
downs; only 5 (15%) processes have stopped before democracy broke down and 10
(30%) were still ongoing in 2017.
85
Conclusion: the third wave of autocratization
This article presents the rst systematic empirical analysis of contemporary autocrati-
zation in historical perspective. The article, rst, contributes with a new method to
identify not only sudden but also gradual autocratization episodes, providing a compre-
hensive empirical overview of adverse regime change from 1900 to today across the
democracy-autocracy spectrum. This new operationalization pinpoints the start and
end year of autocratization processes, which facilitates a new generation of studies
for instance on the drivers of autocratization onset and sequential pathways during
episodes.
Second, we provide evidence that contemporary declines of democracy amount to a
third wave of autocratization. A key nding is that the present reverse wave starting
after 1993 mainly aects democracies, unlike prior waves. What is especially worrying
ARG
ARG
ARG
ARM
AUT
BEL
BEL
CHL
CZE
DNK
EST
FJI
FRA
DEU
GHA
IND
IDN
LVA
LTU
LUX
NLD
NOR
PER
POL
RUS
ESP
LKA
SUR
TUR
URY
VUT
BGD
BLR
BOL BRA
BFA
COM
COM HRV
DOM ECU
FJI
FJI
GHA
HND
HUN
LSO
LBY
MKD
MDG
MWI
MDV
MLI
MDA
MDA
NPL
NIC
NIC
NER
NER
NER
PHL
POL
SRB
SLB
KOR ESP
LKA
THA
THA
TUR
UKR
UKR
VEN
ZMB
Median Pre−Third Wave
Median Third Wave
−.8
−.7
−.6
−.5
−.4
−.3
−.2
−.1
0
Total EDI Drop during Autocratization Episode
0 .1 .2 .3 .4 .5 .6 .7 .8 .9 1
Electoral Democracy Index (EDI, Final Year of Episode)
Pre−Third Wave (<1994) Third Wave (1994−2017)
Figure 5. The consequences of autocratization on the level of democracy.
DEMOCRATIZATION 13
about this trend is that historically, very few autocratization episodes starting in democ-
racies have been stopped short of turning countries into autocracies.
Furthermore, we present a series of descriptive tests corroborating key claims found
in the extent literature but not tested before on systematic evidence: Contemporary
autocratizers mainly use legal and gradual strategies to undermine democracies.
Based on original data, we show that about 68% of all contemporary autocratization epi-
sodes starting in democracies are led by incumbents who came to power legally and
typically by democratic elections. Conversely, during the pre-third wave period most
autocratization episodes included an illegal power grab, such as a military coup.
Whereas autocratizers before the third wave took clearly recognizable moves such as
issuing a new non-democratic constitution or dissolved the legislature, most contem-
porary autocratizers do not change the formal rules. Thus, also the way incumbents
undermine democracy has become more informal and clandestine.
Finally, we devise a new metric the autocratization rate capturing how fast
regimes lose their democratic quality from one year to the other measured as a percen-
tage change of the highest possible value of V-Dems EDI. We can then show that auto-
cratization has become much more gradual than before. Its maximum rate declined
from a median of about 31% in the pre-third wave period to about 8% in the third
wave. This trend is strongly correlated with the changes in the global share of demo-
cratic regimes. As democracy spread around the globe in the 1990s and 2000s, autocra-
tization became more gradual.
By now, most regimes even autocracies hold some form of multiparty elections.
Sudden and illegal moves to autocracy tend to provoke national and international oppo-
sition. The tests we present suggest that contemporary autocratizers have learned their
lesson and thus now proceed in a much slower and much less noticeable way than their
historical predecessors. Thus, while democracy has undoubtedly come under threat, its
normative power still seems to force aspiring autocrats to play a game of deception.
Consequently, states hit by the third wave of autocratization remain much more
democratic than their historical cousins. On the one hand, this gives hope that the
current wave of autocratization might be milder than the rst and second waves. On
the other hand, the third wave may still be picking up. It has aected 22 countries in
2017 and more are on the threshold. For these countries, two scenarios are plausible:
Because autocratization is more gradual, democratic actors may remain strong
enough to mobilize resistance. This happened for instance in South Korea in 2017,
when mass protests forced parliament to impeach the president, which reversed the
prior autocratization trend.
86
Conversely, initial small steps towards autocracy
brought other countries such as Turkey, Nicaragua, Venezuela and Russia on a slip-
pery slope deep into the authoritarian regime spectrum. Future research needs to inves-
tigate what distinguishes these two scenarios and how autocratization can be stopped
and reversed. Yet, one conclusion is clear: As it was premature to announce the end
of historyin 1992, it is premature to proclaim the end of democracynow.
Notes
1. See for example Altman and Perez-Liñan, Explaining the Erosion of Democracy; Bermeo, On
Democratic Backsliding; Cassini and Tomini, Reversing Regimes and Concepts; Coppedge,
Eroding Regimes; Diamond, Facing Up Democratic Recession; Haggard and Kaufmann,
Dictators and Democrats; Levitsky and Ziblatt, How Democracies Die; Lührmann et al., State
14 A. LÜHRMANN AND S. I. LINDBERG
of the World; Mounk, The People vs. Democracy; Runciman, How Democracy Ends; Snyder, On
Tyranny; Tomini and Wagemann, Varieties of Democratic Breakdown; Waldner and Lust,
Unwelcome Change.
2. E.g. Bermeo, On Democratic Backsliding; Runciman, How Democracy Ends.
3. Linz, Breakdown of Democratic Regimes.
4. Lührmann et al., State of the World.
5. Levitsky and Ziblatt, How Democracies Die; Snyder, On Tyranny.
6. Mechkova, Lührmann, and Lindberg, How Much Backsliding?
7. Runciman, How Democracy Ends.
8. Norris, Is Western Democracy Backsliding?; Brunkert et al., Culture-bound Regime
Evolution.
9. Dahl, Polyarchy; Dahl, On Democracy.
10. Huntington, The Third Wave.
11. Fukuyama, The End of History.
12. Svolik, Politics of Authoritarian Rule.
13. Schedler, The Politics of Uncertainty; Diamond, Thinking about Hybrid Regimes.
14. Freedom House, Freedom in the World, 2008.
15. Diamond, The Democratic Rollback; Freedom House, Freedom in the World, 2008.
16. Merkel, Are Dictatorships Returning?; Levitsky and Way, Myth of Democratic Recession.
17. Freedom House, Freedom in the World, 2018.
18. Lührmann et al., State of the World.
19. Diamond, Facing Up Democratic Recession; Levitsky and Ziblatt, How Democracies Die; Kur-
lantzick, Democracy in Retreat.
20. Mechkova, Lührmann, and Lindberg, How Much Backsliding?
21. Waldner and Lust, Unwelcome Change,14.
22. Bermeo, On Democratic Backsliding,6.
23. Bermeo, On Democratic Backsliding,14; Diamond, Facing Up Democratic Recession.
24. Svolik, Which Democracies will Last?
25. Mechkova, Lührmann, and Lindberg, How Much Backsliding?
26. Coppedge, Eroding Regimes.
27. Bermeo, On Democratic Backsliding,10.
28. Levitsky and Ziblatt, How Democracies Die.
29. Bermeo, On Democratic Backsliding.
30. Mounk, The People vs. Democracy.
31. Diamond, Facing Up Democratic Recession; Lührmann et al., State of the World.
32. E.g. Svolik , Authoritarian Reversals and Democratic Consolidation; Bernhard, Nordstrom,
and Reenock, Economic Performance and Democratic Survival; Ulfelder and Lustik, Model-
ing Transitions to Democracy; Przeworski, Democracy and Development.
33. E.g. Merkel, Are Dictatorships Returning?; Erdmann, Decline of Democracy; Levitsky and
Way, Myth of Democratic Recession.
34. Lueders and Lust, Multiple Measurements of Regime Change.
35. Closed autocracies are typically dened in the literature as regimes where the chief executive is
not subjected to de jure multiparty elections. Thus, this category includes monarchies, military
regimes, as well as one-party states.
36. Schedler, The Politics of Uncertainty; Levitsky and Way, Competitive Authoritarianism.
37. Zakaria, The Future of Freedom.
38. Lührmann, Tanneberg, and Lindberg, Regimes of the World.
39. Norris, Does the World Agree?; Hyde, The Pseudo-democratsDilemma.
40. Diamond, Liberal Democratic Order Crisis.
41. Schedler, The Politics of Uncertainty.
42. Bunce and Wolchik, Defeating Dictators; Kuntz and Thompson, More than Final Straw.
43. Kim and Kroeger, Rewarding Introduction of Multiparty Elections.
44. See Searcey and Yaya Barry, As Yahya Jammeh Entered Exile.
45. Bermeo, On Democratic Backsliding.
46. Hall and Ambrosio, Authoritarian Learning.
47. Bermeo, On Democratic Backsliding,17.
48. Levitsky and Ziblatt, How Democracies Die; Runciman, How Democracy Ends.
DEMOCRATIZATION 15
49. See Przeworski, Democracy and Development.
50. See Collier and Adcock, Democracy and Dichotomies; Lindberg, Democracy and Elections in
Africa,2427.
51. While these are the most commonly used terms, it is important to note that others exist as well
such as democratic erosion(Coppedge, Eroding Regimes), de-democratization(Tilly,
Inequality, Democratization, and De-democratization), democratic recession(Diamond,
Facing Up Democratic Recession)orclosing space(Carothers and Brechenmacher,
Closing Space). For a more extensive list of terms used in the debate see Cassani and
Tomini Reversing Regimes and Concepts,4.
52. E.g. Linz, Breakdown of Democratic Regimes.
53. Przeworski, Democracy and Development.
54. Bermeo, On Democratic Backsliding,5.
55. Waldner and Lust, Unwelcome Change,5.3.
56. Sartori, Concept Misinformation in Comparative Politics.
57. Runciman, How Democracy Ends,3.
58. Sartori, Concept Misinformation in Comparative Politics.
59. Lindberg, Democratization by Elections, 12; Cassani and Tomini (Cassani and Tomini, Rever-
sing Regimes and Concepts)dene autocratization positively as a process of regime change
towards autocracy that makes politics increasingly exclusive and monopolistic, and political
power increasingly repressive and arbitrary.This denition diers from our approach to
think about autocratization negatively as a move away from democracy. We prefer our
approach for two reasons. First, it is in-line with the common understanding of autocracy as
non-democracy (e.g. Schedler, The Politics of Uncertainty). Second, our approach allows us to
understand autocratization and democratization as mutually exclusive, which allows us to oper-
ationalize them unambiguously.
60. Powell and Thyne, Global Instances of Coups.
61. E.g. Bernhard, Nordstrom, and Reenock, Economic Performance and Democratic Survival;
Haggard and Kaufman, Dictators and Democrats.
62. Coppedge et al., V-Dem Dataset v8.
63. Approximately half of the indicators in the V-Dem dataset are based on factual information
from ocial documents such as constitutions. The remainder consists of expert assessments
on topics like the quality of elections and de facto compliance with constitutional standards.
On such issues, typically ve experts provide ratings for the country, thematic area and time
period for which they are specialists (Coppedge et al., V-Dem Codebook v8).
64. Dahl, Polyarchy; Dahl, On Democracy; Teorell et al., Measuring Polyarchy Across the Globe.
65. Lührmann et al. (Lührmann et al., State of the World) use V-Dems Liberal Democracy Index
to identify democratic declines. The advantage of this alternative strategy is that it provides an
early warning tool because liberal aspects of democracy often are the rst to erode (see Cop-
pedge, Eroding Regimes). However, the aim of this article is dierent. Namely, we want to
provide a heuristic device, which facilitates the analysis of questions such as how liberal con-
straints inuence the likelihood of autocratization. Therefore, we need to operationalize autocra-
tization in a way that is parsimonious and does not include liberal aspects of democracy.
66. V-Dem aggregates the expert assessments using Bayesian IRT model (Pemstein et al., The V-
Dem Measurement Model; Marquardt and Pemstein, IRT Models).
67. Robustness checks with dierent thresholds yield similar results in regression analysis (see
Appendix A4).
68. A lower threshold of 0.01 for ending episodes would for instance lead to the episode in Russia to
be limited to the years 20112017, even though already the years since 1993 saw major cumu-
lative declines (.19).
69. An alternative option would have been to use a rolling ve-year average change on the EDI as for
instance Coppedge (Coppedge, Eroding Regimes). However, our strategy gives us a start or
end point of more gradual processes.
70. Lührmann, Tannenberg, and Lindberg, Regimes of the World.
71. This count includes only countries still in existence in 2017.
72. Albania, Romania, Belgium, Denmark, France, Netherlands and Norway.
73. Huntington, The Third Wave.
74. Dahl, Polyarchy; Dahl, Democracy and its Critics.
16 A. LÜHRMANN AND S. I. LINDBERG
75. See Doorenspleet (Reassessing the Three Waves) for a similar point.
76. Huntington, The Third Wave, 16.
77. It is somewhat ambiguous to precisely delineate a start and end point of autocratization and
democratization waves every year given that at almost any given year there are a number of
countries changing in both directions. Thus, the dates we suggest here should be understood
to delineate the main part of a wave. But the issue about whether a wave started in exactly
this year or that is not a critical issue in a century-long perspective. We identify an autocratiza-
tion wave starting when the number of democratization episodes begins to decrease at the same
time as autocratization episodes increase for two years in a row. It ends when autocratization
episodes decline in number and democratization episodes increase over the next four years.
78. Following Doorenspleet (Reassessing the Three Waves) we focus our analysis on the number
of countries. In Appendix D we show that the three reverse waves are also manifest when basing
the graphical analysis on the share of countries. Democratization episodes are based on Lindberg
et al. (Successful and Failed Democratization).
79. Carothers, End of the Transition Paradigm.
80. 48 cases occurred in-between the waves.
81. Schedler, The Politics of Uncertainty; Levitsky and Way, Competitive Authoritarianism.
82. Brunkert et al. base their analysis on V-Dem data, but operationalize democratic trends dier-
ently, that is, focusing on a combination of liberal, electoral and participatory aspects. Conse-
quently, their ndings in detail dier from ours. For instance, they argue that the climax of
democracy has been reached in about 2000 when 21% of the worlds population was living in
what they term full democracies. Brunkert et al., Culture-bound Regime Evolution,7.
83. The coding process proceeded in three steps: First, we used V-Dem data to identify whether or
not the appointment of the Head of the Executive involved force (v2expathhs/v2expathhg; Cop-
pedge et al., V-Dem Codebook v8). Second, taking this information into account, a research
assistant coded the four sub-categories based on standard references such as Nohlen and Stöver
(Elections in Europe), and Lentz (Encyclopedia of Heads of states), as well as case specic litera-
ture. Third, we veried the coding choices in particular with regards to borderline cases. Table
A1 in the appendix shows the categorization of individual episodes.
84. Lentz, Heads of States 1945 Through 1992, 633; Levitsky and Ziblatt, How Democracies Die.
85. The following ve episodes stopped before breakdown: Bolivia (2015), Ecuador (2010), Nicar-
agua (1999), South Korea (2014) and Vanuatu (1996). Brazil, Croatia, Dominican Republic,
Ghana, Hungary, Lesotho, Moldova, Niger, Poland and Spain were ongoing in 2017.
86. Shin and Moon, South Korea after Impeachment,130.
Acknowledgements
For helpful comments, we thank Scott Gates, Kyle Marquardt, Jan Teorell, Joe Wright, three anon-
ymous reviewers and participants of the APSA Conferences (8/2017; 8/2018), the ECPR General Con-
ferences (9/2017; 8/2018), the V-Dem Research conference (5/2018) the post-doctoral working group at
the University of Gothenburg and the HU/Princeton workshop on constitutionalism, dissent and
resistance (6/2018), where earlier versions of this article were discussed. In particular, we are grateful
to Rick Morgen who helped to better operationalize our idea of autocratization episodes. We also
beneted immensely from Philipp Tönjesand Sandra Grahns skillful research assistance.
Disclosure statement
No potential conict of interest was reported by the authors.
Funding
This research was supported Vetenskapsrådet [grant number 2018-016114], PI: Anna Lührmann; by
the European Research Council [grant number 72419], PI: Staan I. Lindberg; as well as by internal
grants from the Vice-Chancellorsoce, the Dean of the College of Social Sciences, and the Department
of Political Science at University of Gothenburg.
DEMOCRATIZATION 17
Notes on the contributors
Anna Lührmann is Deputy Director of the V-Dem Institute and Assistant Professor at the Department
of Political Science at the University of Gothenburg.
Staan I. Lindberg is Professor of Political Science, Director of the V-Dem Institute at University of
Gothenburg and one of six Principal Investigators for V-Dem.
ORCID
Anna Lührmann http://orcid.org/0000-0003-4258-1088
Staan I. Lindberg http://orcid.org/0000-0003-0386-7390
Bibliography
Freedom House. Freedom in the World. Freedom House, 2008. https://freedomhouse.org/report/
freedom-world/freedom-world-2008.
Freedom House. Freedom in the World. Freedom House, 2008. https://freedomhouse.org/sites/default/
les/FH_FITW_Report_2018_Final_SinglePage.pdf.
Altman, D., and A. Pérez-Liñán. Explaining the Erosion of Democracy: Can Economic Growth Hinder
Democracy?Varieties of Democracy (V-Dem) Institute Working Paper No. 42, 2017.
Bermeo, N. On Democratic Backsliding.Journal of Democracy 27, no. 1 (2016): 519.
Bernhard, M., T. Nordstrom, and C. Reenock. Economic Performance, Institutional Intermediation,
and Democratic Survival.The Journal of Politics 63, no. 3 (2001): 775803.
Brunkert, L., S. Kruse, and C. Welzel. A Tale of Culture-bound Regime Evolution: The Centennial
Democratic Trend and its Recent Reversal.Democratization 26, no. 3 (2018): 422443. doi:10.
1080/13510347.2018.1542430.
Bunce, V. J., and S. L. Wolchik. Defeating Dictators: Electoral Change and Stability in Competitive
Authoritarian Regimes.World Politics 62, no. 1 (2010): 4386.
Carothers, T. The End of the Transition Paradigm.Journal of Democracy 13, no. 1 (2002): 521.
Cassani, A., and L. Tomini. Reversing Regimes and Concepts: From Democratization to
Autocratization.European Political Science 57, no. 3 (2018): 687716.
Collier, D., and R. Adcock. Democracy and Dichotomies: A Pragmatic Approach to Choices About
Concepts.Annual Review of Political Science 2 (1999): 537565.
Coppedge, M. Eroding Regimes: What, Where, and When?Varieties of Democracy (V-Dem)
Institute Working Paper No. 57, 2017.
Coppedge, M., J. Gerring, C. H. Knutsen, S. I. Lindberg, S.-E. Skaaning, J. Teorell, D. Altman, et al. V-
Dem Codebook v8.Varieties of Democracy (V-Dem) Project (2018).
Coppedge, M., J. Gerring, C. H. Knutsen, S. I. Lindberg, S.-E. Skaaning, J. Teorell, D. Altman, et al. V-
Dem Dataset v8.Varieties of Democracy (V-Dem) Institute (2018).
Dahl, R. A. Democracy and its Critics. New Haven: Yale University Press, 1989.
Dahl, R. A. On Democracy. New Haven, CT: Yale University Press, 1998.
Dahl, R. A. Polyarchy: Participation and Opposition. New Haven, CT: Yale University Press, 1971.
Diamond, L. Facing up to the Democratic Recession.Journal of Democracy 26, no. 1 (2015): 141155.
Diamond, L. The Democratic Rollback.Foreign Aairs 87, no. 2 (2008): 3648.
Diamond, L. The Liberal Democratic Order in Crisis.The American Interest, 2018. https://www.the-
american-interest.com/2018/02/16/liberal-democratic-order-crisis/.
Diamond, L. Thinking about Hybrid Regimes.Journal of Democracy 13, no. 2 (2002): 2135.
Erdmann, G. Decline of Democracy.Comparative Governance and Politics 1 (2011): 2158.
Fukuyama, F. The End of History and the Last Man. New York: Free Press, 1992.
Haggard, S., and R. R. Kaufman. Dictators and Democrats - Masses, Elites and Regime Change.
Princeton, NJ: Princeton University Press, 2016.
Hall, S. G. F., and T. Ambrosio. Authoritarian Learning: A Conceptual Overview.East European
Politics 33, no. 2 (2017): 143161.
Huntington, S. P. The Third Wave of Democratization in the Late Twentieth Century. London:
University of Oklahoma Press, 1991.
18 A. LÜHRMANN AND S. I. LINDBERG
Hyde, S. D. The Pseudo-democrats Dilemma. Why Election Observation Became an International
Norm. Ithaca: Cornell University Press, 2011.
Kim, N. K., and A. Kroeger. Rewarding the Introduction of Multiparty Elections.European Journal of
Political Economy 49 (2017): 164181.
Kurlantzick, J. Democracy in Retreat: The Revolt of the Middle Class and the Worldwide Decline of
Representative Government. New Haven, CT: Yale University Press, 2013.
Lentz, H. M. Encyclopedia of Heads of States and Governnments, 1900 Through 1945.Jeerson, NC:
McFarland & Company, 1999.
Lentz, H. M. Heads of States and Governments - A Worldwide Encyclopedia of Over 2,300 Leaders, 1945
Through 1992.Jeerson, NC: McFarland & Company, 1994.
Levitsky, S., and L. A. Way. Competetive Authoritariansism: Hybrid Regimes after the Cold War,
Problems of International Politics. New York: Cambridge University Press, 2010.
Levitsky, S., and L. A. Way. The Myth of Democratic Recession.Journal of Democracy 26, no. 1
(2015): 4558.
Levitsky, S., and D. Ziblatt. How Democracies Die. London: Penguin Books, 2018.
Lindberg, S. I., P. Lindenfors, A. Lührmann, L. Maxwell, J. Medzihorsky, R. Morgan, and M. C. Wilson.
Successful and Failed Episodes of Democratization: Conceptualization, Identication, and
Description.Varieties of Democracy (V-Dem) Institute Working Paper No. 79, 2018.
Linz, J. J. The Breakdown of Democratic Regimes - Crisis, Breakdown, & Reequilibration. Edited by
J. J. Linz and A. Stepan. Baltimore, MD: The Johns Hopkins University Press, 1978.
Lührmann, A., V. Mechkova, S. Dahlum, L. Maxwell, M. Olin, C. S. Petrarca, R. Sigman, M. C. Wilson,
and S. I. Lindberg. State of the World 2017: Autocratization and Exclusion?Democratization 25,
no. 8 (2018): 13211340.
Lührmann, A., M. Tanneberg, and S. Lindberg. Regimes of the World (RoW): Opening New Avenues
for the Comparative Study of Political Regimes.Politics and Governance 6, no. 1 (2018): 6077.
Marquardt, K., and D. Pemstein. IRT Models for Expert-coded Panel Data.Varieties of Democracy
(V-Dem) Institute Working Paper No. 41, 2017.
Mechkova, V., A. Lührmann, and S. I. Lindberg. How Much Democratic Backsliding?Journal of
Democracy 28, no. 4 (2017): 162169.
Merkel, W. Are Dictatorships Returning? Revisiting the Democratic RollbackHypothesis.
Contemporary Politics 16, no. 1 (2010): 1731.
Mounk, Y. The People vs. Democracy: Why Our Freedom is in Danger. Cambridge, MA: Harvard
University Press, 2018.
Nohlen, D., and P. Stöver. Elections in Europe: A Data Handbook. Nomos, Germany: Baden-Baden, 2010.
Norris, P. Is Western Democracy Backsliding? Diagnosing the Risks.Journal of Democracy (Online
Exchange) (2017). https://www.journalofdemocracy.org/online-exchange-%E2%80%9Cdemocratic-
deconsolidation%E2%80%9D.
Pemstein, D., K. Marquardt, E. Tzelgov, Y. Wang, J. Krusell, and F. Miri. The V-Dem Measurement
Model: Latent Variable Analysis for Cross-national and Cross-temporal Expert-coded Data.
Varieties of Democracy (V-Dem) Institute Working Paper No. 21, 2nd ed., 2017.
Przeworski, A. Democracy and Development: Political Institutions and Material Well-being in the
World, 19501990. Cambridge: Cambridge University Press, 2000.
Runciman, D. How Democracy Ends. London: Prole Books, 2018.
Schedler, A. The Politics of Uncertainty: Sustaining and Subverting Electoral Authoritarianism. Oxford:
Oxford University Press, 2013.
Snyder, T. On Tyranny: Twenty Lessons from the Twentieth Century. 1st ed. New York: Tim Duggan
Books, 2017.
Svolik, M. W. Which Democracies will Last? Coups, Incumbent Takeovers and the Dynamic of
Democratic Consolidation.British Journal of Political Science 45, no. 4 (2015): 715738.
Svolik, M. W. The Politics of Authoritarian Rule. Cambridge: Cambridge University Press, 2012.
Tomini, L., and C. Wagemann. Varieties of Contemporary Democratic Breakdown and Regression: A
Comparative Analysis.European Journal of Political Research 57, no. 3 (2018): 687716.
Waldner, D., and E. Lust. Unwelcome Change: Coming to Terms with Democratic Backsliding.
Annual Review of Political Science 21, no. 1 (2018): 93113.
DEMOCRATIZATION 19
... To contribute to this debate, I bridge scholarship on the late 20th-century transitions from authoritarianism that led only to semi-democratic regimes (Diamond, 2002) and on the more recent processes of backsliding that have resulted in the erosion (yet not the abolishment) of democracy (Lührmann and Lindberg, 2019), and I advance the idea of a 'great convergence' towards hybrid regimes. More specifically, I argue that, rather than an alternation between democratization waves and authoritarian ebbs, the three decades that followed the end of the Cold War could be described more fruitfully as a phase characterized by a tendency of both democracies and autocracies to shift towards hybrid forms of political regime that are neither fully democratic nor fully authoritarian. ...
... Huntington's 'waves and ebbs' model survived these criticisms. During the 2010s, it retook centre stage again as the fear arose that the third wave of democratization, similar to previous waves, could be followed by a new authoritarian tide (Plattner, 2015;Lührmann and Lindberg, 2019). As the wave metaphor returned to the spotlight, a new heated debate unfolded. ...
... Building on scholars who investigated the outcomes of late 20th-century transitions from authoritarian rule as well as of more recent processes of autocratization (Carothers, 2002;Diamond, 2002;Bermeo, 2016;Lührmann and Lindberg, 2019), I argue that the proliferation of so-called hybrid regimes is one such regularity. Accordingly, I contend that rather than an alternation between democratization waves and authoritarian ebbs, the three decades that have followed the end of the Cold War could be described more fruitfully as a phase of 'great convergence' towards hybrid forms of political regime in which institutional elements typical of democracy and authoritarianism coexist. ...
Article
Full-text available
The ‘waves and ebbs’ model proposed by Huntington in his 1991's The Third Wave has profoundly shaped how scholars interpret global trends of democratization and autocratization, but has also received criticisms, especially concerning its ability to explain regime change in the three decades following the end of the Cold War. I contend that, rather than an alternation between democratization waves and authoritarian ebbs, the post-Cold War period could be more fruitfully described as a phase of ‘regime convergence’ characterized by a tendency of both democracies and autocracies to shift towards hybrid forms of political regime. By showing that between 1990 and 2023 transitions to hybrid regimes significantly exceeded transitions in other directions, I demonstrate the empirical relevance of hybridization as a process affecting both democracies and autocracies, and I encourage renewed attention to this phenomenon distinct from both democratization and autocratization.
... Tunisia's democratic backsliding mirrors the contemporary wave of autocratization and is well-suited for a contextual case study of the dynamics of autocratization and resistance. Recent years have seen increased scholarly attention to such autocratization processes, but the existing literature has relatively little to say about the dynamics of resistance (Lührmann & Lindberg 2019). Although civil society is often mentioned as a pivot of anti-autocratic resistance (e.g. ...
... Autocratization changes the formal and informal institutions that regulate how political power is exercised and can occur within minimalist or substantive democracies and in partial or full autocracies (Varieties of Democracy Institute 2023). Quantitative studies show that there is an ongoing global 'wave of autocratization', which is mainly taking the form of gradual changes led by elected leaders who dismantle democratic institutions and norms step by step (Lührmann & Lindberg 2019). This mode of autocratization contrasts with the abrupt transitions to closed autocracy through military coups that were more common during the Cold War (Waldner & Lust 2018;Svolik 2019). ...
... Tunisia's autocratization has thus unfolded as a sequential process that is in broad agreement with contemporary patterns of democratic decline (Lührmann & Lindberg 2019;Lührmann 2021). While the autocratic changes have occurred at a fast pace since July 2021, it has been a partial process that in its first year produced electoral autocracy rather than a closed authoritarian regime. ...
Article
Full-text available
The contemporary rise of autocratization calls for increased attention to the role of civil society in resisting authoritarianism. While the literature on autocratization is growing, there are fewer analyses of the dynamics of civil society resistance. The purpose of the article is to address this knowledge gap through a contextual case study of civil society responses to autocratization in Tunisia. Through a relational analysis of autocratization and resistance, the authors show that Tunisia’s autocratization process was met with resistance but not in the form of a broad-based and forceful resistance movement. The response can instead be characterized as being relatively slow, soft, and fragmented. This finding may be explained with reference to civil society fragmentation and state co-optation, resentments against the democratic system and political elite that emerged from the Arab Spring, and the incremental but swift changes that characterized the autocratization process. The authors conclude that the Tunisian case highlights the need for analytical attention to the question of civil society responses to autocratization but challenges simplistic assumptions about politically unified and effective resistance movements. This in turn calls for further contextual, comparative, and conceptual research on the varieties of resistance against autocratization in different political and societal contexts.
... The far right puts Europe at risk of autocratization; once started, this process almost always leads to autocracies (Lührmann and Lindberg 2019). Therefore, it is increasingly necessary to develop tools that allow researchers to distinguish between radical and extreme parties as well as their commitment to democratic and liberal principles, together with tools for democratic resilience. ...
Article
Full-text available
In contemporary Europe, far-right parties threaten liberal democratic principles such as pluralism, media freedom and minority rights. Despite the stigma they normally face, far-right parties have experienced electoral breakthroughs even in countries where they remained electorally marginal such as Germany, Portugal, Spain and Sweden. We advance the idea that this happened because the level of stigmatization faced by these parties decreased before their electoral breakthrough. Therefore, we form a theoretical framework based on a threefold mechanism: far-right parties manage to reduce the stigma they face because of a reputational shield or by moderating their message; the media help the far right gain visibility and legitimacy by accommodating its views; established parties accommodate far-right parties without ostracizing them. Then, we test the framework by looking at the electoral breakthroughs of four parties: the results confirm the expectations except for the role of established parties, which is inconclusive.
... It is widely argued that democracy is in crisis (Lührmann and Lindberg 2019). According to the latest Democracy Report, published annually by the Varieties of Democracy (V-Dem) Institute, democracy is down to 1985 levels (Nord et al. 2024). ...
... Also, there is a growing acknowledgment that countries may move back and forth along this continuum, and thus find themselves in different categories even within just a few years. As the prominent diagnosis of a "third wave of autocratization" (Lührmann & Lindberg, 2019) suggests, the overall global trend in the more recent past has been clearly towards the autocratic end of the continuum, if starting from very different locations. This general assessment is shared also by most authors that have voiced skepticism about the more particular argument of a third wave. ...
Article
Full-text available
p>Leadership and followership have long been considered to be defining features of democratic politics. However, more recently, both conceptual redefinitions and real-world developments have put issues of leadership and followership in regimes from beyond the family of established liberal democracies center-stage. This article looks into the nature of authoritarian leadership and followership from a comparative perspective and in light of theories of democratic political leadership. As the inquiry suggests, the rise and nature of leadership activities in contemporary authoritarian regimes reflects both the turn towards more competitive types of autocracy and the aspiration of many authoritarian powerholders to be seen as democratic. At the same time, some of the most spectacular manifestations of autocratic leadership relate to democratic rather than established authoritarian regimes. While followers of autocratic leaders can control their leaders only to the extent that a regime provides mechanisms of vertical accountability, authoritarian followers, even in established autocracies, are not in all regards less important or powerful than their democratic counterparts. Many authoritarian followers do not just support autocrats, but actively attack and chase non-followers or followers of other leaders, and thus play an independent role in the legitimation or de-legitimation of leaders and regimes.</p
Chapter
The chapter examines the key features of growing authoritarianism and democratic regression at a global level over the last two decades. The opportunities presented by the COVID-19 pandemic to authoritarian governments to silence dissent, and entrench their hold on power, are analysed, with attention drawn to similarities in the methods employed. In the cases of Ethiopia, Uganda and Rwanda (the focus of this study), during the pandemic their ruling parties resorted to increasingly repressive and violent methods to deal with political opposition, which included the pursuit by Uganda and Rwanda of highly destructive interventions in the neighbouring Democratic Republic of Congo and the launching of a war in Tigray by the Ethiopian government. Such conduct led to Western expressions of opprobrium and threats to withhold critical development assistance. The new Biden administration declared democracy promotion would once again figure prominently as an objective in Washington’s foreign policy and in decisions on aid allocations. The absence of the leaders of these three countries at a Washington summit on democracy in 2021 is highlighted, as are their reactions to their apparent fall from grace in Western eyes.
Article
Full-text available
O artigo examina a conjuntura e o devir da política externa e da inserção internacional brasileira na macrorregião latino-americana, por ocasião do início do novo governo de Brasília, encabeçado pelo presidente Luiz Inácio Lula da Silva. O objetivo geral da pesquisa consiste em analisar as iniciativas sobre política regional apresentadas durante os cem primeiros dias do novo governo de Brasília, especialmente no que tange aos temas da promoção da democracia e de defesa do meio ambiente. Em termos teóricos, a pesquisa é pautada na influência do pluralismo e do neorrepublicanismo nas Relações Internacionais, com destaque para o conceito da paz democrática. Igualmente, é importante assinalar o diálogo com as teorias de alcance intermediário disponíveis para o estudo da inserção de potências regionais, com o ambientalismo global, e com as denominadas teorias não-ocidentais. Metodologicamente, trata-se de um ensaio de interpretação. O texto também se apoia na análise de política externa, formação de cenários prospectivos e na análise documental. As principais fontes utilizadas são documentos diplomáticos, hemerográficos e literatura especializada. Ainda que não isento de contradições e percalços, conclui-se que os resultados operativos observados nos primeiros meses são alentadores para as autoridades político-diplomáticas incumbidas de propor a revisão da agenda.
Chapter
Full-text available
At first glance, some might find it odd that a qualitative researcher like me would be commenting on a section about data aggregation. However, data aggregation as it is called here, or elsewhere called “mapping” the Third Sector has emerged as an important component to my research. As a practice, mapping can georeference information (a literal map). As a metaphor, mapping can signal the gathering and collating of standardized information about the third sector. As an outcome, mapping can facilitate sampling frames and encourage comparative research. . This commentary considers the section of ‘data aggregation’ in this volume.
Book
Full-text available
Why did election monitoring become an international norm? Why do pseudo-democrats-undemocratic leaders who present themselves as democratic-invite international observers, even when they are likely to be caught manipulating elections? Is election observation an effective tool of democracy promotion, or is it simply a way to legitimize electoral autocracies? In The Pseudo-Democrat's Dilemma, Susan D. Hyde explains international election monitoring with a new theory of international norm formation. Hyde argues that election observation was initiated by states seeking international support. International benefits tied to democracy give some governments an incentive to signal their commitment to democratization without having to give up power. Invitations to nonpartisan foreigners to monitor elections, and avoiding their criticism, became a widely recognized and imitated signal of a government's purported commitment to democratic elections. Hyde draws on cross-national data on the global spread of election observation between 1960 and 2006, detailed descriptions of the characteristics of countries that do and do not invite observers, and evidence of three ways that election monitoring is costly to pseudo-democrats: micro-level experimental tests from elections in Armenia and Indonesia showing that observers can deter election-day fraud and otherwise improve the quality of elections; illustrative cases demonstrating that international benefits are contingent on democracy in countries like Haiti, Peru, Togo, and Zimbabwe; and qualitative evidence documenting the escalating game of strategic manipulation among pseudo-democrats, international monitors, and pro-democracy forces.
Article
Full-text available
Using a new measure of “comprehensive democracy,” our analysis traces the global democratic trend over the last 116 years, from 1900 until 2016, looking in particular at the centennial trend’s cultural zoning. As it turns out, democracy has been proceeding and continues to differentiate the world’s nations in a strongly culture-bound manner: high levels of democracy remain a distinctive feature of nations in which emancipative values have grown strong over the generations. By the same token, backsliding and autocratization are limited to cultures with under-developed emancipative values. In line with this finding, public support for democracy neither favours democratization, nor does it prevent autocratization in disjunction from emancipative values. On the contrary, public support for democracy shows such pro-democratic effects if – and only if – it co-exists in close association with emancipative values. The reason is that – in disconnect from emancipative values – support for democracy frequently reverts its meaning, indicating the exact opposite of what intuition suggests: namely, support for autocracy. In conclusion, the prospects for democracy are bleak where emancipative values remain weak.
Article
Full-text available
This article presents evidence of a global trend of autocratization. The most visible feature of democracy – elections – remains strong and is even improving in some places. Autocratization mainly affects non-electoral aspects of democracy such as media freedom, freedom of expression, and the rule of law, yet these in turn threaten to undermine the meaningfulness of elections. While the majority of the world’s population lives under democratic rule, 2.5 billion people were subjected to autocratization in 2017. Last year, democratic qualities were in decline in 24 countries across the world, many of which are populous such as India and the United States. This article also presents evidence testifying that men and wealthy groups tend to have a strong hold on political power in countries where 86% of the world population reside. Further, we show that political exclusion based on socio-economic status in particular is becoming increasingly severe. For instance, the wealthy have gained significantly more power in countries home to 1.9 billion of the world’s population over the past decade.