The effects of IMF conditional programs on the unemployment rate

Highlights

• We examine the effect of IMF conditional programs on the unemployment rate.

• IMF program participation increases the unemployment rate of recipient countries.

• Accounting for IMF conditions, a detrimental effect is found only for conditions.

• We provide a disaggregated analysis of IMF conditionality by four policy areas.

Abstract

The fundamental mission of the International Monetary Fund (IMF) is to ensure global financial stability and to assist countries in economic turmoil. Although there is a consensus that IMF-supported programs can have a direct effect on the labor market of recipient countries, it remains unclear how IMF participation decision and conditionalities attached to IMF loans can affect the unemployment rate of borrowing countries. Using a world sample of countries from 1980 to 2014, we investigate how lending conditional programs of the IMF affect the unemployment rate. Our analyses account for the selection bias related to, first, the IMF participation decision and, second, the conditions included within the program. We show that IMF program participation significantly increases the unemployment rate of recipient countries. Once we control for the number of conditions, however, we find that only IMF conditions have a detrimental and highly significant effect on the unemployment rate. There is evidence that the adverse short-run effect of IMF conditions holds robust in the long-run. Disaggregating IMF conditionality by issue area, we find adverse effects on the unemployment rate for four policy areas: labor market deregulation, reforms requiring privatization of state-owned enterprises, external sector reforms stipulating trade and capital account liberalization, and fiscal policy reforms that restrain government expenditure. Our initial results are found to be robust across alternative empirical specifications.

Introduction

The International Monetary Fund (IMF) has become an almost universal financial institution, with its membership rising from 44 states in 1946 to 190 at present. The basic conception of the IMF's role is to uphold global financial stability, and it is often considered as the “crisis manager” for developing and developed countries. With its loan programs the IMF promoted policy reforms, such as the privatization of state-owned enterprises, trade and financial liberalization, economic deregulation, social policy, labor market reforms, and ‘good governance’, able to restructure the domestic economy of borrowing countries (Kentikelenis et al., 2016). Previous studies highlight that IMF loan programs can have multi-level effects on borrowing countries, including their growth rates (Barro and Lee, 2005; Dreher, 2006), poverty and income inequality (Oberdabernig, 2013; Forster et al., 2019; Lang, 2021; Stubbs et al., 2021; Chletsos and Sintos, 2022), labor rights (Lloyd and Weissman, 2002; Reinsberg et al., 2019), tax revenue (Crivelli and Gupta, 2016; Reinsberg et al., 2020), public spending (Rickard and Caraway, 2019) and health outcomes (Forster et al., 2020).

A strand of literature highlights the impact of international financial institutions (IFIs) on labor (Blanton et al., 2015; Martin and Brady, 2007; Nooruddin and Vreeland, 2010; Vreeland, 2002). Previous work points out that these organizations have notoriously sought to increase the “flexibility” of labor markets – that is, increasing employers' ease in hiring, firing, or setting working hours (Burgess, 2010, p.202) – in borrowing countries, with the promise of future economic growth. Furthermore, previous studies examine the link between IMF programs and labor rights. While Abouharb and Cingranelli (2007) and Blanton et al. (2015) conclude that IMF programs significantly reduce both individual and collective labor rights, Reinsberg et al. (2019) demonstrate that collective labor rights increase in the wake of IMF programs. In this context, studies also examine the effects of IMF program on the public sector. Research results on the effects of IMF loans on public spending are mixed and depend on the sample used and estimated procedure (e.g. Bulír and Moon, 2006; Clements et al., 2013; Kentikelenis et al., 2015; Nooruddin and Simmons, 2006; Nooruddin and Vreeland, 2010). More specific results are reported in Rickard and Caraway (2019), who find that the implementation of reforms related to the public sector to a country's IMF program reduces government spending on the public sector wage bill. Li et al. (2015) point out that IMF policy reforms often lead to an inactive business investment, poorer government service, severe social instability, and a higher unemployment rate – all of which may damage the economic development of bailed-out countries in the long term. The IMF also promotes policies to reduce external and domestic arrears by imposing restrictions on available sources of credit (Kentikelenis et al., 2016). These monetary and debt measures typically lead to higher interest rates, increasing the cost of borrowing and making it more expensive for businesses to expand (Beck et al., 2008; Chletsos and Sintos, 2021a; Marques et al., 2017), which in turn may affect the creation and destruction of private sector jobs. Moreover, previous research documents that IMF conditionality has adverse effects on a range of labor related indicators, including employment levels, wages and employment protection policies (Lloyd and Weissman, 2002; Blanton et al., 2015). Pastor (1987) and Vreeland (2002) empirically investigate the impact of IMF programs on the labor share of income and show that IMF programs are linked to a decline in labor's share of income. In addition, Ohanyan and Androniceanu (2017) show that IMF participation decreases the level of employment using a combination of propensity score matching with the differences-in-differences estimator for the EU-28 between 1993 and 2013. Arguably, the IMF through its policy reforms can alter the underlying structure of an economy, including the labor market (Reinsberg et al., 2019; Chletsos and Sintos, 2021b). Therefore, we focus our attention on labor market issues, and specifically on the level of unemployment.

Should we care about unemployment? During the last century, unemployment has been considered one of the most important factors behind the rise in poverty (e.g. Corcoran and Hill, 1980; Osinubi, 2005; Saunders, 1990). Furthermore, the level of unemployment is linked to various health outcomes. An increased level of unemployment is associated with several mental health issues, such as mixed symptoms of distress, depression, anxiety, psychosomatic symptoms, subjective well-being, and self-esteem (e.g. Backhans and Hemmingsson, 2012; Björklund and Eriksson, 1998; McKee-Ryan et al., 2005; Paul and Moser, 2009). There is also evidence that unemployment increases mortality rates and suicidal behavior (e.g. Córdoba-Doña et al., 2014; Gerdtham and Johannesson, 2003; Lundin et al., 2010; Platt, 1984). Moreover, previous research points out that unemployment increases alcohol consumption and drug use (Henkel, 2011). In addition, several studies show that there is a causal positive relationship between unemployment and both property and violent crime (Lee and Holoviak, 2006; Lin, 2008; Papps and Winkelmann, 2000; Raphael and Winter-Ebmer, 2001). All of these suggest that unemployment is damaging for those who experience it.

In this paper, we explore the influence of IMF programs on the unemployment rate. In exchange for low-cost financing, the IMF requires governments to implement a set of reforms, known as conditionalities (Dreher, 2009). IMF's structural adjustment involves a series of reforms in sensitive policy areas, such as privatization of state-owned enterprises, trade and financial liberalization, economic deregulation, social policy, labor market reforms, and ‘good governance’ (Kentikelenis et al., 2016). These required structural adjustments designed by the IMF to its borrowers can affect the labor market and in turn the unemployment rate of recipient countries. Our study follows the recent strand of the literature (e.g. Forster et al., 2019; Kern et al., 2019; Reinsberg et al., 2021; Stubbs et al., 2020), by investigating both the effect of IMF program participation and conditionality. Most studies treat IMF programs as homogenous (e.g. they include only a binary indicator to capture IMF program participation) and are therefore unable to distinguish specific pathways via which IMF programs affect the outcome of interest. Moreover, to account for the endogeneity bias related to the IMF participation decision and the conditions attached to IMF loans, we follow a recently developed econometric technique (Stubbs et al., 2020), by estimating a system of three equations including instrumental variables via maximum likelihood and allowing for correlated errors across equations (Roodman, 2011).

To account for the aforementioned challenges, this study meticulously assesses the impact of IMF loan programs (conditional on their mandated policy reforms) on the unemployment rate using annually compiled panel data for a world sample of countries between 1980 and 2014. Our baseline findings show that IMF program participation significantly increases the unemployment rate of recipient countries. Once we control for the number of conditions, however, we find that only IMF conditions have a detrimental and highly significant effect on the unemployment rate. In addition, we show that IMF conditions have detrimental long-term implications for the unemployment rate. A more detailed analysis identifies adverse effects on the unemployment rate for four policy areas: labor market deregulation, reforms requiring privatization of state-owned enterprises, external sector reforms stipulating trade and capital account liberalization, and fiscal policy reforms that restrain government expenditure. Our initial results are robust across various specifications.

The rest of the paper proceeds as follows. Section 2 describes the potential pathways through which IMF programs affect the unemployment rate. In Section 3, we discuss the relationship between IMF programs, labor market institutions, and unemployment. Section 4 analyzes our data and the empirical methodology used. Section 5 reports the results. Finally, Section 6 concludes.