Electoral Cycles and Minimum Wages in Argentina

Abstract

We document the evolution of minimum salaries in Argentina during electoral cycles. We argue that incumbent politicians tend to significantly increase the growth rates of minimum salaries a few days before primary elections, and they generally keep them high for a long period until general elections conclude. This is due to the fact that minimum wages are known to be best justified by their social contributions rather than their economic effects. Our findings provide evidence of the strategic tools that are used by incumbent politicians to increase their chances of being re-elected. Our results highlight the importance that incumbents and challengers compete on a level playing field during electoral campaigns, and there are mechanisms in place that guarantee that incumbents do not use strategic tools (such as increases in minimum salaries) for their own benefit just before elections.

1. Introduction

This paper contributes to two strands of the literature. First, Lewis-Beck and Stegmaier (2000) suggest that in developing countries and in young democracies in particular, economic variables (such as minimum wages) may not determine political preferences due to, for example, the inability of the state to effectively communicate its actions. On the other hand, Jimenez (2023) has argued that, recently in Peru, a country that has not updated its minimum salary very frequently in the analyzed period, minimum wages have boosted approval of presidential and executive power, warning that “…incumbent politicians have perverse incentives to increase the minimum wage indiscriminately to increase their chances of getting re-elected…”. However, Jimenez (2023) did not document a specific electoral cycle for minimum salaries in countries such as Argentina, where increases in minimum wages occur very often. It is notable that Brummund and Strain (2020) showed that minimum salaries may have different effects depending on the frequency of adjustment.

Second, there is another strand in the literature that documents the strategic use of different specific actions by politicians during the electoral cycle. These actions are reviewed in Table 1. One of the most recent papers documents the use of corruption lawsuits, demonstrating the existence of an electoral cycle in filing new corruption accusations against and by politicians in the periods immediately preceding an election (Feierherd et al. 2024) in Argentina. Feierherd et al. (2024) investigated the differences between primary and general elections in Argentina from July 2013 to December 2021. They found no significant differences in their empirical results and concluded that the incentive to use corruption accusations is similar across primary and general elections. All of the studies in Table 1 show the presence of political business cycles and the opportunistic timing of other political strategies, but none of them have investigated minimum wage policies. One of the main objectives of this paper is to fill this important empirical gap.

Table 1. Literature review of the use of strategic tools by politicians during political campaigns. Source: Own elaboration.

Reference	Tools/Actions	Countries
Mintz (1988)	Defense spending	Israel and USA
Keech and Pak (1989)	Veterans’ benefit programs	USA
Mintz and Ward (1989)	Military spending	Israel
Dyke (2007)	Criminal justice	USA
Eckstein (2009)	Ethnic policies	Cuba
Teele (2014)	Female suffrage	Britain
Min and Golden (2014)	Electricity losses	India
Foremny and Riedel (2014)	Corporate tax rates	Germany
Figueroa (2016)	Political involvement of bureaucrats	Argentina
Ahnen (2017)	Microcredit lending programs	Argentina
Parthasarathy (2017)	Ethnic quotas	India
Englmaier and Stowasser (2017)	Savings bank lending	Germany
Nanes (2017)	Counterterrorism policies	Israel/Palestine
Álvarez-Mingote (2019)	Remittance recipients	México
Westman et al. (2020)	Contributions of small- and medium-sized enterprises	Canada
Araújo (2021)	Conditional cash transfers	Latin America and Asia
Slee and Desmond (2021)	Evictions	USA
Cooperman (2022)	(Un)Natural disasters	Brazil
Han (2022)	Autocratic redistribution	African countries
Abofarha and Nasreldein (2022)	Austerity economic policies	Brazil, Argentina and Ecuador
Barrientos (2023)	Social assistance	Latin America
Finnegan (2023)	Fossil fuel taxation	High-income democracies
Sanford (2023)	Deforestation	Worldwide
Feierherd et al. (2024)	Corruption lawsuits	Argentina
Ellinas and Lamprianou (2024)	Social protests	Greece
Davies (2025)	Teacher absenteeism	India
Douglas et al. (2025)	Taxes	USA
Hamel (2025)	Local education spending	USA
Klüver (2025)	Social influencers	Germany
Snarski (2025)	Environmental protests	Russia
Su and Buerger (2025)	Traffic fines	USA
Urvoy (2025)	Funding of nonprofits	France
Washida (2025)	Statistical manipulation in autocracies	Autocracies
Landgrave et al. (2025)	Political action committees’ money	USA
Sajuria and Collignon (2025)	Local media and local candidates	UK

The main objective of our paper is to unite the two strands of the literature mentioned above and discover whether the use of minimum wages as a political strategy is influenced by the electoral cycle in Argentina. Argentina was selected as a case study because it updates its minimum wages much more often than Peru, and this provides us with a larger sample size with which we can analyze the relationship of changes in minimum wages versus possible electoral cycles. One interesting feature of Argentina that bolsters our case selection rationale is that it features incumbents from different sides of the ideological spectrum (Feierherd et al. 2024). Finding that the business cycle persists despite the ideological orientation of the president signals broader scope conditions and may challenge certain assumptions about who is more likely to engage in this kind of strategy. Finally, it is important to stress that the government sets the minimum wage (i.e., statutory) in every Latin American country. Argentina has a statutory national minimum wage and sets higher floors for some sectors via collective bargaining. However, although social actors do have a voice in countries where the government sets a statutory minimum wage, in practice, this voice is relatively weak except in those cases where there is an institutionalized tripartite consultation body. Both Peru (where Jimenez (2023) documented the perverse use of minimum wages by incumbent politicians) and Argentina are countries where the minimum wage is set unilaterally by the government without any formal consultation with the so-called “social partners” (see column (3) of Table 1 in Gindling and Ronconi (2025)).

We rely heavily on the methodology employed by Feierherd et al. (2024), who provided a dataset documenting election dates in the period of 2013–2021 in Argentina (see Table 2), which is a crucial part of our main argument. Our paper applies this methodology to define electoral cycles and expands on Jimenez’s (2023) contributions on the effects of minimum wage policies on incumbents’ political support, documenting it in the context of Argentina. We argue that, given our knowledge of the existence of political business cycles and other strategically timed political strategies given in Table 1, and their pervasive nature, incumbents will try to manipulate the minimum wage for electoral gain, exactly as they do with other policies. Schemmel and Picot (2025) showed that the best justification for minimum salaries does not lie in their economic effects but in the respect for social contributors that they express. Therefore, minimum wages are also regarded as paying necessary respect to those who fulfill their obligation to social cooperation (see Schemmel and Picot 2025) and who are involved in issues of justice in public discourses (Biermann and Kalfagianni 2020). These studies suggest that incumbent politicians make use of increases in the minimum wage to increase their chances of being re-elected, not only for their economic effects, but also to show their respect for social contributors, which they use in their public discourses. This justifies the increase in minimum wages by incumbents just a few days before primaries, and they expect this to increase their chances of re-election following the increase in minimum salaries, without needing to wait to feel the economic effects.

Table 2. Election dates in Argentina in the period 2013–2021. Source: Table A9 in Feierherd et al. (2024).

Date	Type
2013-08-11	Primary
2013-10-27	General
2015-08-09	Primary
2015-10-25	General
2015-11-22	Run-off
2017-08-13	Primary
2017-10-22	General
2019-08-11	Primary
2019-10-27	General
2021-09-12	Primary
2021-11-14	General

Figure 1 shows the evolution of nominal growth rates of minimum wages in Argentina from July 2013 to December 2021. In the same figure, we also plot solid (dashed) blue (red) vertical lines corresponding to the dates of primary (general) elections in Table 2. We can observe how, in general, exactly a few days before primary elections, there is a very significant increase in the growth rate of minimum wages, creating an asymmetry before and after primary elections. On the other hand, before and after general elections (red lines), we generally observe that the growth rates of minimum wages exhibit symmetric behavior, and there is no incentive to maintain high growth rates for minimum wages. September 2014 is an outlier in terms of minimum salary growth rates; Argentina was experimenting with an annual inflation of 30%, which was unrelated to elections. Figure 2 shows the evolution of real growth rates in minimum wages, and confirms that September 2014 was an outlier due to the high inflation shown in Figure 1. Moreover, the asymmetric behavior of growth rates before and after primary elections and the symmetrical behavior before and after general elections is generally confirmed, as shown in Figure 1 and Figure 2. It is especially interesting to note that in Figure 2, positive growth rates of real minimum salaries are generally obtained when primary elections are approaching. “Although Figure 2 does not show that real growth rates increase a few days before primary elections as clearly as Figure 1, our empirical analysis will confirm the results in the following sections of the paper”. In short, both Figure 1 and Figure 2 indicate the possible existence of an electoral cycle of minimum wages in Argentina (mainly focused on primary elections).

Figure 1. Solid black line shows the evolution of daily nominal growth rate of minimum salary in Argentina from July 2013 to December 2021 in Argentina. Solid (dashed) blue (red) vertical lines correspond to the dates of primary (general) elections in Table 1. Source: Own elaboration and Ministerio de Capital Humano de Argentina1.

Figure 2. Solid black line shows the evolution of daily real (adjusted for inflation) growth rate of minimum salary in Argentina from July 2013 to December 2021 in Argentina. Solid (dashed) blue (red) vertical lines correspond to the dates of primary (general) elections in Table 1. Source: Own elaboration, Instituto Nacional de Estadística y Censos (INDEC) de Argentina2 and Ministerio de Capital Humano de Argentina.

The plan of the paper is as follows. We first describe our data and methodology. Later, we show our main results, robustness checks, and finally, our conclusions.

2. Materials and Methods

2.1. Data

We obtained our daily nominal and real data on minimum wages between July 2013 and December 20213 from Ministerio de Capital Humano and Instituto Nacional de Estadística y Censos (INDEC) in Argentina. Following Campos-González and Balcombe (2024), we applied growth rates to the minimum salary (denoted GMinimumSalary) in our sample in order to avoid spurious results due to possible nonstationarities. Figure 1 and Figure 2 show the evolution of nominal and real growth rates, respectively.

2.2. Methods

2.2.1. Baseline Model

Following Feierherd et al. (2024), we start by constructing a simple regression model for varying temporal windows on one-day intervals around elections:

$${\mathit{G}\mathit{M}\mathit{i}\mathit{n}\mathit{i}\mathit{m}\mathit{u}\mathit{m}\mathit{S}\mathit{a}\mathit{l}\mathit{a}\mathit{r}\mathit{y}}_{\mathit{t}}={\mathit{\beta }}_0+{\mathit{\beta }}_1\mathit{C}\mathit{a}\mathit{m}\mathit{p}\mathit{a}\mathit{i}\mathit{g}\mathit{n} \mathit{S}\mathit{e}\mathit{a}\mathit{s}\mathit{o}\mathit{n} \mathit{t}+{\mathit{\epsilon }}_{\mathit{t}},$$

(1)

where ${\mathit{G}\mathit{M}\mathit{i}\mathit{n}\mathit{i}\mathit{m}\mathit{u}\mathit{m}\mathit{S}\mathit{a}\mathit{l}\mathit{a}\mathit{r}\mathit{y}}_{\mathit{t}}$ corresponds to nominal/real growth rates of minimum wages, $\mathit{C}\mathit{a}\mathit{m}\mathit{p}\mathit{a}\mathit{i}\mathit{g}\mathit{n} \mathit{S}\mathit{e}\mathit{a}\mathit{s}\mathit{o}\mathit{n} \mathit{t}$ is a dummy variable that indicates whether growth rates of minimum wages correspond to before or after an election (being 1 before election), and ${\mathit{\epsilon }}_{\mathit{t}}$ is a disturbance. When focusing on primary elections, according to our Figure 1 and Figure 2, we expect ${\mathit{\beta }}_1$ = 0 or ${\mathit{\beta }}_1$ < 0 depending on the temporal window on one-day intervals. We argue that growth rates of both nominal and real minimum wages outside electoral windows are lower than during campaign periods, and that there is an initial symmetry since growth rates of minimum wages increase significantly a few days before primary elections, and they remain the same for the same few days after primary elections. However, later we observe an asymmetry since, after those initial days, growth rates are significantly higher after primary elections than before. We used the Argentine electoral law regarding the official start of campaigns to define campaign seasons (Feierherd et al. 2024). We also expected that the effect of primary elections would be different than general elections, since we argue that incumbents increase growth rates as soon as primary elections approach, causing an important asymmetry. We argue that before and after the general election, in a window of approximately half a year, we cannot find important differences in the growth rates due to the expected symmetry. We argue that incumbents who are in power after the general election keep high wage growth rates for a while right after general elections because of the special nature of the minimum salary discussed above (see Schemmel and Picot 2025). Following Feierherd et al. (2024), we estimate Equation (1) using ordinary least squares (OLS) with Huber–White robust standard errors (of the HC2 type).

Therefore, the main hypothesis to be tested from (1) in this paper is the following:

H₁: 

For primary elections, we expect ${\mathit{\beta }}_1$  = 0 if the varying temporal window is for only the first few days that we expect growth rates to increase before primary elections and ${\mathit{\beta }}_1$  < 0 when the temporal window is larger. For general elections, we expect ${\mathit{\beta }}_1$  = 0.

We set our temporal window on one-day intervals to be 150 days, since we expect the impact of the change in the growth rate before/after the election to be more persistent than the effect analyzed in Feierherd et al. (2024), where a range of thirty-seven was selected when analyzing the file of complaints. The reason, we argue, is that incumbents who lost, or even those who won, keep high wage growth rates for a while right after general elections, as discussed above (see Schemmel and Picot 2025). We also check that our results are robust to analyzing windows of 180 days before and after the election, corresponding to a complete year. Note that this methodology, which allows for varying temporal windows, is especially useful in our case, since we can investigate whether, in primary elections and across separate regressions for varying windows, ${\mathit{\beta }}_1$ = 0 for the first days that we expect growth rates to increase before primary elections and ${\mathit{\beta }}_1$ < 0 after that.

2.2.2. Robustness Checks: Local-Linear Regression

Following Feierherd et al. (2024), we also estimate a local-linear regression using the number of days away from the nearest election to compute triangular weights. For a given bandwidth, we fitted two separate regressions for the days before and after the nearest election, and estimated the difference between the two intercepts. Bias-corrected confidence intervals and p-values proposed by Calonico et al. (2014) are provided. We use a bandwidth equal to the temporal window used in estimating Equation (1), and therefore a maximum of 150 days is allowed before and after elections. We also check that our results are robust to a bandwidth of 180 days. Note that this methodology allows us only to check that ${\mathit{\beta }}_1$ = 0 for general elections, but does not allow us to check if ${\mathit{\beta }}_1$ = 0 or ${\mathit{\beta }}_1$ < 0 for primary elections, depending on the temporal window.

3. Results

3.1. Baseline Model

Figure 3 shows the results of estimating Equation (1) with nominal growth rates when focusing on primary (panel (a)), general (panel (b)) or both types of elections (panel (c)). As expected, since we argue that incumbents significantly increase growth rates of minimum wages a few days before primary elections, we find that, at the beginning of the cycle, growth rates of minimum wages are not significantly different before and after primary elections (during the few days of the initial increase before primary elections); however, after those few days have passed, we find that growth rates are statistically significantly higher after primary elections than before (see panel (a)). Panel (b) shows that growth rates before and after general elections are not statistically different. The effect when employing both types of elections in panel (c) covers the two types of effects of panels (a) and (b). Figure 4 shows that the results are robust when using real growth rates.

Figure 3. Difference in the number of days of nominal growth rates before and after the nearest election. Point estimates come from separate regressions for varying windows as given in (1). Bars represent robust 95 per cent confidence intervals. Panel (a) corresponds to primary elections, while panels (b,c) focus on general and both types of elections. Source: Own elaboration.

Figure 4. Difference in the number of days of real growth rates before and after the nearest election. Point estimates come from separate regressions for varying windows as given in (1). Bars represent robust 95 per cent confidence intervals. Panel (a) corresponds to primary elections, while panels (b,c) focus on general and both types of elections. Source: Own elaboration.

3.2. Robustness Checks: Local-Linear Regression

Table 3, Table 4, Table 5 and Table 6 show the local-linear regression results. By using nominal and real growth rates, along with robust confidence intervals (RCI) following the procedure of Calonico et al. (2014), Table 3 and Table 5 confirm the expected results of Figure 3 and Figure 4 that ${\mathsf{\beta }}_1$ = 0 for general elections. For primary elections, Table 3 also shows evidence in favor that ${\mathit{\beta }}_1$ < 0, although this methodology does not allow us to find statistically significant evidence that growth rates increased a few days before primary elections. Table 5 shows that ${\mathit{\beta }}_1$ = 0 for primary elections, but it cannot verify if ${\mathit{\beta }}_1$ < 0 for some particular time varying windows as the baseline model. Table 4 and Table 6 also confirm the results of Table 3 and Table 5 for a bandwidth of 180 days.

Table 3. Local linear estimates for a bandwidth of 150 days. Nominal growth rates. Source: Own elaboration.

Types of Elections	Estimate	95% RCI	p-Value	N Before/N After	Window
Primary	−0.04	(−0.05, −0.01)	0.009	507/452	150
General	0.00	(−0.02, 0.02)	0.780	462/507	150
Primary/General (both)	−0.02	(−0.02, 0)	0.100	604/590	150

Table 4. Local linear estimates for a bandwidth of 180 days. Nominal growth rates. Source: Own elaboration.

Types of Elections	Estimate	95% RCI	p-Value	N Before/N After	Window
Primary	−0.04	(−0.05, −0.01)	0.001	595/540	180
General	0.00	(−0.02, 0.01)	0.858	550/595	180
Primary/General (both)	−0.02	(−0.03, −0.01)	0.004	692/678	180

Table 5. Local linear estimates for a bandwidth of 150 days. Real growth rates. Source: Own elaboration.

Types of Elections	Estimate	95% RCI	p-Value	N Before/N After	Window
Primary	−0.01	(−0.01, 0.01)	0.470	507/452	150
General	−0.01	(−0.01, 0.01)	0.912	462/507	150
Primary/General (both)	−0.01	(−0.02, 0)	0.138	604/590	150

Table 6. Local linear estimates for a bandwidth of 180 days. Real growth rates. Source: Own elaboration.

Types of Elections	Estimate	95% RCI	p-Value	N Before/N After	Window
Primary	−0.02	(−0.02, 0.00)	0.225	595/540	180
General	−0.01	(−0.02, 0.01)	0.560	550/595	180
Primary/General (both)	−0.01	(−0.02, −0.01)	0.001	692/678	180

4. Discussion

There is an extensive and diverse literature showing the effects of minimum wages on income inequality in Latin America (see, e.g., Maurizio and Vázquez 2016; Feierherd et al. 2023; Francis and Valodia 2021; Anria and Bogliaccini 2022; Cámara et al. 2024). Maurizio and Vázquez (2016) argued that increasing minimum wages decreased income inequality in Argentina, Brazil, and Uruguay, but not in Chile. However, Anria and Bogliaccini (2022) showed that positive results also exist in Chile, while Axelsson and Martins (2024) showed the dangers of the lack of wage indexation in Brazil. More recently, it has been shown that repeated minimum wage increases did not lead to increased unemployment in Argentina (Abbate and Jiménez 2026); however, as minimum wages rise, firms increase investments in tangible assets (da Fonseca et al. 2026). In this paper, we show that minimum wages can also be used by incumbent politicians with perverse incentives (as in Jimenez 2023) by documenting an electoral cycle in the evolution of minimum salaries in Argentina. In more detail, we focus on the potential strategic manipulation of a core economic policy around elections by identifying a political business cycle related to changes in minimum wage policy in Argentina between 2013 and 2021. The findings are interesting, especially because Argentina features incumbents from different sides of the ideological spectrum (Feierherd et al. 2024). Therefore, we find that the business cycle persists despite the ideological orientation of the president, and this signals broader scope conditions.

We argue that incumbent politicians tend to significantly increase growth rates of minimum wages a few days before primary elections, and they tend to keep them high for a period after general elections conclude. We argue that incumbents who lost, or even those who won, keep high wage growth rates for a while right after general elections, because minimum salary is regarded as paying necessary respect to those who fulfill their obligation to social cooperation (see Schemmel and Picot 2025). We show that our analysis is robust to the use of nominal and real minimum wages. However, note that Schemmel and Picot (2025) showed that the best justification of minimum salaries does not lie in their economic effects but in the respect for social contributors that they express. Therefore, we also argue that incumbent politicians benefit from the increase in minimum wages a few days before elections, even without needing to wait to feel the economic effects.

These results are in line with previous studies on populism in Latin America (see, e.g., Posner 2016; Montero 2024), and how increasing the minimum wage can improve citizens’ assessment of governance performance (Jimenez 2023). Our results are also in line with previous literature on the strategic use of different tools used by politicians aiming to be re-elected (Eckstein 2009; Álvarez-Mingote 2019; Barrientos 2023; Feierherd et al. 2024). Our results highlight the importance of incumbents and challengers competing on a level playing field during electoral campaigns (Johnson 2011), and that there are mechanisms in place that guarantee that incumbents do not use strategic tools (such as increases in minimum salaries) for their own benefit just before elections. The following are some examples of specific mechanisms that could be put in place: First, having institutionalized tripartite consultation bodies in order to force consultation with other social actors before changing the minimum wage. Second, installing a caretaker (i.e., a temporary ad hoc) government: an interim, non-partisan, and neutral administration to oversee elections, preventing the use of government resources for campaigns. Third, forcing regulations and legal restrictions to prohibit massive minimum salary increases before an election, and laws that mandate transparency in the actions of incumbents during campaigns and the regulation of state resources. Fourth, making use of autonomous independent electoral management bodies acting as a check on executive power, ensuring unbiased administration of elections, including enforcement of campaign finance laws. Fifth, setting rules requiring public media to provide equitable, balanced coverage to both opposition and incumbent parties, since incumbents often control state media, providing them with disproportionate coverage and the ability to marginalize opposition voices, especially during campaigns.

Our methodology has several important limitations. First, we make use of daily data in order to characterize the election dates (primary and general) in Argentina; however, minimum wages in Argentina are not adjusted daily. They are typically changed through discrete policy decisions. Therefore, using daily growth rates may generate long periods of zero changes punctuated by large jumps, and this may create a data structure poorly suited for OLS regressions. Second, we assume that higher growth rates of minimum salaries around elections are equal to electoral manipulation with perverse incentives, and we recommend electoral fairness mechanisms. However, alternative explanations could include pre-scheduled wage adjustments, bargaining cycles with unions, or institutional wage review calendars. Without controlling for these factors, the results may reflect coincidence rather than strategy. Third, our dataset covers only the period 2013–2021 in order to make use of the primary and general election dates given in Feierherd et al. (2024). However, this includes very few election cycles, and it raises concerns about small sample bias and lack of statistical power. The use of an alternative methodology is left for future research.

Finally, it is important to note that many Latin American countries have a statutory national minimum wage that is set unilaterally by the government (see Table 1 in Gindling and Ronconi (2025)). Additionally, 30 out of 38 OECD countries have a statutory minimum wage in place, and minimum wages also exist in most non-OECD emerging economies (OECD 2025). All of this supports the applicability of our results to other countries.