Inflation and Cross-National Homicide: Assessing Nonlinear and Moderation Effects Across 65 Countries,1965

Abstract

Criminologists have long speculated that economic conditions play a role in driving crime trends. Emerging research finds that inflation rates are associated with crime rates both within the United States and cross-nationally. Inflation is hypothesized to increase crime by incentivizing illegal markets and organized criminal activity and by reducing the legitimacy of social institutions. Existing research on the association between inflation and homicide rates has been limited to single-country studies or multicountry studies consisting of developed countries only. Moreover, there has been limited attention to the potential complexity of this relationship, including whether it is nonlinear, as crime rates may only increase after a certain threshold of inflation is reached, and whether the criminogenic impact of inflation may be moderated by socioeconomic development, as developing countries are anticipated to be more adversely impacted by the criminogenic influence of inflation. Drawing on a sample of 65 economically diverse countries from 1965 to 2015, we find a positive direct relationship between inflation and homicide rates, although we do not find evidence that this association is nonlinear. Finally, contrary to expectation, we find that the inflation–homicide relationship is most impactful in countries with higher levels of development. We discuss these findings in the context of cross-national predictors of crime.

Keywords

homicide trends inflation development cross-national criminology

Social scientists have long theorized that economic conditions such as unemployment rates, deprivation, business cycles, consumer sentiment, and economic growth play a role in driving crime trends (Arvanites & Defina, 2006; Bushway, 2011; Bushway et al., 2012; Chiricos, 1987; Rosenfeld & Fornango, 2007; Rosenfeld & Messner, 2013). Macro-level research has generally found mixed support for the relationship between these economic indicators and crime rates. In recent years, Rosenfeld (2014, 2018) has called attention to the role of inflation (i.e., the decrease in purchasing power of a nation’s currency) as an understudied economic indicator that may serve as a driving force behind crime trends (see also Rosenfeld & Levin, 2016; Rosenfeld et al., 2019). While some research provides support for the proposition that increasing inflation corresponds with rising crime, the research on this topic remains limited and has been characterized as “small, scattered, theoretically underdeveloped, and limited to the United States” (Rosenfeld, 2014, p. 342). In his 2017 Sutherland Address to the American Society of Criminology, Rosenfeld (2018) called for further research into the relationship between inflation and crime, noting that the extant literature “is not a large body of research, and inflation is generally included as a control variable without much theoretical development” (p. 11).

While several studies have advanced our understanding of the macro-level relationship between inflation and crime, research on this relationship remains limited in four substantive ways. First, existing research has largely focused on the relationship between inflation and acquisitive crime. However, in recent decades, a growing body of research has emerged that aims to explain cross-national variation in homicide rates, yet few studies assess whether inflation influences levels of homicide (Rosenfeld, 2014). Indeed, the dearth of attention to the role of inflation is demonstrated by the fact that recent systematic reviews and meta-analyses of the cross-national homicide literature do not list inflation as a correlate of homicide rates (see Nivette, 2011; Trent & Pridemore, 2012). Second, current cross-national research on the relationship between inflation and homicide has continued to focus primarily on the United States and other wealthy, Western democracies that have experienced relatively low inflation rates over recent decades. In contrast, less developed countries often experience high inflation, greater economic deprivation, and higher homicide rates (Ha et al., 2019; Roncaglia de Carvalho et al., 2018; United Nations [UN] Office on Drugs and Crime, 2019). Given research showing that homicide trends between Western and non-Western countries tend to follow divergent paths (Weiss et al., 2016), this raises important questions regarding the extent to which the findings from these studies are generalizable.

Third, existing cross-national research on inflation and homicide typically covers time frames since the 1980s (see Rosenfeld, 2014; Tuttle, 2018), which omits key decades of the 1960s and 1970s that witnessed both higher global inflation (Ciccarelli & Majon, 2010) and higher global homicide rates (LaFree et al., 2015). Hence, extant research has not investigated the link between inflation and homicide rates during periods of dramatic rises and more extreme values in both phenomena. Finally, prior work has largely focused on exploring a direct, linear relationship between inflation and homicide despite Rosenfeld’s (2014) proposition that the relationship between inflation and crime may be (1) nonlinear, such that crime rates only begin to increase after a certain threshold of inflation is reached, or (2) moderated by other economic indicators such as economic development or income levels, which may mitigate the criminogenic influences of inflation.

The current study addresses these gaps in the literature on the relationship between inflation and cross-national homicide rates by (1) investigating the direct association between inflation and homicide rates in a large, diverse sample of both developed and developing countries, over a nearly 5-decade period since 1965, (2) assessing whether the relationship between inflation and homicide is nonlinear, and (3) exploring whether economic development moderates any relationship between inflation and homicide.

Literature Review

Inflation and Crime: A Direct Association

Rosenfeld (2014, 2018) posited several reasons why higher inflation should result in increased crime. First, in times of high inflation, there will be a general decline of confidence in the legitimacy of economic institutions among a country’s population. Foster (1981) suggesed one possible consequence of inflation is that it:

may destroy confidence in the existence of a government in control of our destiny and capable of dealing with the problems that confront us. It creates fear that the social contract under which we have operated in the past is no longer operative. Coupled with that is resentment…Those who feel that resentment seem ready to cast about for any effective means to recapture their lost economic power. (p. 44)

In turn, decreased institutional legitimacy and trust should increase crime by reducing informal social control and conformity to the law (Elgar & Aitken, 2011; LaFree, 1998; Nivette & Eisner, 2013; Tyler, 2006) and by removing legitimate outlets to settle interpersonal disputes and thereby increasing the chances that individuals will seek out alternative means to settle such disputes (Kirk & Papachristos, 2011; Nivette, 2016).

Second, inflation often spurs economic adversity among a country’s population by increasing the prices of goods and services. This adversity may be particularly pronounced among the most economically deprived segments of the population, who lack accumulated wealth and who may rely exclusively on their hourly wages for their subsistence. As the readjustment of wages will usually lag increases in prices, inflation operates as a tax on earnings (Easterly & Fischer, 2001). Increased economic adversity may intensify the propensity for interpersonal violence, as growing resource deprivation enhances frustration and ultimately aggression (Hsieh & Pugh, 1993; Nivette, 2011). Indeed, countries with higher levels of economic deprivation tend to have substantially higher homicide rates (Elgar & Aitken, 2011; Pridemore, 2008; Santos et al., 2018). Third, by driving up prices in legitimate economic markets, inflation may increase crime by enhancing the demand for cheaper goods in illicit markets. While this latter proposition has often been interpreted to impact acquisitive crime rates (Rosenfeld & Levin, 2016; Rosenfeld et al., 2019), a strengthened illicit economy may generate increased violence by promoting organized criminal activity and the interpersonal violence that stems from it (Bagley & Rosen, 2015; Blumstein, 1995). Finally, Devine et al. (1988) suggested that inflation may be criminogenic as “inflation fosters anomie and fuels a general climate of uncertainty and fear” in addition to limiting the ability of the government to control crime through deterrent (e.g., incarceration) and placative (public assistance) mechanisms (p. 408). The results of cross-national research indicate that homicide rates are higher in the face of weak institutional and governmental legitimacy (Nivette, 2016; Nivette & Eisner, 2013; Testa et al., 2017).

These propositions suggest that inflation should have a direct and positive association with homicide rates. Existing research that has explored this relationship at national and subnational levels within the United States generally provides support for this direct positive association. Devine and colleagues (1988) found a positive relationship between inflation and changes in crime rates for homicide, robbery, and burglary in the United States from 1948 to 1985. Rosenfeld and Levin (2016) used Uniform Crime Report data to examine the relationship between five economic indicators (inflation, unemployment, median household income, real gross domestic product [GDP], and consumer sentiment) and acquisitive crime (robbery, burglary, larceny, and motor vehicle theft) in the United States from 1960 to 2012. They found inflation to be the only economic indicator that consistently explained both short- and long-term changes in crime rates. Nunley and colleagues (2016) found a robust, positive link between inflation and four types of property crime (larceny, burglary, motor vehicle theft, and robbery) in the United States from 1950 to 2000. More recently, Rosenfeld and colleagues (2019) examined inflation and acquisitive crime in 17 U.S. cities between 1960 and 2013. While they found an overall positive relationship between inflation and acquisitive crime consistent with prior studies, they also found that the size and significance of this relationship differed across cities.

Several studies conducted in non-Western countries also provide evidence of a positive relationship between inflation and crime. A study on Nigeria between 1970 and 2013 found that each percentage point increase in inflation was associated with a 0.4% increase in armed robbery, 0.3% increase in arson, and 0.2% increase in false pretense/cheating (i.e., lying for the purpose of stealing; Adekoya & Abdul Razak, 2016). Two studies conducted in Pakistan, one between 1984 and 2012 (Ahad, 2016) and another between 1975 and 2007 (Gillani et al., 2009), were consistent in finding that several economic indicators including inflation were associated with both short- and long-term national crime trends. In a study across regions in China, Cheong and Wu (2015) found that crime trends between 1997 and 2007 were strongly and positively associated with inflation. Using national-level data from India between 1991 and 2015, Hazra and Cui (2018) similarly found that macroeconomic conditions, including inflation, were related to crime trends. Interestingly, the authors found evidence for reverse causality, as crime rates also influenced economic outcomes over the long term. Finally, Tang (2009) found that inflation was a particularly strong predictor of long-term crime trends in Malaysia between 1970 and 2006. Overall, the findings across all these studies are consistent in showing a positive association between inflation and crime in non-Western countries.

Although the role of inflation has gained general support as a correlate of crime in single-country studies (Devine et al., 1988; Nunley et al., 2016; Rosenfeld & Levin, 2016; Rosenfeld et al., 2019), there has been limited attention to its role in a cross-national comparative context beyond that of Western democracies. In the first study on this topic, Rosenfeld (2014) found a strong positive association between inflation and both homicide and burglary rates in a sample of 14 nations composed of the United States and European countries from 1981 to 2010. In a study of 31 member countries of the Organization for Economic Cooperation and Development from 1990 to 2005, Tuttle (2018) found that inflation was not associated with changes in homicide within countries over time, although a between-country analysis revealed that countries with high levels of inflation had greater homicide rates. While generally consistent with the findings of single-country studies, the use of samples composed solely of Western democracies in these two studies raises a question regarding the generalizability of these findings beyond such countries.

Inflation and Crime: A Nonlinear Association

While prior research on inflation and homicide has largely focused on determining whether a direct linear association exists, there are reasons to expect a nonlinear relationship between inflation and homicide (Rosenfeld, 2014). In particular, it is plausible that inflation rates must reach a certain threshold in order to become criminogenic. As inflation rises, changes in prices may not generate societal disruptions necessary to spur changes in economic confidence or interpersonal behavior. Rather, it may be that inflation rates need to reach a specific threshold in order to generate a level of disruption that would impact crime rates. As Rosenfeld (2014, p. 347) proposed, “crime rates should not increase appreciably, and may even decline, until inflation reaches a minimum threshold.”

Despite this contention, there has been limited investigation into the possibility of a nonlinear relationship between inflation and crime. Rosenfeld (2014) found no support for a nonlinear association between inflation and either homicide or burglary rates in his study of 14 developed nations. However, Rosenfeld found that robbery rates began to increase after inflation hit a threshold of approximately 3.5%. Other studies focusing on inflation and acquisitive crime rates in the United States have found no evidence of a nonlinear relationship (Rosenfeld et al., 2019).

Inflation and Crime: Moderation by Development

Another proposition that has received limited attention thus far is the possibility that any relationship between inflation and homicide is moderated by other economic conditions, namely the socioeconomic development of a country. Rosenfeld (2014) notes that compared to other economic indicators, such as unemployment or economic growth, the criminogenic effects of inflation are unique because inflation is “widespread, instantaneous, and direct” (p. 349). In other words, while shifts in unemployment or economic growth only impact a segment of a population, inflation should affect all members of society. While inflation does impact all persons within a country, the negative consequences of inflation may not be uniformly distributed, given that its most harmful effects should be greatest for the poor (Easterly & Fischer, 2001). Jaravel (2019) found that inflation is, on average, approximately 0.44% points higher for individuals living in the bottom 20% of an income distribution relative to those at the top 20%. Easterly and Fischer (2001) also found that the poor are more likely than the wealthy to mention inflation as a top national concern using household survey data from 38 countries. Recent work has demonstrated that a similar process exists at the population level as developing countries have higher levels of inflation than developed economies (Roncaglia de Carvalho et al., 2018).

There are reasons to suspect that more developed countries—where the citizens are wealthier, and governments have greater economic resources on average—may be more capable of absorbing shocks related to inflation. First, wealthier populations may be better equipped to weather moments of economic instability by relying on their domestic and international investments, which are often indexed to, or insured from, inflation. In contrast, citizens of developing countries have, on average, less access to global markets and such financial services, which translates into great disparities in the financial inclusion of citizens across countries globally (Demirguc-Kunt & Klapper, 2012). Second, governments of wealthier countries are able to provide more resources to ensure social and economic stability during times of crisis (Benatar et al., 2011). In contrast, developing countries have fewer resources to counter growing inflation and therefore should be more negatively impacted. Indeed, Rosenfeld (2014) has proposed that a fruitful area of future research is to determine whether income conditions the effect of inflation on crime, suggesting that “increasing incomes should weaken the contribution of inflation to crime increases” (p. 362).

Baumer et al. (2013) investigated potential moderators of the link between economic adversity and crime in a sample of 82 U.S. cities between 1980 and 2009. They found that adverse economic conditions were related to both property crime and homicide. While inflation did not consistently yield an independent linear relationship with homicide, the relationship between other economic indicators and crime was conditional on the level of inflation. In particular, inflation was found to amplify the negative association between changes in wages and homicide rates while reducing the strength of the relationship between unemployment and homicide. This latter finding was contradictive of the author’s expectations, who recommended a deeper investigation of the conditional effects underlying the relationship between economic conditions and crime. Still, researchers have yet to investigate whether the relationship between inflation and homicide is moderated by other economic conditions including socioeconomic development at the cross-national level.

Current Study

Existing studies on the relationship between inflation and homicide rates have been limited by primarily focusing on determining whether a direct linear relationship exists in the United States and other developed, Western countries. In this study, we aim to expand upon prior research on the link between inflation and cross-national homicide rates by drawing on data from a more economically diverse sample of 65 countries across a longer time frame (1965–2015). Specifically, we extend the prior research in this area by examining the following three questions:

Is there a direct linear association between inflation and cross-national homicide rates?

Does the relationship between inflation and homicide operate in a strictly linear fashion or is the relationship nonlinear?

Does socioeconomic development moderate any relationship between inflation and homicide?

Data and Methods

This study uses longitudinal data on a variety of country-level indicators including homicide rates, population demographics, and economic conditions. The analytical sample consists of unbalanced panel data for 65 countries up to 51 years between 1965 and 2015. Appendix lists the countries included in the analytical sample, along with a description of the years of data availability for each country. Countries that dissolved during this time frame, such as East Germany, the Soviet Union, Czechoslovakia, and Yugoslavia, are not part of the analytical sample.¹

Dependent Variable: Homicide Rate

Homicide rates reflect the number of homicides per 100,000 population and were collected from the World Health Organization (WHO) Mortality Database. The database is part of a long-standing compilation of the frequency of deaths in each country aggregated by cause. These deaths are recorded by medical professionals and statisticians, who work closely with the WHO to ensure that deaths are recorded accurately and consistently across the world. Causes of deaths are classified according to the International Classification of Diseases, a standardized table listing a code for any possible cause of death. Homicide is defined as any “death caused by injuries inflicted by another person with the intent to injure or kill, by any means, excluding injuries due to legal interventions or operations of war” (WHO, 2014). All homicide counts correspond to actual observed data and do not include any model-based estimates (Kanis et al., 2017).² Homicide rates were log transformed in order to reduce skew and minimize the influence of outliers.

Focal Independent Variable: Inflation

Inflation rates were collected from the World Bank and represent the annual percent change in consumer prices for each country. Price changes are calculated using the Laspeyres price index, which compares the total cost of a set basket of goods and services at current prices with the cost for the same basket in the previous year (Diewert, 1998).

Moderator Variable: Socioeconomic Development

To reflect the multidimensional aspect of development and to avoid issues with multicollinearity, we used principal component analysis to produce a single measure of socioeconomic development (Messner & Rosenfeld, 1997; Weld & Roche, 2017). Consistent with prior research, we measure socioeconomic development using an index composed of several correlated measures including life expectancy, GDP per capita, infant mortality, percent urban, and percent youth (Messner & Rosenfeld, 1997; Savolainen, 2000; Testa et al., 2020; Weld & Roche, 2017; Weiss et al., 2020). The first four variables were collected from the World Bank, while percent youth was collected from the UN World Population Prospects.

Life expectancy measures the average number of years newborns are projected to live if current mortality probabilities of their population were to remain constant. GDP per capita measures the aggregated value of all goods and services produced, divided by the population of a country in a given year. To account for differences in the pricing of the same goods (e.g., inflation, exchange rates), values are adjusted to reflect the corresponding purchasing power of U.S. dollars in 2010. Infant mortality is the rate of newborn deaths before reaching the age of 1 per 1,000 live births. Percent urban reflects the proportion of the population residing in urban areas. Percent youth measures the proportion of individuals between the ages of 15–29 years relative to the entire population in each country–year observation.

We reduced these five measures into a single component of socioeconomic development using principal component analysis (α = .877, eigenvalue = 3.37). The resulting component was strongly and positively correlated with life expectancy (ρ = .921), GDP per capita (ρ = .838), and percent urban (ρ = .707). Conversely, the component was strongly and negatively correlated with infant mortality (ρ = −.849) and percent youth (ρ = −.776) as developed countries tend to have fewer infant deaths and generally older populations. According to this operationalization, higher scores represent greater levels of socioeconomic development. To facilitate the interpretation of the component, values were transformed to a scale ranging from 0 to 10.³

Control Variables

We include three controls that are commonly used in comparative homicide research (Nivette, 2011).⁴ Income inequality is measured by the Gini index obtained from the Standardized World Income Inequality Database (Solt, 2016). The Gini index is a statistical measure of the income distribution, ranging from a hypothetical zero (i.e., perfect equality in the income of a population) to a maximum value of 100 (i.e., perfect income inequality). Percent male measures the percentage of a country’s population that is male and is obtained from the UN World Population Prospects.

Finally, we utilized data from the Polity IV project (Marshall et al., 2019) to generate three dichotomous indicators classifying the system of governance of each country in a given year. The Polity database contains a 21-point scale indicator (−10 to +10), which measures governance on a continuum from autocracy to democracy. We follow the procedure of LaFree and Tseloni (2006) and generate three variables that classify the democracy scale into three mutually exclusive categories: institutionalized democracy (10 points), transitional democracy (1–9 points), and autocracy (−10 to 0 points). An institutionalized autocracy describes a government by a political elite, which restricts competitive political participation and which has a high degree of interference over social and economic activity. An institutionalized democracy is a country where (1) citizens can express their preference for alternative leaders and policies, (2) the power of the executive branch has institutionalized constraints, and (3) civil liberties are guaranteed to all. Third, a transitional democracy designates countries that share some but not all of the characteristics of the above two typical ideal institutionalized systems and is indicative of a process of change in the system of government, which can be socially disruptive and anomic (LaFree & Tseloni, 2006).

Descriptive Statistics

Table 1 presents the descriptive statistics for the 65 countries and 51 years included in the analytic sample. The average homicide rate prior to log transformation is 8.8 deaths per 100,000 population with a standard deviation (SD) of 22.4. The distribution of homicide is characterized by a strong positive skew due to extremely high homicide rates in some country–year observations, for example, Russia in 1996 and Honduras in 2011 (skewness = 6.84, kurtosis = 62.98). In contrast, the log-transformed homicide rate has an approximately normal distribution with an average of 1.05 log deaths per 100,000 population and an SD of 1.35 (skewness = 0.63, kurtosis = 2.44). The average inflation rate is 2.8% per year, with a small number of countries experiencing deflation in some years (e.g., Ireland in 2009, Singapore in 1976, Greece in 2015) and a small number of countries experiencing extremely high levels of inflation such as Peru in 1990 (at 74.8%) and Armenia in 1994 (at 33.7%). The socioeconomic development index is 6.38 on average. Most countries have relatively balanced sex compositions, which vary by only a few percentage points from the mean of 49.2% male. The average Gini index is 35.16. About half of the analytical sample are institutional democracies (48.8%), and most of the remaining cases (40.3%) represent country–years of transitioning governance. Approximately 11% of country–years are autocracies.

Table 1.

Descriptive Statistics.

Variable	Mean	SD	Minimum	Maximum	α/Factor Loadings
Homicide rate (per 100,000)	8.80	22.41	0.07	280.58
Homicide rate (logged)	1.05	1.35	−2.62	5.64
Inflation (%)	0.28	2.31	−0.04	74.82
Development index	6.38	1.76	0	10.00	α = .88
Live expectancy (years)	73.67	5.20	52.57	83.79	.92
GDP per capita (US$)	20,078	17,858	564	91,418	.84
Infant mortality (per 1,000 births)	16.51	16.62	1.90	143.80	−.85
Urban population (%)	67.79	16.31	18.29	100.00	.71
Youth population (%)	23.91	3.73	14.75	34.71	−.78
Male population (%)	49.19	1.02	45.79	52.35
Gini index	35.16	8.80	19.60	60.30
System of governance
Institutional democracy	0.49	0.50	0	1
Institutional autocracy	0.11	0.31	0	1
Transitional democracy	0.40	0.49	0	1

Note. N = 2,122. SD = standard deviation; GDP = gross domestic product.

Analytical Strategy

Our data include longitudinal country-level data from 1965 to 2015 (see Appendix). We utilize a fixed-effects model with country fixed effects to predict the within-country change in homicide rates while controlling for any unobserved time-invariant factors that may confound the relationship between inflation and homicide.⁵ The fixed-effects model estimating the average effect of inflation on homicide rates was specified using the following formula:

L n ({Homicide rate}_{i t}) = β_{0} + β_{1} Inflation + β_{j} X_{i t} + α_{i} + ε_{i t},

where the left-hand side of the equation is the natural log of the homicide rate of country i at time t, β₁is the coefficient for the focal independent variable (inflation), and X_it represents all time-varying controls included in the model. Next, ∊ _it is the error term of each individual data point, and α _i refers to the country fixed effects, which account for time-invariant country-level characteristics. Standard errors were clustered to account for the dependence of year observations within each country (Wooldridge, 2002).

We test if the relationship between inflation and homicide rate is nonlinear by including a squared term of inflation to the original fixed-effects model. An either positive or negative coefficient size for the squared term of inflation is indicative that any effect of inflation on homicide rates is nonlinear.

L n ({Homicide rate}_{i t}) = β_{0} + β_{1} {Inflation}_{i t} + β_{2} {({Inflation}_{i t})}^{2} + β_{j} X_{i t} + α_{i} + ε_{i t} .

Finally, we evaluate the extent to which economic development moderates the association between inflation and homicide rates by including a product-term interaction. A positive coefficient for the interaction term suggests that the effect of inflation on homicides increases with development, while the hypothesized negative interaction term would support the conclusion that increased socioeconomic development mitigates the association between inflation and homicide rates.

L n ({Homicide rate}_{i t}) = β_{0} + β_{1} {Inflation}_{i t} + β_{2} {Development}_{i t} + β_{3} {Inflation}_{i t} \times {Development}_{i t} + β_{j} X_{i t} + α_{i} + ε_{i t} .

Results

Table 2 displays the results of the fixed-effects regression models of the natural log of the homicide rate on inflation and other controls. Model 1 indicates a positive association between inflation and homicide (β = .020, p = .027) with each percentage point increase in inflation corresponding to an increase of 2.06% in the homicide rate. Model 2 includes controls for percentage male, the Gini index, and the system of governance. Of these variables, only the Gini index is associated with the homicide rate (β = −.041, p = .018) with each one-unit increase in the Gini index associated with a 3.98% decrease in the homicide rate. Furthermore, the coefficient for inflation remains approximately the same as in Model 2 (β = .022, p = .014) when compared to Model 1, suggesting that including these control variables does not significantly interfere with the association between inflation and homicide rates. In contrast, the inclusion of the development index as an additional control in Model 3 yields both a sizable and negative association between socioeconomic development and homicide rates (β = −.156, p = .001) as well as a reduction in the size of the coefficient for inflation, which declines from 0.022 in Model 2 to 0.016 in Model 3 (p = .075).⁶

Table 2.

Fixed Effects (FE) Regression of the Natural Log of the Homicide Rate.

Variables	Model 1	Model 2	Model 3
Inflation (%)	.020** (.009)	.022** (.009)	.016* (.009)
Development index			−.156*** (.045)
Male population (%)		−.194 (.193)	−.294* (.152)
Gini index		−.041** (.017)	−.027* (.014)
System of governance (institutional democracy = 0)
Transitional democracy		−.064 (.157)	−.141 (.171)
Institutional autocracy		−.286 (.249)	−.565** (.270)
Constant	1.047*** (0.003)	12.05 (9.424)	17.58** (7.337)
Observations	2,122	2,122	2,122
R ²	.007	.051	.104
Number of countries	65	65	65
Country FE	Yes	Yes	Yes

Note. Robust standard errors clustered by country are in parentheses.

*p < .1. **p < .05. ***p < .01 (two-tailed tests).

Table 3 presents the results of the models testing for nonlinear effects and moderation by socioeconomic development. Model 1 includes a squared term for inflation to test the hypothesis regarding nonlinear effects. The results demonstrate a substantively null and statistically nonsignificant squared term, suggesting that the association between inflation and homicide is not nonlinear. Model 2 tests the moderation hypothesis by interacting inflation with the measure of socioeconomic development. The results demonstrate a positive and statistically significant interaction term (β = .027, p = .030), suggesting that socioeconomic development moderates the relationship between inflation and homicide.⁷

Table 3.

Fixed Effects (FE) Regression of the Natural Log of the Homicide Rate With Interaction and Squared Terms.

Variables	Model 1	Model 2
Inflation centered (%)	.031 (.026)	.079** (.033)
Development index centered (%)	−.154*** (.045)	−.148*** (.046)
Inflation × Development		.027** (.012)
Inflation centered squared (%)	−.0003 (.0003)
Male population (%)	−.299* (.152)	−.313** (.152)
Gini index	−.027* (.015)	−.025 (.015)
System of governance (institutional democracy = 0)
Transitional democracy	−.142 (.171)	−.149 (.171)
Institutional autocracy	−.561** (.271)	−.549** (.273)
Constant	16.86** (7.346)	17.47** (7.312)
Observations	2,122	2,122
R ²	.105	.111
Number of countries	65	65
Country FE	Yes	Yes

Note. Robust standard errors clustered by country are in parentheses.

*p < .1. **p < .05. ***p < .01 (two-tailed tests).

We graphically display the results of this interaction in Figure 1. A few notable patterns emerge. First, the results show that countries with lower levels of socioeconomic development have higher levels of homicide on average. Second, the generally positive relationship between inflation and homicide becomes stronger at higher levels of economic development (i.e., 1 and 2 SDs above the mean). In contrast, the predicted homicide rate remains relatively stable across varying levels of inflation for countries characterized by lower levels of socioeconomic development (i.e., 1 and 2 SDs below the mean). Overall, Figure 1 indicates that the criminogenic effects of inflation become more pronounced as economic development increases such that the association between inflation and homicide is strongest for the most developed country–years in our sample.

Figure 1.

Predicted logged homicide rate by development index by levels of development. Note. SD = standard deviation.

Discussion

The current study advances extant literature on the contribution of economic conditions to cross-national homicide rates by empirically examining the association between inflation, socioeconomic development, and homicide rates in a large sample of countries over a 5-decade period. This study extends prior work on inflation and homicide by (1) using a larger sample, which is composed of both developed and developing countries and (2) exploring this relationship over a longer time range, which includes the high inflationary periods of the 1960s and 1970s (Rosenfeld, 2014; Tuttle, 2018). We examined whether there is a direct association between inflation and cross-national homicide and whether this relationship is nonlinear or moderated by socioeconomic development. This analysis was informed by contemporary research on inflation as a criminogenic force and by theoretical work on the relationship between economic conditions and homicide in a cross-national context.

In regard to our first research question, we found a generally positive association between inflation and cross-national homicide rates. Substantively, our results suggest that inflation is a meaningful predictor of cross-national homicide, with estimates indicating that a one-percentage point increase in inflation rates corresponds to an approximately 2% increase in homicide rates. This result is consistent with prior research showing that inflation is positively associated with cross-national homicide rates (Rosenfeld, 2014). Coupled with extant research, the results from the current study suggest that inflation is an important economic correlate of variation in homicide trends.

This study also considered whether the relationship between inflation and homicide is nonlinear. We found no support for a nonlinear association between inflation and homicide rates in our sample. This result is consistent with prior research at both the cross-national level (Rosenfeld, 2014) and research on inflation and acquisitive crime rates in the United States (Rosenfeld et al., 2019), which has found no evidence for a nonlinear association. This finding suggests that inflation does not necessarily have to reach a certain threshold before generating a risk for increasing homicide. While there is a general lack of evidence of nonlinear effects as demonstrated by multiple studies, future research should investigate whether irregular patterns of inflation correspond with changes in homicide rates. One fruitful area for future research may be to investigate whether homicide rates are impacted by rapid and sustained periods of high inflation. For instance, LaFree and Drass (2002) previously conceptualized homicide booms, which are characterized by (1) positive direction, (2) rapid growth, and (3) sustained change. Although there is limited evidence of a nonlinear association, future research that assesses whether homicide rates increase during periods of rapid, sustained, and positive increases in inflation would be beneficial in further understanding this relationship.

Finally, we examined whether socioeconomic development moderates the relationship between inflation and homicide. We anticipated that inflation would have a greater impact on homicide in less developed countries, given that inflation is typically more harmful to the poor than the wealthy and that these countries would have fewer resources to mitigate the negative ramifications stemming from high levels of inflation. However, the results were contrary to our expectations as inflation appears to have a more criminogenic effect in countries with higher levels of development. While less developed countries have higher rates of homicide on average, the results suggest increasing inflation does not exacerbate levels of homicide in these less developed countries. In contrast, inflation was more strongly associated with homicides at higher levels of development. This finding adds nuance to extant research on the relationship between inflation and homicide, as it raises the possibility that the strong relationship found in prior studies may partially be an artifact of sample selection in that prior studies used samples composed of highly developed countries. Hence, our findings suggest that while there exists a positive association between inflation and homicide rates on average, this association is not necessarily generalizable throughout the world and may, in fact, be stronger in more socioeconomically developed countries.

One potential explanation for this finding may stem from the disadvantage saturation hypothesis (Hannon, 2003). According to this hypothesis, countries at the greatest risk for homicide rates should be least impacted by adverse events such as increases in inflation rates. In other words, the additional strain stemming from growing levels of inflation contributes little to make the criminogenic atmosphere worse in countries already experiencing multiple disadvantages. This proposition is consistent with individual-level research that finds the impact of specific risk factors become muted for individuals who have accumulated a concentration of disadvantages (Hannon, 2003; Raine, 2002; Turanovic, 2019). A similar pattern may exist at a macro-level where additional socioeconomic adversity may have little impact on homicide in the world’s most violent countries, where a multitude of other criminogenic disadvantages (e.g., corruption, relative and absolute deprivation, political and economic instability) mitigates the observable influence of any single factor (Santos et al., 2018).

In contrast, a single indicator of economic adversity—inflation—may be more strongly related to homicide in developed countries, which tend to be more stable in comparison. This is consistent with cross-national homicide research, which finds that individual risk factors have a greater impact on homicide in the safest nations (Santos et al., 2018), where the absence of competing drivers of homicide enables these risk factors to operate and be observed in isolation. The criminogenic effect of inflation may be especially apparent in developed countries where inflation tends to be relatively low and stable (Roncaglia de Carvalho et al., 2018). Accordingly, while periods of high inflation in developed countries may be low relative to inflation levels in developing countries, the fact that individuals living in developed countries are more accustomed to low levels of inflation means that even relatively small increases in inflation rates may be more criminogenic.

There are several limitations of the current study that future research can expand upon. First, the focus of this study was limited to assessing the nexus between inflation, socioeconomic development, and cross-national homicide rates. However, a key proposition linking inflation to increasing homicide rates is through a decline in the legitimacy of institutions among a country’s population. While the mechanism by which inflation contributes to increased homicide rates was outside the scope of our study, future research may assess this proposition by investigating whether the relationship between inflation and homicide is either moderated or mediated by measures such as the level of trust, confidence, and legitimacy of social institutions such as criminal justice systems within countries (LaFree, 1998; Nivette, 2016; Nivette & Eisner, 2013; Testa et al., 2017). Second, this study used only homicide rates to measure crime. Homicide is generally considered the most accurate measure of crime in a cross-national context and is among the only indicators of crime available for a large sample of countries over time (LaFree, 1999; Lynch & Pridemore, 2011). Still, future research can investigate the association between inflation, economic development, and other forms of criminal activity using alternative sources of cross-national data, such as the International Crime Victimization Survey, or other innovative measures of deviance, such as civic honesty (Cohn et al., 2019).

Third, we used country-level socioeconomic development in order to test the proposition that inflation rates would be more harmful in less developed countries. While we believe this measure captures important features of development that may moderate the relationship between inflation and homicide, it would be ideal to have alternative measures such as median levels of income for individuals or households (see Rosenfeld, 2014). Indeed, the selection of variables in the current study was limited to measures that were readily available throughout the time frame. We assessed the possibility of including additional measures in our development index, such as the percent of the population living on less than US$2 per day, literacy rates, and average education levels. However, these measures are available for only a few countries prior to the 1980s. Accordingly, the inclusion of these factors and other similar measures would limit our ability to analyze long-term trends.

Despite these limitations, the findings of this study in conjunction with prior research (Rosenfeld, 2014) suggest that inflation plays a substantive role in explaining shifts in cross-national homicide rates over time. Understanding the contribution of inflation to international homicide trends is an increasingly important endeavor given the growth of a globalized economy and the interconnectedness of economic conditions across countries (Ha et al., 2019; Rosenfeld, 2014). As such, the efforts of central banks to maintain low inflation rates may also serve as effective criminal justice policy, particularly in economically developed countries.

Footnotes

Appendix

Table A1.

Countries and Years in the Analytical Sample.

Country	Total Years	Years		Country	Total Years	Years
Country	Total Years	First	Last	Country	Total Years	First	Last
Armenia	20	1994	2015	Lithuania	21	1995	2015
Australia	48	1967	2015	Malaysia	15	2000	2014
Austria	33	1983	2015	Mauritius	26	1987	2012
Belarus	17	1993	2014	Mexico	51	1965	2015
Belgium	37	1979	2015	Moldova	21	1995	2015
Brazil	35	1981	2015	The Netherlands	39	1977	2015
Bulgaria	26	1989	2014	New Zealand	33	1982	2014
Canada	45	1971	2015	Nicaragua	15	2000	2014
Chile	45	1971	2015	North Macedonia	20	1994	2013
Colombia	40	1970	2015	Norway	46	1970	2015
Costa Rica	50	1965	2014	Panama	40	1970	2015
Croatia	21	1995	2015	Paraguay	22	1994	2015
Czechia	23	1993	2015	Peru	34	1972	2015
Denmark	40	1976	2015	Philippines	30	1965	2011
Dominican Republic	27	1986	2013	Poland	24	1990	2015
Ecuador	29	1987	2015	Portugal	47	1968	2015
Egypt	23	1975	2015	Romania	25	1991	2015
El Salvador	23	1991	2014	Russian Federation	21	1993	2013
Estonia	23	1993	2015	Singapore	43	1973	2015
Finland	50	1966	2015	Slovakia	19	1993	2014
France	51	1965	2015	Slovenia	25	1991	2015
Germany	26	1990	2015	South Africa	20	1996	2015
Greece	42	1974	2015	South Korea	31	1985	2015
Guatemala	31	1981	2014	Spain	42	1974	2015
Hungary	25	1991	2015	Sri Lanka	23	1977	2006
Ireland	43	1973	2015	Sweden	51	1965	2015
Israel	37	1979	2015	Switzerland	36	1980	2015
Italy	49	1967	2015	Thailand	47	1965	2015
Jamaica	17	1988	2014	Trinidad and Tobago	31	1975	2005
Japan	51	1965	2015	United Kingdom	50	1965	2015
Kazakhstan	22	1994	2015	United States	51	1965	2015
Kyrgyzstan	20	1996	2015	Uruguay	32	1981	2015
Latvia	22	1994	2015

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Notes

References

Adekoya

A. F.

Abdul Razak

N. A.

(2016). Inflation, deterrence and crime: Evidence from Nigeria using bounds test approach. Journal of Economics and Sustainable Development, 7(18), 23–32.

Ahad

(2016). Nexus between income inequality, crime, inflation and poverty: New evidence from structural breaks for Pakistan. International Journal of Economics and Empirical Research, 4(3), 133–145.

Andersson

Kazemian

(2017). Reliability and validity of cross-national homicide data: A comparison of UN and WHO data. International Journal of Comparative and Applied Criminal Justice, 42(4), 287–302.

Arvanites

T. M.

Defina

R. H.

(2006). Business cycles and street crime. Criminology, 44(1), 139–164.

Bagley

B. M.

Rosen

J. D.

(2015). Drug trafficking, organized crime, and violence in the Americas today. University Press of Florida.

Baumer

E. P.

Rosenfeld

Wolff

K. T.

(2013). Are the criminogenic consequences of economic downturns conditional? Assessing potential moderators of the link between adverse economic conditions and crime rates. In Rosenfeld

Edberg

Fang

Florence

C. S.

(Eds.), Economics and youth violence: Crime, disadvantage, and community (pp. 53–84). NYU Press.

Benatar

S. R.

Gill

Bakker

(2011). Global health and the global economic crisis. American Journal of Public Health, 101(4), 646–653.

Blumstein

(1995). Youth violence, guns, and the illicit-drug industry. Journal of Criminal Law & Criminology, 86(1), 10–36.

Bushway

S. D.

(2011). Labor markets and crime. In Wilson

J. Q.

Petersilia

(Eds.), Crime and public policy (pp. 183–209). Oxford University Press.

10.

Bushway

S. D.

Cook

P. J.

Phillips

(2012). The overall effect of the business cycle on crime. German Economic Review, 13(4), 436–446.

11.

Chamlin

M. B.

Cochran

J. K.

(2006). Economic inequality, legitimacy, and cross-national homicide rates. Homicide Studies, 10(4), 231–252.

12.

Cheong

T. S.

(2015). Crime rates and inequality: A study of crime in contemporary China. Journal of the Asia Pacific Economy, 20(2), 202–223.

13.

Chiricos

T. G.

(1987). Rates of crime and unemployment: An analysis of aggregate research evidence. Social Problems, 34(2), 187–212.

14.

Ciccarelli

Mojon

(2010). Global inflation. The Review of Economics and Statistics, 92(3), 524–535.

15.

Cohn

Maréchal

M. A.

Tannenbaum

Zünd

C. L.

(2019). Civic honesty around the globe. Science, 365(6448), 70–73.

16.

Demirguc-Kunt

Klapper

(2012). Measuring financial inclusion: The global findex database. World Bank.

17.

Devine

J. A.

Sheley

J. F.

Smith

M. D.

(1988). Macroeconomic and social-control policy influences on crime rate changes, 1948-1985. American Sociological Review, 53(3), 407–420.

18.

Diewert

W. E.

(1998). Index number issues in the consumer price index. Journal of Economic Perspectives, 12(1), 47–58.

19.

Easterly

Fischer

(2001). Inflation and the poor. Journal of Money, Credit and Banking, 33(2), 160–178.

20.

Elgar

F. J.

Aitken

(2011). Income inequality, trust and homicide in 33 countries. European Journal of Public Health, 21(2), 241–246.

21.

Foster

(1981). Who loses from inflation? The Annals of the American Academy of Political and Social Science, 456(1), 32–45.

22.

Gillani

S. Y. M.

Rehman

H. U.

Gill

A. R.

(2009). Unemployment, poverty, inflation and crime nexus: Cointegration and causality analysis of Pakistan. Pakistan Economic and Social Review, 47(1), 79–98.

23.

Kose

M. A.

Ohnsorge

(Eds.). (2019). Inflation in emerging and developing economies: Evolution, drivers, and policies. World Bank.

24.

Hannon

(2003). Poverty, delinquency, and educational attainment: Cumulative disadvantage or disadvantage saturation? Sociological Inquiry, 73(4), 575–594.

25.

Hazra

Cui

(2018). Macroeconomic determinants of crime: Evidence from India. Journal of Quantitative Economics, 16(1), 187–198.

26.

Hsieh

C. C.

Pugh

M. D.

(1993). Poverty, income inequality, and violent crime: A meta-analysis of recent aggregate data studies. Criminal Justice Review, 18(2), 182–202.

27.

Jaravel

(2019). The unequal gains from product innovations: Evidence from the US retail sector. Quarterly Journal of Economics, 134(2), 715–783.

28.

Kanis

Messner

S. F.

Eisner

M. P.

Heitmeyer

(2017). A cautionary note about the use of estimated homicide data for cross-national research. Homicide Studies, 21(4), 312–324.

29.

Karlson

K. B.

Holm

Breen

(2012). Comparing regression coefficients between same-sample nested models using logit and probit: A new method. Sociological Methodology, 42(1), 286–313.

30.

Kirk

D. S.

Papachristos

A. V.

(2011). Cultural mechanisms and the persistence of neighborhood violence. American Journal of Sociology, 116(4), 1190–1233.

31.

Krahn

Hartnagel

T. F.

Gartrell

J. W.

(1986). Income inequality and homicide rates: Cross-national data and criminological theories. Criminology, 24(2), 269–294.

32.

LaFree

(1998). Losing legitimacy: Street crime and the decline of social institutions in America. Westview Press.

33.

LaFree

(1999). A summary and review of cross-national comparative studies of homicide. In Smith

M. D.

Zahn

M. A.

(Eds.), Homicide: A sourcebook of social research (pp. 125–145). Sage.

34.

LaFree

Curtis

McDowall

(2015). How effective are our ‘better angels’? Assessing country-level declines in homicide since 1950. European Journal of Criminology, 12(4), 482–504.

35.

LaFree

Drass

K. A.

(2002). Counting crime booms among nations: Evidence for homicide victimization rates, 1956 to 1998. Criminology, 40(4), 769–800.

36.

LaFree

Tseloni

(2006). Democracy and crime: A multilevel analysis of homicide trends in forty-four countries, 1950-2000. The Annals of the American Academy of Political and Social Science, 605(1), 25–49.

37.

Lynch

J. P.

Pridemore

W. A.

(2011). Crime in international perspective. In Wilson

J. Q.

Petersilia

(Eds.), Crime and public policy (pp. 5–52). Oxford University Press.

38.

Marshall

M. G.

Gurr

T. R.

Jaggers

(2019). Polity IV project: Political regime characteristics and transitions, 1800-2018. Center for Systemic Peace.

39.

Messner

S. F.

Raffalovich

L. E.

Shrock

(2002). Reassessing the cross-national relationship between income inequality and homicide rates: Implications of data quality control in the measurement of income distribution. Journal of Quantitative Criminology, 18(4), 377–395.

40.

Messner

S. F.

Rosenfeld

(1997). Political restraint of the market and levels of criminal homicide: A cross-national application of institutional-anomie theory. Social Forces, 75(4), 1393–1416.

41.

Nivette

A. E.

(2011). Cross-national predictors of crime: A meta-analysis. Homicide Studies, 15(2), 103–131.

42.

Nivette

A. E.

(2016). Institutional ineffectiveness, illegitimacy, and public support for vigilantism in Latin America. Criminology, 54(1), 142–175.

43.

Nivette

A. E.

Eisner

(2013). Do legitimate polities have fewer homicides? A cross-national analysis. Homicide Studies, 17(1), 3–26.

44.

Nunley

J. M.

Stern

M. L.

Seals

R. A.

Zietz

(2016). The impact of inflation on property crime. Contemporary Economic Policy, 34(3), 483–499.

45.

Ouimet

(2012). A world of homicides: The effect of economic development, income inequality, and excess infant mortality on the homicide rate for 165 countries in 2010. Homicide Studies, 16(3), 238–258.

46.

Pridemore

W. A.

(2008). A methodological addition to the cross-national empirical literature on social structure and homicide: A first test of the poverty-homicide thesis. Criminology, 46(1), 133–154.

47.

Pridemore

W. A.

(2011). Poverty matters: A reassessment of the inequality–homicide relationship in cross-national studies. The British Journal of Criminology, 51(5), 739–772.

48.

Raine

(2002). Biosocial studies of antisocial and violent behavior in children and adults: A review. Journal of Abnormal Child Psychology, 30(4), 311–326.

49.

Santos

M. R.

Testa

Porter

L. C.

Lynch

J. P.

(2019). The contribution of age structure to the international homicide decline. PLoS One, 14(10), e0222996.

50.

Santos

M. R.

Testa

Weiss

D. B.

(2018). Where poverty matters: Examining the cross-national relationship between economic deprivation and homicide. The British Journal of Criminology, 58(2), 372–393.

51.

Roncaglia de Carvalho

Ribeiro

R. S.

Marques

A. M.

(2018). Economic development and inflation: A theoretical and empirical analysis. International Review of Applied Economics, 32(4), 546–565.

52.

Rosenfeld

(2014). Crime and inflation in cross-national perspective. Crime and Justice, 43(1), 341–366.

53.

Rosenfeld

(2018). Studying crime trends: Normal science and exogenous shocks. Criminology, 56(1), 5–26.

54.

Rosenfeld

Fornango

(2007). The impact of economic conditions on robbery and property crime: The role of consumer sentiment. Criminology, 45(4), 735–769.

55.

Rosenfeld

Levin

(2016). Acquisitive crime and inflation in the United States: 1960–2012. Journal of Quantitative Criminology, 32(3), 427–447.

56.

Rosenfeld

Messner

S. F.

(2013). Crime and the economy. Sage.

57.

Rosenfeld

Vogel

McCuddy

(2019). Crime and inflation in US cities. Journal of Quantitative Criminology, 35(1), 195–210.

58.

Savolainen

(2000). Inequality, welfare state, and homicide: Further support for the institutional anomie theory. Criminology, 38(4), 1021–1042.

59.

Solt

(2016). The standardized world income inequality database. Social Science Quarterly, 97(5), 1267–1281.

60.

Tang

C. F.

(2009). The linkages among inflation, unemployment and crime rates in Malaysia. International Journal of Economics and Management, 3(1), 50–61.

61.

Testa

Santos

M. R.

Weiss

D. B.

(2020). Incarceration rates and hospital beds per capita: A cross-national study of 36 countries, 1971-2015. Social Science & Medicine, 263, 113262.

62.

Testa

Young

J. K.

Mullins

(2017). Does democracy enhance or reduce lethal violence? Examining the role of the rule of law. Homicide Studies, 21(3), 219–239.

63.

Trent

C. L.

Pridemore

W. A.

(2012). A review of the cross-national empirical literature on social structure and homicide. In Liem

M. A.

Pridemore

W. A.

(Eds.), Handbook of European homicide research (pp. 111–135). Springer.

64.

Turanovic

J. J.

(2019). Heterogeneous effects of adolescent violent victimization on problematic outcomes in early adulthood. Criminology, 57(1), 105–135.

65.

Tuttle

(2018). Specifying the effect of social welfare expenditures on homicide and suicide: A cross-national, longitudinal examination of the stream analogy of lethal violence. Justice Quarterly, 35(1), 87–113.

66.

Tyler

T. R.

(2006). Why people obey the law. Princeton University Press.

67.

United Nations Office on Drugs and Crime. (2019). Global study on homicide 2019. https://www.unodc.org/unodc/en/data-and-analysis/global-study-on-homicide.html

68.

Weiss

D. B.

Santos

M. R.

Testa

Kumar

(2016). The 1990s homicide decline: A western world or international phenomenon? A research note. Homicide Studies, 20(4), 321–334.

69.

Weiss

D. B.

Testa

Santos

M. R.

(2018). Hazardous alcohol drinking and cross-national homicide rates: The role of demographic, political, and cultural context. Journal of Drug Issues, 48(2), 246–268.

70.

Weiss

D. B.

Testa

Santos

M. R.

(2020). Institutional anomie and cross-national differences in incarceration. Criminology, 58(3), 454–484.

71.

Weld

Roche

S. P.

(2017). A matter of time: A partial test of institutional anomie theory using cross-national time use data. Journal of Quantitative Criminology, 33(2), 371–395.

72.

Wooldridge

J. M.

(2002). Econometric analysis of cross section and panel data. MIT Press.

73.

World Health Organization. (2014). Global status report on violence prevention 2014. https://www.who.int/violence_injury_prevention/violence/status_report/2014/en/

Inflation and Cross-National Homicide: Assessing Nonlinear and Moderation Effects Across 65 Countries,1965–2015

Abstract

Keywords

Literature Review

Inflation and Crime: A Direct Association

Inflation and Crime: A Nonlinear Association

Inflation and Crime: Moderation by Development

Current Study

Data and Methods

Dependent Variable: Homicide Rate

Focal Independent Variable: Inflation

Moderator Variable: Socioeconomic Development

Control Variables

Descriptive Statistics

Analytical Strategy

Results

Discussion

Footnotes

Appendix

Declaration of Conflicting Interests

Funding

Notes

References