The extant literature suggests a significant association between oil prices and unemployment. However, the relevant literature is unclear on how oil prices impact unemployment in oil-importing economies. In this study, we empirically examine the impact of oil prices on unemployment in 29 African oil-importing economies employing the linear and nonlinear panel autoregressive distributed lag (ARDL) techniques. Our findings demonstrate a negative and significant relationship between oil prices and unemployment in the short run. In contrast, we observe a direct but weak association between unemployment and oil prices in the long run. Besides, unemployment reacts negatively and positively, respectively, to positive and negative changes in oil prices in the short run. In the long run, we detect that an increase in oil prices aggravates unemployment significantly, but a fall in oil prices improves unemployment significantly. All our findings prove robust to data of different frequencies (quarterly and yearly) and Brent and West Texas Intermediate (WTI) oil price indices.

Keywords:

1. Introduction

Unemployment is a problem confronting economies around the world as a result of its attendant social, political, and economic consequences. These consequences are much more prevalent in developing countries where productivity is low, macroeconomic policies are weak, and the tax base is low. Besides, unemployment benefits in these countries are either absent or minuscule and reach only an insignificant proportion of the unemployed. The social consequences of unemployment are diverse. These include low life expectancy, a high rate of divorce, deteriorating health conditions, widening income inequality, and high crime rates, among others. Its economic consequences include lower output, revenue losses for the government, higher costs of unemployment benefits, a loss of human capital, low investment, and high poverty. Decreased political participation and engagement (Saint-Paul, 1999) and civil unrest are among its political consequences. Hence, the goal of reducing unemployment, since it is impossible to eradicate it, to a level around which their economy gravitates in the long run (the steady state of unemployment) is one of the macroeconomic objectives that every country pursues. Mankiw (2010) and Chan and Dong (2022) noted that achieving unemployment rate stability is among the policymakers’ primary objectives. However, for every country or group of countries, achieving this macroeconomic objective requires a good and accurate understanding of the factors that cause unemployment. Also, accurate insights on the empirical relationship between unemployment and those factors that cause it are necessary for addressing unemployment effectively.

Several factors cause unemployment. These include workers’ remuneration (wages and salaries), marginal productivity of workers or labor, prices of other factors of production that complement energy, and productivity of other factors of production. Besides, local factors such as the level of technology, population demographics, the state of an economy, the business cycle, and global factors such as energy prices cause unemployment (Dogrul and Soytas, 2010). Rising energy prices, for example, increase the cost of production, thereby decreasing aggregate supply and, by extension, the aggregate output of goods and services (Uri, 1996; Nusair, 2020). The relevant literature documents that an increase in the price of oil leads to an economic downturn, particularly in the small oil-importing countries (Ahmed et al., 2023a). This potentially results in rising unemployment in those countries. An increase in the oil price translates into goods and services output fall, triggering a contraction in labor demand, which inevitably aggravates the unemployment situation in an economy.

Oil is indispensable in goods and service production, but its price is inherently unstable. This instability could potentially impact unemployment since, in many instances, it is unpredictable. Evidence from Chan and Dong (2022) indicates that unpredicted oil price instability translates into a persistent rise in the unemployment rate. Thus, it is imperative to evaluate empirically the effect of oil prices on unemployment. The literature indicates that studies have focused on the relationship between oil prices and unemployment. Some of these studies cover groups of countries (for example, Ran and Voon, 2012; Katircioglu et al., 2015; Cuestas and Gil-Alana, 2017; Cheratian et al., 2019; Nusair, 2020). While others, for example, Uri (1996), Caporale and Gil-Alana (2002), Loschel and Ulrich (2009), Dogrul and Soytas (2010), Senzangakhona and Choga (2015), Bocklet and Baek (2017), Karlsson et al. (2018), and Kocaarslan et al. (2020), cover individual countries. Different findings have emerged from these studies. For instance, Caporale and Gil-Alana (2002), Ran and Voon (2012), and Senzangakhona and Choga (2015) establish a positive relationship between oil prices and unemployment. But Katircioglu et al. (2015) detect a negative association between the variables. Ahmed (2013) and Karlsson et al. (2018) report causality from oil prices to unemployment. Another study by Cuestas and Gil-Alana (2017) establish that an increase in oil prices spurs unemployment, while a decrease in those prices reduces unemployment. Similarly, Caruth et al. (1998) validate that at equilibrium, while a rise in oil prices increases the unemployment rate, a fall in the prices reduces the unemployment rate. Bocklet and Baek (2017); Kocaaslan (2019); Kocaarslan et al. (2020), and Nusair (2020) demonstrate evidence of asymmetries in the relationship between oil prices and unemployment. This inconsistent effect of oil prices on unemployment necessitates further evaluation of the association between the variables.

Oil prices’ influence on unemployment could differ among countries as a result of several factors. One potential reason is whether a country is an oil-exporting or importing country. Also, the relationship between these variables may vary depending on whether the oil-importing or exporting country is developing or not. Furthermore, the association between oil price and unemployment may differ across regions and groups of countries due to many factors. These factors are level of development, macroeconomic policies, natural resource(s) availability, level of technology, etc. For example, Ahmed et al. (2023b) suggest that the effects of oil prices on unemployment are a function of the state of the economy. Thus, assessing the impact of the oil prices on unemployment for these groups of countries — oil-importing and exporting countries (developed or developing), for a particular region, and group of countries is highly essential. This will potentially benefit these countries in terms of providing them with invaluable insights for policymaking focused on addressing underemployment. Against this backdrop, this study examines the relationship between oil prices and unemployment in African oil-importing economies. It answers the following question. How does unemployment react to positive and negative oil price changes in oil-importing economies in Africa in the short and long run?

Oil-importing countries in Africa are our focus since unemployment has been high in most of these countries recently. For instance, in Botswana, Cabo Verde, Djibouti, Eritrea, Mauritius, Namibia, and Rwanda, the unemployment rates in 2019 and 2020 were, respectively, 20.09% and 21.02%, 12.21% and 15.68%, 26.32% and 28.05%, 7.28% and 6.80%, 6.33% and 8.63%, 20.00% and 21.24%, and 12.43% and 13.01% (World Bank, 2022). Besides, the wealth effect theory postulates that an increase in the price of oil potentially transfers income from oil-importing countries to their counterparts that export. This potentially reduces the purchasing power in the former group of countries, ceteris paribus. The potential consequence is a decline in aggregate demand for goods and services in the oil-importing countries, exacerbating the unemployment situation in these countries. An understanding of the impact of oil prices on unemployment in African oil-importing countries based on empirical evidence will, therefore, provide a useful guide to the governments of these countries and policymakers in those countries. This understanding will enable them to formulate and implement policies that are potentially effective in reducing unemployment in their countries.

The structure of the remainder of our paper is as follows: Section 2 presents the relevant literature. Sections 3 and 4 provide our methodology; results, and discussion, respectively. Section 5 presents our summary and recommendations.

2. Relevant Literature

While literature abounds on the association between oil prices and unemployment, a consensus has yet to be reached on the nature of the relationship between these variables. For example, Ahmed (2013) investigates the relationship between oil prices and unemployment in Pakistan using monthly data from 1991:01 to 2010:12. Results from Toda and Yamamoto causality tests show that, in the long run, oil prices cause unemployment. Similarly, Karlsson et al. (2018) assessed the causal association between oil prices and unemployment in Norway using monthly data spanning 1997–2015. The findings of the wavelet multi-resolution analysis (MRA), causality tests, and impulse response functions show the existence of causality between oil prices and unemployment. Focusing on Dogrul and Soytas, (2010) investigates the causality between unemployment and input prices (crude oil and the real interest rate). Evidence from the Toda Yamamoto causality test indicates that during 2005:01–2009:08 oil prices improved unemployment forecasts in the long run.

Adeosun et al. (2023) assess the nexus between return on oil price and unemployment in Japan, Italy, Canada, and France from 2000:12 to 2022:2. They found that oil prices predict unemployment. Cuestas and Gil-Alana (2017) examine the effect of oil price movements on unemployment in Central and Eastern Europe using quarterly data from 2000:Q1 to 2015:Q4. Results from the autoregressive distributed lag (ARDL) model revealed that an increase in oil prices increases the natural unemployment rate. Besides, a decrease in oil prices decreases the natural rate of unemployment. Michieka and Gearhart (2019) examine the long- and short-run relationship between oil prices and employment in four sectors of the top oil-producing countries in the US. The results from the panel ARDL model and monthly data from 1990:01–2017:09 show causality from oil prices to employment and the absence of short- and long-run causality from oil prices to service employment.

In Canada and the US, Nusair (2020) examines the effect of oil price changes on unemployment rates using data spanning 1960:01–2018:04. The findings from the nonlinear ARDL model indicate that changes in oil prices have a minor impact on unemployment in the short run. In the long run, oil price changes have a significant positive influence on unemployment. Besides, there is evidence of short- and long-term asymmetries in unemployment’s reaction to positive and negative changes in oil prices. Furthermore, a rise in oil prices is associated with increased unemployment rates, while falling oil prices lower unemployment rates. In a similar study, Cheratian et al. (2019) analyzed the role of price movements on unemployment rates in oil-importing and exporting countries in the MENA region from 1991 to 2017. The finding from the nonlinear ARDL model suggests that positive changes in oil prices raise unemployment in the long run. In contrast, negative changes have an insignificant impact on the unemployment rate in both the short and long run.

Bocklet and Baek (2017) test the symmetric and asymmetric effects of oil price changes on unemployment in Alaska from 1987:Q3 to 2014:Q4. Observations from nonlinear ARDL reveal evidence of a short-run asymmetric relationship between oil price and unemployment. In addition, increases in oil prices have more influence on unemployment than decreases. Kocaarslan et al. (2020) examine the presence of asymmetric interactions among oil prices, oil price uncertainty, interest rates, and unemployment in the US from 2007:05 to 2019:04. Evidence from nonlinear ARDL indicates an asymmetric response of unemployment to changes in oil prices. Furthermore, an increase in oil prices significantly raises unemployment, while a decrease in oil prices insignificantly reduces unemployment.

In South Africa, Senzangakhona and Choga (2015) investigate the impact of crude oil volatility on unemployment using quarterly data from 1990 to 2010. The finding from the vector autoregression (VAR) technique reveals that a negative association exists between oil price and unemployment in the short run. However, a positive relationship exists between the variables in the long run. Using the same model and data frequency covering 2000:Q1–2015:Q4, Trang et al. (2017) analyze the influences of oil prices on macroeconomic variables in Vietnam. The result shows that a positive oil price shock has a positive but insignificant impact on unemployment in the short run. Similarly, Loschel and Ulrich (2009) analyzed the impact of oil prices on unemployment in Germany from 1973:10 to 2008:01. The results from the VAR model show that oil price increases induce a rise in unemployment. Cuestas and Ordonez (2018) analyze the role of oil price movements in the UK’s unemployment evolution from 2000:Q1 to 2014:Q4. The Bayesian Structural Vector Autoregressive (SVAR) technique result reveals that negative oil price innovations contribute positively to preventing further increases in unemployment after the beginning of the 2008 financial crisis. Ahmed et al. (2023b) assessed low and high oil price uncertainty on the US’s unemployment rate from 1973 to 2018. The result from the logistic smooth transition autoregressive process and the exponential generalized autoregressive heteroskedastic models demonstrates that uncertainty in oil prices has no significant impact on unemployment in low oil price uncertainty states. However, oil price uncertainty raises the unemployment rate in a state of high price uncertainty.

Ordonez et al. (2019) analyze the effects of oil price shocks on unemployment from 2000:Q1 to 2014:Q4. The findings from the SVAR model indicate that positive oil price shocks contributed negatively to unemployment during the post-2007–2009 financial crises. However, a decrease in oil prices had less influence on increasing unemployment. Ran and Voon (2012) examine whether oil price shocks significantly affect unemployment in Hong Kong, Singapore, South Korea, and Taiwan economies during 1984: Q1-2007: Q3. Results from the VAR and Vector Error Correction Model (VECM) techniques demonstrate that oil prices have significant positive impacts on unemployment after three-time lags. In evaluating the impact of real oil price shocks on labor market flows in the US, Ordonez et al. (2010) employ the Smooth Transition Regression (STR) approach and quarterly data that cover 1957:Q1–2003:Q3 and establish that oil price shocks are important forces of job market flows.

In the US, Kocaaslan (2019) searches for the effects of oil price uncertainty and oil price shocks on unemployment using data from 1974:Q2 to 2017:Q4. The result from the GARCH-in-mean VAR method suggests that oil price uncertainty increases the unemployment rate. Also, a positive oil price shock increases unemployment, while a negative oil price shock raises unemployment. Furthermore, unemployment asymmetrically responds to positive and negative oil price shocks. In the US, Uri (1996) assessed whether fluctuations in crude oil prices have affected employment and the unemployment rate by utilizing annual data from 1890 to 1994. Evidence from the Ordinary Least Squares (OLS) approach suggests that a secular trend in unemployment explains changes in oil prices. Caporale and Gil-Alana (2002) evaluated the relationship between unemployment, real oil prices, and interest rates in Canada using data covering 1966:Q1–2000:Q2. The findings from the OLS reveal that oil prices have a significant positive impact on unemployment.

Having searched the earlier literature extensively, to the best of our knowledge, the relationship between oil price and unemployment for oil-exporting economies in Africa had yet to receive attention. Accordingly, we examine the impact of oil prices on unemployment in oil-importing economies in Africa. We therefore add to the earlier literature in the following ways: First, we investigate how oil prices impact short- and long-run unemployment in oil-importing economies in Africa. That is, we investigate the short- and long-run linear (symmetric) response of unemployment to oil prices in oil-importing economies in Africa. Second, we account for positive and negative oil price changes in the short- and long-term relationship between oil prices and unemployment in this group of economies. In particular, in addition to assessing the short- and long-run linear relationship between oil prices and unemployment, we evaluate the short- and long-run nonlinear (asymmetric) reaction of unemployment to increases and decreases in oil prices in oil-importing economies in Africa. To our knowledge, we are the first to capture both the short- and long-term symmetric and asymmetric influence of oil prices on unemployment at the regional level. A proper understanding of the empirical linear (symmetric) relationship between oil price and unemployment, as well as how unemployment reacts to increases and falls in the price of oil in oil-importing economies in Africa, provides a guide for both governments and policymakers in these countries in their efforts to fight unemployment and its attendant consequences.

2.1. Theoretical framework

The literature documents that oil prices affect economic activities and, by extension, influence unemployment (see, for example, Hamilton, 1983; Burbidge and Harrison, 1984; Hamilton, 1988; Caruth et al., 1998). Brown and Yucel (2002) provide different channels through which oil prices and unemployment are related. These channels are the supply-side shock, the transfers of income and aggregate demand, the real balance effect, and the role of monetary policy. In this study, however, we present the supply-side shocks and transfers of income channels described by Brown and Yucel (2002). In addition to these channels, we present the impact of the oil price on labor market dynamics.

Brown and Yucel (2002) suggest that supply-side shocks, for example, increases in oil prices, are energy (an essential input in goods and services production) shortage indications. Ceteris paribus, the scarcity of this key input (oil) lowers both productivity and output, reduces employers’ real wages, and decreases demand for labor, thus raising the rate of unemployment in a country. Similarly, an increase in production costs resulting from an increase in oil prices reduces productivity growth, thereby reducing real wage growth and translating into an increase in unemployment.

The transfer of income channel suggests that an increase in the price of oil shifts purchasing power from countries that import to countries that export oil (Fried and Schultze, 1975; Dohner, 1981; Brown and Yucel, 2002). A rise in the price of oil increases the amount of money oil-importing countries spend on oil, dwindling their income and ability to buy goods and services. In contrast, the case is different in the oil-importing countries, as a rise in the price of oil translates into an increase in revenue from oil sales, strengthening their purchasing power. All else being constant, the purchasing power transfer from countries that import to their counterparts that export oil reduces the aggregate demand for goods and services in the former group of countries. But the converse is the case in the latter group of countries. This fall in demand for goods and services in the oil-importing countries potentially worsens the unemployment situation in those countries. Mankiw (2010) and Mian and Sufi (2012) suggest that a decline in aggregate demand translates into rising unemployment, since this forces employers to minimize losses by cutting down on production, necessitating reducing their workforce. Mian and Sufi (2012) found evidence indicating that between 2007 and 2009, the fall in aggregate demand accounted for the largest part of unemployment in the US.

Oil prices also affect unemployment via their impact on labor market dynamics. An increase in the price of oil affects the relative cost of production in many industries and changes the production structure. Especially if the increase in the oil price is persistent, hence strengthening unemployment (see Loungani, 1986). If there is an increase in the price of oil, the marginal cost of production for industries or firms that are oil-intensive rises, forcing those industries or firms to adopt other production methods that are less or not oil-intensive. All other things being constant, this will potentially have an effect on unemployment in the long run since there will be a rapid reallocation of both labor and capital across activities or sectors. Workers will have to move from one sector or activity to another, and because workers have acquired skills and training that are specific to certain activities or industries, it will take a long time before they are absorbed. This aggravates the unemployment situation in a country.

3. Methodology

3.1. Data and data description

Our data consists of quarterly data from 1991Q to 2019Q4 for 29 oil-importing economies in Africa. We also use annual data for the same period to strengthen the credibility of our findings. Our variables are as follows: (1) unemployment as a percentage of the total labor force unemployed (unemployment rate) based on an International Labour Organisation (ILO) estimate. (2) Brent oil price index. (3) West Texas Intermediate (WTI) oil price index. The WTI and Brent prices indices measurement units are US dollars per barrel. We use the Brent oil price index for our main estimation, while the WTI price index is for reliability testing. Our data source for data on unemployment is the 2022 World Bank’s World Development Indicators (WDI) database.¹ We sourced data on oil prices from the United States Energy Information Administration (EIA).² Countries that we consider for the study are as follows: Benin, Botswana, Burkina Faso, Burundi, Cabo Verde, the Central African Republic, Djibouti, Eritrea, Eswatini, and Ethiopia. Others are Kenya, Lesotho, Madagascar, Malawi, Mali, Mauritius, Mozambique, Namibia, Rwanda, Senegal, Seychelles, Sierra Leone, Somalia, Tanzania, The Gambia, Togo, Uganda, Zambia, and Zimbabwe. These countries depend entirely on imported oil since they are not oil-producing economies. Consistent with the wealth effect theory, the economies are potentially more affected by increases than decreases in oil prices.

3.2. Model specification

To achieve the objectives of this study, we employ the symmetric ARDL model of Pesaran and Shin (1999) and Pesaran et al. (2001), as well as the asymmetric ARDL model of Shin et al. (2014). However, because our study is a panel study, we present the panel version of both Pesaran and Shin (1999) and Pesaran et al. (2001) and Shin et al. (2014) models. Several reasons justify the choice of our model. Results produced from panel data techniques are better when compared to those made from cross-sectional and time series data techniques (Munir and Iftikhar, 2021). One reason for this is that the panel data technique accounts for heterogeneity and cross-specific effects, making the estimates produced robust (Baltagi, 2013; Munir and Iftikhar, 2021). Besides, the panel symmetric and asymmetric ARDL procedures enable short- and long-run symmetric and asymmetric relationship modeling between variables in panel data analysis. The nonlinear approach is flexible, allowing the use of variables that are I(1), I(0), or a combination of I(0) and I(1) (Shin et al., 2014), without compromising results quality. Furthermore, the panel nonlinear ARDL model accounts for the effects of heterogeneity in panel data (Munir and Riaz, 2019). These benefits potentially make the insights presented in this study reliable. Equation (1) is our panel ARDL model for the linear relationship between unemployment and oil prices.

where

$u_{i, t}$ ,

$p_{t}$ ,

$μ_{t}$ , and

$υ_{i t}$ represent unemployment rate for country i at time t, oil price at time t, the group specific effect at time t, and the error correction term, respectively. Similarly, i, and t, respectively, stand for units of sample and time periods we considered for this study. The assumption about the error correction term (

$υ_{i t}$ ) is that it is independently distributed across i and t. Note that in the long run

$Δ u_{i, t - r}$ and

$Δ p_{t - s}$ in Eq. (1) do not exist. Hence, our long run coefficient of oil price for every cross section for the symmetric ARDL model is

$- \frac{ρ_{3 i}}{ρ_{2 i}}$ . In the equation,

$ψ_{i s}$ denotes the coefficient of oil price for the model. Our error correction term model for the linear panel ARDL model is given by the following equation :

where

$u_{i, t} - λ_{1 i} - λ_{2 i} p_{t - 1} = ς_{i t - t}$ is the error correction term, while

$ϕ_{i}$ represents the coefficient of the error correction term for every unit. For clarification,

$λ_{1 i} = \frac{ρ_{1 i}}{ρ_{2 i}}$ and

$λ_{2 i} = \frac{ρ_{3 i}}{ρ_{2 i}}$ .

Having stated the symmetric ARDL model, we now turn to the model that accounts for the reaction of unemployment to the positive and negative changes in oil prices. As stated earlier, our asymmetric ARDL model is a panel version of Shin et al. (2014). Thus, we decompose our decision variable (oil price) into positive and negative exchanges as follows :

p t = p 0 + p + t + p - t, <math display="block" altimg="eq-00016.gif"><msub><mrow><mi>p</mi></mrow><mrow><mi>t</mi></mrow></msub><mo>=</mo><msub><mrow><mi>p</mi></mrow><mrow><mn>0</mn></mrow></msub><mo>+</mo><msubsup><mrow><mi>p</mi></mrow><mrow><mi>t</mi></mrow><mrow><mo>+</mo></mrow></msubsup><mo>+</mo><msubsup><mrow><mi>p</mi></mrow><mrow><mi>t</mi></mrow><mrow><mo>-</mo></mrow></msubsup><mo>,</mo></math> (3)

where

$p_{t}^{+}$ and

$p_{t}^{-}$ are, respectively, positive and negative partial sum processes of positive and negative changes in oil price. Hence, we construct our nonlinear (asymmetric) panel ARDL model as follows :

u i t = ρ + p + t + ρ - p - t + ε i t . <math display="block" altimg="eq-00019.gif"><msub><mrow><mi>u</mi></mrow><mrow><mi>i</mi><mi>t</mi></mrow></msub><mo>=</mo><msup><mrow><mi>ρ</mi></mrow><mrow><mo>+</mo></mrow></msup><msubsup><mrow><mi>p</mi></mrow><mrow><mi>t</mi></mrow><mrow><mo>+</mo></mrow></msubsup><mo>+</mo><msup><mrow><mi>ρ</mi></mrow><mrow><mo>-</mo></mrow></msup><msubsup><mrow><mi>p</mi></mrow><mrow><mi>t</mi></mrow><mrow><mo>-</mo></mrow></msubsup><mo>+</mo><msub><mrow><mi>ε</mi></mrow><mrow><mi>i</mi><mi>t</mi></mrow></msub><mo>.</mo></math> (4)

Defining the positive and negative partial sum processes, thus

p + t = S \sum s = 1 Δ p + s = S \sum s = 1 max (Δ p s, 0), <math display="block" altimg="eq-00020.gif"><msubsup><mrow><mi>p</mi></mrow><mrow><mi>t</mi></mrow><mrow><mo>+</mo></mrow></msubsup><mo>=</mo><munderover accentunder="true" accent="true"><mrow><mo>\sum</mo></mrow><mrow><mi>s</mi><mo>=</mo><mn>1</mn></mrow><mrow><mi>S</mi></mrow></munderover><mi mathvariant="normal">Δ</mi><msubsup><mrow><mi>p</mi></mrow><mrow><mi>s</mi></mrow><mrow><mo>+</mo></mrow></msubsup><mo>=</mo><munderover accentunder="true" accent="true"><mrow><mo>\sum</mo></mrow><mrow><mi>s</mi><mo>=</mo><mn>1</mn></mrow><mrow><mi>S</mi></mrow></munderover><mo>max</mo><mo stretchy="false">(</mo><mi mathvariant="normal">Δ</mi><msub><mrow><mi>p</mi></mrow><mrow><mi>s</mi></mrow></msub><mo>,</mo><mn>0</mn><mo stretchy="false">)</mo><mo>,</mo></math> (5)

p - t = S \sum s = 1 Δ p - s = S \sum s = 1 min (Δ p s, 0) . <math display="block" altimg="eq-00021.gif"><msubsup><mrow><mi>p</mi></mrow><mrow><mi>t</mi></mrow><mrow><mo>-</mo></mrow></msubsup><mo>=</mo><munderover accentunder="true" accent="true"><mrow><mo>\sum</mo></mrow><mrow><mi>s</mi><mo>=</mo><mn>1</mn></mrow><mrow><mi>S</mi></mrow></munderover><mi mathvariant="normal">Δ</mi><msubsup><mrow><mi>p</mi></mrow><mrow><mi>s</mi></mrow><mrow><mo>-</mo></mrow></msubsup><mo>=</mo><munderover accentunder="true" accent="true"><mrow><mo>\sum</mo></mrow><mrow><mi>s</mi><mo>=</mo><mn>1</mn></mrow><mrow><mi>S</mi></mrow></munderover><mo>min</mo><mo stretchy="false">(</mo><mi mathvariant="normal">Δ</mi><msub><mrow><mi>p</mi></mrow><mrow><mi>s</mi></mrow></msub><mo>,</mo><mn>0</mn><mo stretchy="false">)</mo><mo>.</mo></math> (6)

Equations (5) and (6) represent the positive and negative partial sum processes, respectively. From Eqs. (3)–(6), we state our asymmetric panel ARDL model as follows :

Where

$p_{t}^{+}$ is positive change (rise) in oil price,

$p_{t}^{-}$ is negative change (fall) in oil price,

$ρ_{3 i}^{+}$ and

$ρ_{3 i}^{-}$ are, respectively, the long run coefficients of positive and negative changes in oil price for the asymmetric model. Similarly,

$ψ_{i s}^{+}$ and

$ψ_{i s}^{-}$ , respectively, denote the short run coefficients of positive and negative changes in oil price for our nonlinear model. Our long run coefficients of positive and negative changes in oil price are arrived at thus:

$ρ_{3 i}^{+} = - \frac{ρ_{3 i}^{+}}{ρ_{2 i}}$ and

$ρ_{3 i}^{-} = - \frac{ρ_{3 i}^{-}}{ρ_{2 i}}$ Equation 8 is our error correction model for the nonlinear panel ARDL model.

In Eq. (8), $u_{i, t - 1} - ρ_{1 i} - ρ_{3 i}^{+} p_{t - 1}^{+} - ρ_{3 i}^{-} p_{t - 1}^{-} = μ_{i t - 1}$ . Coefficient of this expression for every unit ( $λ_{i}$ ) is the speed at which the variables converge to their long run equilibrium in the presence of short run disequilibrium or shock(s). The higher the coefficient of this term, the shorter the time taken by the variables to converge to the long-run equilibrium when a shock occurs, and vice versa (Zumba, 2022).

4. Results and Discussion

We inspect some of the important futures of all our series and present the results in Table 1. As the table depicts, the total number of observations is 3364, suggesting that our panel is strongly balanced since all the variables have a complete number of observations. The mean values of all the series are higher than their respective minimum values but smaller than their respective maximum values, suggesting the consistency of all our series. Table 1 also depicts that, on average, the Brent oil price index is greater than the WTI index, while unemployment has the smallest average value. Furthermore, it shows that unemployment has the smallest, while the Brent oil price index has the highest variability.

**Table 1. Summary statistics.**
Variable	Obs.	Mean	Std. dev.	Min	Max
u	3364	7.9330	7.4347	0.2854	37.9760
Brent	3364	50.0861	32.8992	11.09	122.2439
WTI	3364	48.5446	29.1600	12.8967	124.0671

We prioritize the stationary of all our data since it is traditional for series to be exposed to stationarity inspection, particularly in studies that involve macro panels. For this reason, we subject all our series to panel unit root tests. We utilize several sets of unit root tests to achieve this objective. These are Im et al. (2003), Maddala and Wu (1999), Pesaran (2007), Levin et al. (2002), Harris and Tzavalis (1999), Breitung (2001), and Hadri (2000). Im et al. (2003), Maddala and Wu (1999) examine the null hypothesis of a unit root with an individual unit process. Pesaran (2007) investigates the null hypothesis of a unit root in the presence of cross-sectional dependence. Harris and Tzavalis (1999), Levin et al. (2002), and Breitung (2001) evaluate the null hypothesis of no unit root with a common unit root process. Table 2 shows the results of our unit root test.

**Table 2. Unit root test result.**
Test	Brent	WTI	u
Null hypothesis: unit root with individual unit root process
Im, Pesaran, and Shin W-stat.	$-$ 32.6985*** $^{, b}$	$-$ 1.4606* $^{, a}$	$-$ 16.2659*** $^{, a}$
ADF Fisher Chi-square	503.7333*** $^{, b}$	671.6448*** $^{, b}$	101.6372*** $^{, a}$
Null hypothesis: unit root in the presence of cross-sectional dependence
Pesaran CD test²	26.038 ( $-$ 6.379*** $^{, b}$ )	26.038 ( $-$ 4.841*** $^{, b}$ )	$-$ 3.628*** $^{, a}$
Null hypothesis: unit root with common process
Levin, Lin, and Chu t*	$-$ 4.8456*** $^{, a}$	$-$ 6.4671*** $^{, a}$	$-$ 8.2754*** $^{, b}$
Harris–Tzavalis rho	$-$ 2.4885*** $^{, a}$	$-$ 4.6194*** $^{, a}$	$-$ 69.4086*** $^{, b}$
Breitung t-stat.	$-$ 4.5969*** $^{, a}$	$-$ 5.5507*** $^{, a}$	$-$ 19.4144*** $^{, b}$
Null hypothesis: no unit root with common unit root process
Hadri Z-stat.	$-$ 3.1044^b	$- 3.559 6^{b}$	$-$ 2.6455^a
No. of cross-sections	29	29	29
No. of periods	115	115	115
No. of observations	3190	3,190	3,190

Notes: * and *** denote statistically significant at 10% and 1%, respectively, while ^a and ^b, respectively, denote stationary at level and at first difference.

The result in Table 2 shows that for all our variables, we find a mixture of different integration orders — order zero (I[0]) and order one ([1]). This result is consistent with panel nonlinear ARDL. It is instructive, however, to state that notwithstanding the order of integration of our series, Table 2 demonstrates that all the series are stationary. Based on our unit root test result, we find it necessary that, in examining how unemployment reacts to changes in oil prices, we account for non-stationarity in our series as well as possible heterogeneity that is intrinsic in the panel data series. Thus, the panel ARDL technique is suitable for this study since it accounts for both the non-stationarity and heterogeneity of series in the panel data framework. We utilize the symmetric panel ARDL model to assess the impact of oil prices on unemployment in oil-importing economies in Africa. However, to account for asymmetries in the relationship between oil prices and unemployment in this group of economies, we also utilize the asymmetric panel ARDL model.

We employ the pooled mean group (PMG) and mean group (MG) to estimate both the symmetric and asymmetric panel ARDL models. Afterwards, we expose the results to the Hausman test to decide the best (efficient) estimator of the two. In deciding the best estimator between the MG and the PMG using the Hausman test, if we reject the null hypothesis of the Hausman statistic, we report estimates of the MG since it is the efficient estimator. By implication, if we accept the test’s null hypothesis, we present the estimates produced by the PMG since it is the efficient estimator. We report the efficient results obtained from the symmetric and asymmetric panel ARDL techniques in Table 3. It is important to note that for the symmetric estimates, the Hausman test confirms that the MG is the efficient estimator, whereas for the asymmetric estimates, the PMG is the efficient estimator.

**Table 3. Short- and long-run linear and nonlinear ARDL models estimates — main result.**
Linear model estimates		Nonlinear model estimates
Variable	MG	Variable	PMG
Constant	0.353** ( $-$ 0.153)	Constant	0.115*** ( $-$ 0.0386)
$Δ p_{t}$	$-$ 0.0011*** ( $-$ 0.0004)	$Δ p_{t}^{+}$	$-$ 0.0011*** ( $-$ 0.0004)
		$Δ p_{t}^{-}$	0.0025*** ( $-$ 0.0007)
$p_{t}$	0.0095 ( $-$ 0.0151)	$p_{t}^{+}$	0.0873* ( $-$ 0.0467)
		$p_{t}^{-}$	$-$ 0.411*** ( $-$ 0.125)
${\hat{ς}}_{i, t - 1}$	$-$ 0.0232*** ( $-$ 0.0061)	${\hat{μ}}_{i, t - 1}$	$-$ 0.0086*** ( $-$ 0.0022)
Hausman test $X_{k}^{2}$	159.91 [0.0000]	Hausman test $X_{k}^{2}$	2.07 [0.3851]
No. of cross-sections	29	No. of cross-sections	29
Log likelihood	4432.934	Log likelihood	4490.227
Observations	3,334	Observations	3,334

Notes: ***, **, and * denote statistically significant at 1%, 5%, and 10%, respectively, while ( $\cdot$ ) and [ $\cdot$ ] are, respectively, standard error and probability values.

Estimates of the linear model reported in Table 3 reveal that, in the short run, oil prices exert a strong negative influence on unemployment. This implies that an increase in oil prices greatly reduces unemployment, which is consistent with the findings of Kocaarslan et al. (2020), which show that an oil price increase translates into an increase in unemployment in the US. The finding also conforms to that reported by Keane and Prasad (1996), which shows that oil price increases reduce/raise employment/unemployment in the short run. This finding, however, contradicts the supply-side shock effect, which proposes that an increase in the price of oil worsens the unemployment situation in a country since an increase in the price of oil is synonymous with an increase in the cost of production, lower output, and rising job losses. Similarly, the finding is inconsistent with the proposition of the income (wealth) transfer effect, which suggests that an increase in the price of oil transfers wealth from oil-importing countries to their exporting counterparts. This potentially increases unemployment in the former group of countries. We hypothesize that the reason for the short-run inverse linear (symmetric) relationship between oil prices and unemployment is the willingness of employees, especially in oil-intensive industries, to accept pay cuts, and the willingness of people seeking jobs to accept a low wage when oil prices rise. In addition, governments’ efforts and policies to protect employees from losing their jobs when the price of oil rises might contribute to this short-run inverse linear relationship between the price of oil and unemployment.

In the long run, we observe an insignificant positive linear relationship between oil prices and unemployment. It suggests that oil prices increase unemployment slightly in the long run. This observation supports (Caruth et al., 1998), who observe that real input prices (interest rate and oil price) have a positive effect on unemployment. Our established positive, long-run linear association between oil price and unemployment corroborates Caporale and Gil-Alana (2002). This observation is consistent with the supply-side shock and income (wealth) transfer effects propositions of a positive association between oil price and unemployment. We anticipated this result since a persistent rise in oil prices will, in the long run, leave employers (particularly those whose activities are highly oil-reliant) with no option but to maintain a low workforce, thus aggravating the unemployment situation in these countries.

Looking at our nonlinear result in Table 3, we observe that in the short run, a rise in the price of oil strongly improves the unemployment situation of the oil-importing economies in Africa. This short-run positive relationship between an increase in oil prices and unemployment confirms Nusair (2020), who observes that, in the short run, an increase in oil prices reduces unemployment. We have earlier hypothesized the likely explanations for the short-run negative relationship between a rise in oil prices and unemployment in this group of countries. However, for emphasis, given that oil is a crucial input in many processes of production, an increase in its price will inevitably result in increased production costs for firms, activities, and industries heavily dependent on oil. This potentially prompts producers to downsize their workforce as a measure to reduce costs. In reaction to this potential consequence, governments in those countries are likely to implement measures to protect workers from layoffs. Besides, workers who are fully conversant with the situation in the labor market may accept wage cuts, while people searching for new jobs may accept lower wages. All these, in addition to the time lag between a fall in the price of oil and reactions to the fall in the price of oil by employers by way of production cuts and downsizing of the workforce, potentially account for this short-run inverse relationship between oil price increases and unemployment in these countries.

A fall in the price of oil significantly increases unemployment in this group of countries. The tendency for employees to bargain for a high wage when the cost of inputs, such as oil prices for oil-reliant firms, industries, or activities, falls might explain this. When organized employees succeed in negotiating with employers (most especially organized employers) for wage increases when oil prices fall, the likelihood is that the latter will reduce their workforce to minimize the cost of production. The possible outcome of this is a rise in the rate of unemployment in the short run, ceteris paribus. Besides, the inability of employers to adjust their production scale instantaneously to a fall in the price of oil will cause a fall in the price of oil to have a positive effect on unemployment in the short run.

Turning to the long-run asymmetric model estimates, we detect that an increase in oil prices has a strong positive effect on unemployment, whereas a significant inverse relationship exists between oil price falls and unemployment. Nusair (2020) found a similar relationship between an increase in oil prices and unemployment in the long run. It is necessary to state that our observed long-term relationship between unemployment and an increase or a decrease in the price of oil is natural. Thus, the relationship is expected, particularly for oil-importing economies. The finding is consistent with the supply-side shock theory. The theory suggests that an increase in production costs resulting from an increase in oil prices reduces productivity growth, thereby reducing real wage growth and translating into an increase in unemployment, and vice versa. Likewise, the transfer of wealth theory underpins our findings. The theory postulates that a rise in the price of oil shifts purchasing power from countries that import to countries that export oil (Fried and Schultze, 1975; Brown and Yucel, 2002). The increases in the price of oil raise the amount of money oil-importing economies spend on oil, dwindling their income and ability to buy goods and services. This translates into a fall in aggregate demand and output and raising unemployment, all else equal.

In the long run, a rise in the price of oil increases the marginal cost of production and reduces marginal revenue and total profits for firms. As a result, firms maintain only smaller production scales and workforce, thus strongly aggravating the unemployment situation in oil-importing economies, all else being constant. However, a decline in the price of oil signifies a fall in the cost of production for industries that heavily use oil as a key input. In addition, the income transfer effect proposes that a fall in the price of oil transfers wealth from oil-producing countries to oil-importing countries and vice versa. This indicates that a fall in the price of oil is synonymous with an increase in the demand for goods and services in oil-importing countries. Ceteris paribus, this increase in demand for goods and services provides incentives for producers to expand their scale of operation(s) by employing more inputs, labor, inclusive, in the long run, thus reducing the unemployment rate in these countries in the long run.

4.1. Robustness check

Our short- and long-run coefficients of positive and negative changes in oil prices reveal evidence of an asymmetric response of unemployment to negative and positive changes in oil prices. The estimates show that unemployment does not respond equally to a rise or fall in the oil price since both the short- and long-run coefficients of positive and negative oil price changes are not equivalent. Similar observations were reported by Bocklet and Baek (2017), Kocaaslan (2019), and Nusair (2020). Table 3 shows that unemployment reacts more to negative than to positive oil price changes. The finding is not natural in the relationship between unemployment and oil prices because an increase in oil prices lasts longer than a decrease in oil prices. Our finding suggests that employers of labor respond more in the short and long run to a fall than to a rise in oil prices. It means that employers of labor in oil-importing economies in Africa have more confidence in the continuity of a fall in oil prices over a long period than in the continuity of a rise in those prices.

A robustness check is necessary in this study since our results must be reliable for policymaking. We verify the robustness of our result as follows: First, we replace the Brent oil price index, which is the oil price index for our main estimation, with the WTI index. Second, we estimate our models (linear and nonlinear) using lower frequency data (annual) with the oil price index for our main estimation (Brent). Third, for the lower frequency models, we replace the Brent oil price index with the WTI index. Just like in the case of our main estimation, we employ the MG and the PMG estimators to estimate the robustness models, and the Hausman tests suggest that for all the symmetric models, the MG is the efficient estimator. In contrast, the Hausman test result for all the asymmetric models confirms that the PMG is the most efficient estimator. We report the results of the robustness verification in Tables 4–6. Specifically, Table 4 is the robustness check when quarterly data is used with the WTI oil price index. Table 5 is the robustness result based on annual data with the Brent oil price index. Table 6 reports the robustness check result based on annual data with the WTI oil price.

**Table 4. Short- and long-run linear and nonlinear panel ARDL models estimates (robustness result using quarterly data and WTI oil price index).**
Linear model estimates		Nonlinear model estimates
Variable	MG	Variable	PMG
Constant	$-$ 0.356** ( $-$ 0.161)	Constant	0.0932*** ( $-$ 0.0301)
$Δ p_{t}$	$-$ 0.0009*** ( $-$ 0.0003)	$Δ p_{t}^{+}$	$-$ 0.0011*** ( $-$ 0.0004)
		$Δ p_{t}^{-}$	0.0026*** ( $-$ 0.0008)
$p_{t}$	0.0130 ( $-$ 0.0147)	$p_{t}^{+}$	0.209** ( $-$ 0.0824)
		$p_{t}^{-}$	$-$ 0.520*** ( $-$ 0.168)
${\hat{ς}}_{i, t - 1}$	$-$ 0.0223*** ( $-$ 0.0061)	${\hat{μ}}_{i, t - 1}$	$-$ 0.0072*** ( $-$ 0.0019)
Hausman test $X_{k}^{2}$	303.43 [0.0000]	Hausman test $X_{k}^{2}$	0.30 [0.8622]
No. of cross-sections	29	No. of cross-sections	29
Log likelihood	4426.602	Log likelihood	4527.843
Observations	3,334	Observations	3,334

Notes: *** and ** denote statistically significant at 1% and 5%, respectively, while ( $\cdot$ ) and [ $\cdot$ ] are, respectively, standard error and probability values.

**Table 5. Short- and long-run linear and nonlinear panel ARDL models estimates (robustness result using annual data with Brent oil price index).**
Linear model estimates		Nonlinear model estimates
Variable	PMG	Variable	PMG
Constant	2.024*** ( $-$ 0.577)	Constant	1.708*** ( $-$ 0.431)
$Δ p_{t}$	$-$ 0.0026* ( $-$ 0.0013)	$Δ p_{t}^{+}$	$- 0.0020$ ( $-$ 0.0014)
		$Δ p_{t}^{-}$	0.0011 ( $-$ 0.0008)
$p_{t}$	0.0168 ( $-$ 0.0160)	$p_{t}^{+}$	0.0103* ( $-$ 0.0056)
		$p_{t}^{-}$	$-$ 0.0256*** ( $-$ 0.0068)
${\hat{ς}}_{i, t - 1}$	$-$ 0.190*** ( $-$ 0.0327)	${\hat{μ}}_{i, t - 1}$	$-$ 0.165*** ( $-$ 0.0213)
Hausman test $X_{k}^{2}$	74.81 [0.0000]	Hausman test $X_{k}^{2}$	2.04 [0.3606]
No. of cross-sections	29	No. of cross-sections	29
Log likelihood	200.0434	Log likelihood	194.1328
Observations	812	Observations	811

Notes: ***, **, and * denote statistically significant at 1%, 5%, and 10%, respectively, while ( $\cdot$ ) and [ $\cdot$ ] are, respectively, standard error and probability values.

**Table 6. Short- and long-run linear and nonlinear panel ARDL models estimates (robustness result with annul data and WTI oil price index).**
Linear model estimates		Nonlinear model estimates
Variable	MG	Variable	PMG
Constant	2.034*** ( $-$ 0.624)	Constant	1.650*** ( $-$ 0.428)
$Δ p_{t}$	$-$ 0.0025* ( $-$ 0.0014)	$Δ p_{t}^{+}$	$-$ 0.0036*** ( $-$ 0.0017)
		$Δ p_{t}^{-}$	0.0021** ( $-$ 0.0009)
$p_{t}$	0.0137 ( $-$ 0.0128)	$p_{t}^{+}$	0.0104 ( $-$ 0.0070)
		$p_{t}^{-}$	$-$ 0.0313*** ( $-$ 0.0085)
${\hat{ς}}_{i, t - 1}$	$-$ 0.179*** ( $-$ 0.0337)	${\hat{μ}}_{i, t - 1}$	$-$ 0.158*** ( $-$ 0.0212)
Hausman test $X_{k}^{2}$	72.45 [0.0000]	Hausman test $X_{k}^{2}$	2.30 [0.3170]
No. of cross-sections	29	No. of cross-sections	29
Log likelihood	204.7097	Log likelihood	203.7385
Observations	812	Observations	811

Notes: ***, **, and * denote statistically significant at 1%, 5%, and 10%, respectively, while ( $\cdot$ ) and [ $\cdot$ ] are, respectively, standard error values and probability values.

All the results of our robustness verifications lend credence to our main findings. The results suggest that the relationships (symmetric and asymmetric) between oil prices and unemployment in oil-exporting economies in Africa are insensitive to the oil price index utilized and the frequency of data (quarterly or annually). Those results do not deviate significantly from the main result. For the symmetric estimates, a significant negative relationship exists between oil prices and unemployment in the short run. In the long run, however, there is a positive but insignificant association between the price of oil and unemployment. Likewise, for the asymmetric model, our estimates validate the short-run inverse relationship between a positive change in oil price and unemployment and the short-run negative association between a fall in oil price and unemployment. In the long run, the results confirm that a positive change (increase) in the price of oil worsens the unemployment situation of oil-exporting economies in Africa, whereas a fall in the price of oil declines unemployment in these countries. The robustness verification results authenticate our main findings from the estimates of the symmetric and asymmetric panel ARDL models.

5. Summary and Recommendations

We employ linear and nonlinear panel ARDL models to examine the relationship between oil prices and unemployment in 29 African oil-importing economies. Quarterly and annual data for the period 1991–2019 are utilized for the study. After robustness verification, the findings of our symmetric ARDL model suggest that, in the short run, oil prices have a significant negative effect on unemployment. In the long run, however, we observe that oil prices have a positive but insignificant effect on unemployment in African oil-importing economies. About the asymmetric ARDL model estimates, we detect that in the short run, positive and negative oil price changes, respectively, have significant negative and positive effects on unemployment. But in the long run, positive changes in oil prices have a strong positive effect on unemployment, while negative changes in oil prices have a significant negative influence on unemployment. Our findings indicate a complex association between oil prices and unemployment in oil-importing economies in Africa, with the short-run effect of oil prices on unemployment being different from the long-run effects. The findings highlight the importance of considering the short- and long-run dynamics of the relationship between oil prices and unemployment, as well as the direction of oil price changes, in understanding the effects of oil prices and unemployment in African oil-importing economies.

The insights from our study contribute to relevant literature on several fronts. The insights underscore oil prices’ different effects on unemployment in the short and long run, strengthening the nuanced understanding of the association between the variables. Our study focuses on oil importing economies. Previous research did not focus on evaluating the nexus between oil prices and unemployment in the context of oil-importing economies in Africa. Filling this gap in the literature, our study potentially contributes to the literature. Besides, by using symmetric and asymmetric panel ARDL models, the findings presented reveal that negative and positive oil price changes’ effects on unemployment differ. This symmetric effect of unemployment reaction to oil price emphasizes the complex associations between the variables, adding depth to insights into how different movements in oil prices impact unemployment. Furthermore, most literature on oil prices and unemployment focuses on developed oil-importing or exporting countries or economies. By focusing on 29 oil-importing developing economies, our study potentially widens the literature scope on oil prices and unemployment.

Our findings have some important policy implications. For African oil-importing economies, the oil price is determined exogenously, as the economies have negligible or no influence on global oil prices. This exogenous factor (the oil price) can, therefore, affect the effectiveness of the policies of the various governments of these economies that are intended to address unemployment. Thus, for the effectiveness of policies aimed at reducing unemployment in this group of economies, various governments of these countries should take into account changes in the global oil price, since oil price changes can affect the effectiveness of their policies aimed at addressing unemployment. We thus recommend that policymakers put in place short- and long-run measures to mitigate the adverse effects of oil price increases on unemployment. In the short run, measures such as training initiatives, the creation of job programs, and providing unemployment benefits to the unemployed during times of increasing oil prices have the potential to mitigate the adverse effects of rising oil prices on unemployment in those economies. The long-term measures could include diversifying the African oil-importing countries’ economies. This could involve investing in the services sector, manufacturing, and agriculture that do not rely heavily on energy from oil sources to lessen their dependence on oil and create sustainable employment sources. Besides, they may invest in alternative energy sources, such as renewable energy sources that are cheaper and more sustainable, to further lighten their dependence on oil for energy and, by extension, mitigate the adverse effects of oil price increases on unemployment.

Policymakers in these countries should invest in human capital to address the long-term positive impact of oil price increases on unemployment. They should prioritize investments in human capital, focusing on education, training, and skills acquisition to equip their workforce with the necessary skills to adapt to changing economic situations and transition to new industries or sectors. In addition, measures that promote labor market flexibility to enhance easy adjustments to oil price changes should be considered to ease the adjustment of the labor market to oil price changes. Furthermore, innovation and entrepreneurship should be promoted, and business-friendly environments should be encouraged to encourage job creation and reduce unemployment in those countries.

ORCID

Yunana Ishemu Zumba https://orcid.org/0000-0001-9835-5317

Magai Abe https://orcid.org/0000-0002-8761-1562

Uten Paul https://orcid.org/0009-0009-1039-852X

Notes

¹ https://databank.worldbank.org/source/world-development-indicators.

² http://www.eia.gov/dnav/pet/pet_pri_spt_s1_d.htm.

References

Adeosun, O.A., O. Olayemi, O. Tabash and S.A. Anagreh (2023). The dynamics of oil prices, uncertainty measures and unemployment: a time and frequency approach. China Finance Review International, 13(4), pp. 682–713. Crossref, Google Scholar
Ahmed, F. (2013). The Effect of Oil Prices on Unemployment: Evidence from Pakistan. Business and Economic Research Journal, 4(1), pp. 43–57. Google Scholar
Ahmed, R., X.H. Chen, C. Kumpamool and D.T.K. Nguyen (2023a). Inflation, Oil Prices, and Economic Activity in Recent Crisis: Evidence from the UK. Energy Economics, 126, p. 106918. Crossref, Google Scholar
Ahmed, M.I., Q.F. Farah and R.P. Kishan (2023b). Oil Price Uncertainty and Unemployment Dynamics: Nonlinearities Matter. Energy Economics, 125, p. 106806. Crossref, Google Scholar
Baltagi, B.H. (2013). Econometric Analysis of Panel Data. John Wiley & Sons. Google Scholar
Bocklet, J. and J. Baek (2017). Do Oil Price Changes Have Symmetric or Symmetric Effects on the Unemployment Rate? Empirical Evidence from Alaska. Energy Sources, Part B: Economics, Planning and Policy, 12(4), pp. 1–6. Google Scholar
Breitung, J. (2001). The Local Power of Some Unit Root Tests for Panel Data. In B.H. Baltagi (ed.), Nonstationary Panels, Panel Cointegration and Dynamic Panels, Advances in Econometrics. Elsevier, Amsterdam, pp. 161–177. Google Scholar
Brown, S.P.A. and M.K. Yucel (2002). Energy Prices and Aggregate Economic Activity: An Interpretative Survey. Quarterly Review of Economics and Finance, 42(2), pp. 193–208. Crossref, Google Scholar
Burbidge, J. and A. Harrison (1984). Testing the Effects of Oil Price Rises Using Vector Autoregressions. International Economic Review, 25(2), pp. 459–484. Crossref, Google Scholar
Caporale, G.M. and L.A. Gil-Alana (2002). Unemployment and Input Prices: A Fractional Cointegration Approach. Applied Economics Letters, 9(6), pp. 347–351. Crossref, Google Scholar
Caruth, A.A., M.A. Hooker and A.J. Oswald (1998). Unemployment Equilibria and Input Prices: Theory and Evidence from the United States. Review of Economics and Statistics, 80(4), pp. 621–628. Crossref, Google Scholar
Chan, Y.T. and Y. Dong (2022). How Does Oil Price Volatility Affect Unemployment Rates? A Dynamic Stochastic General Equilibrium Model. Economic Modelling, 114, p. 105935. Crossref, Google Scholar
Cheratian, I., M.R. Farzanegan and S. Goltabar (2019). Oil Price Shocks and Unemployment Rate: New Evidence from the MENA Region. Joint Discussion Paper Series in Economics, No. 31-2019. Google Scholar
Cuestas, J.C. and L.A. Gil-Alana (2017). Oil Price Shocks and Unemployment in Central and Eastern Europe. Economic Systems, 42(1), pp. 164–173. Crossref, Google Scholar
Cuestas, J.C. and J. Ordonez (2018). Oil Prices and Unemployment in the UK Before and After the Crisis: A Bayesian VAR Approach. A Note. Physica A: Statistical Mechanics and its Applications, 510, pp. 200–207. Crossref, Google Scholar
Dogrul, H.G. and U. Soytas (2010). Relationship Between Oil Prices, Interest Rate, and Unemployment: Evidence from an Emerging Market. Energy Economics, 32(6), pp. 1523–1528. Crossref, Google Scholar
Dohner, R.S. (1981). Energy Prices, Economic Activity and Inflation: Survey of Issues and Results. In K.A. Mork (ed.), Energy Prices, Inflation and Economic Activity. Ballinger, Cambridge, MA. Google Scholar
Fried, E.R. and C.L. Schultze (1975). Higher Oil Prices and the World Economy. The Brookings Institution, Washington, D.C. Google Scholar
Hadri, K. (2000). Testing for Stationarity in Heterogeneous Panel Data. Econometrics Journal, 3(2), pp. 148–161. Crossref, Google Scholar
Hamilton, J.D. (1983). Oil and the Macroeconomy Since World War II. Journal of Political Economy, 91(2), pp. 228–248. Crossref, Google Scholar
Hamilton, J.D. (1988). A Neoclassical Model of Unemployment and the Business Cycle. Journal of Political Economy, 96(3), pp. 593–617. Crossref, Google Scholar
Harris, R.D.F. and E. Tzavalis (1999). Inference for Unit Roots in Dynamic Panels Where the Time Dimension Is Fixed. Journal of Economics, 91(2), pp. 201–226. Crossref, Google Scholar
Im, K.S., M.H. Pesaran and Y. Shin (2003). Testing for Unit Roots in Heterogeneous Panels. Journal of Economics, 115(1), pp. 53–74. Crossref, Google Scholar
Karlsson, H., Y. Li and G. Shukur (2018). The Causal Nexus Between Oil Prices, Interest Rates and Unemployment in Norway Using Wavelet Methods. Sustainability, 10(8), pp. 1–15. Google Scholar
Katircioglu, S.T., K. Sertoglu, M. Candemir and M. Mercan (2015). Oil Price Movements and Macroeconomic Performance: Evidence from Twenty-Six OECD Countries. Renewable and Sustainable Energy Review, 44(C), pp. 257–270. Crossref, Google Scholar
Keane, M. and E. Prasad (1996). The Employment and Wage Effects of Oil Price Changes: A Sectoral Analysis. Review of Economics and Statistics, 78(3), pp. 389–400. Crossref, Google Scholar
Kocaarslan, B., M.A. Soytas and U. Soytas (2020). The Asymmetric Impact of Oil Prices, Interest Rates and Oil Prices Uncertainty on Unemployment in the US. Energy Economics, 86, p. 104625, https://doi.org/10.1016/j.eneco.2019.104625. Crossref, Google Scholar
Kocaaslan, O.K. (2019). Oil Price Uncertainty and Unemployment. Energy Economics, 81, pp. 577–583. Crossref, Google Scholar
Levin, A., C. Lin and C.J. Chu (2002). Unit Root Tests in Panel Data: Asymptotic and Finite Sample Properties. Journal of Econometrics, 108(1), pp. 1–24. Crossref, Google Scholar
Loschel, A. and O. Ulrich (2009). Oil and Unemployment in Germany. Journal of Economics and Statistics, 229(2–3), pp. 146–162. Google Scholar
Loungani, P. (1986). Oil Price Shocks and the Macroeconomy: The Role of Price Volatility. Energy Journal, 16, pp. 3–35. Google Scholar
Maddala, G.S. and S. Wu (1999). A Comparative Study of Unit Root Tests with Panel Data and a New Simple Test. Oxford Bulletin of Economics and Statistics, 61(S1), pp. 631–652. Crossref, Google Scholar
Mankiw, N.G. (2010). Macroeconomics, 7th ed. Worth Publishers, New York. Google Scholar
Mian, A. and A. Sufi (2012). What Explains High Unemployment? The Aggregate Demand Channel. Working Paper No. 17830. Available at http://www.nber.org/papers/w17830. Google Scholar
Michieka, N.M. and R.S. Gearhart (2019). Oil Price Dynamics and Sectoral Employment in the US. Economic Analysis and Policy, 62, pp. 140–149. Crossref, Google Scholar
Munir, K. and M. Iftikhar (2021). Asymmetric Impact of FDI and Exchange Rate on Tourism: Evidence from Panel Linear and Nonlinear ARDL Model. SAGE Open, 11, pp. 1–10. Crossref, Google Scholar
Munir, K. and N. Riaz (2019). Energy Consumption and Environmental Quality in South Asia: Evidence from Panel Non-Linear ARDL. Environmental Science and Pollution Research, 26, pp. 29307–29315. Crossref, Google Scholar
Nusair, S.A. (2020). The Asymmetric Effects of Oil Price Changes on Unemployment: Evidence from Canada and the US. The Journal of Economic Asymmetries, 21, p. e00153. Crossref, Google Scholar
Ordonez, J., M. Monfort and J.C. Cuestas (2019). Oil Prices, Unemployment and the Financial Crisis in Oil-Importing Countries: The Case of Spain. Energy, 181, pp. 625–634. Crossref, Google Scholar
Ordonez, J., H. Sala and J. Silva (2010). Oil Prices Shocks and Labour Market Fluctuations. Forschungsinstitut zur Zukunft der Arbeit Institute for the Study of Labour, IZA Discussion Paper No. 5096. Google Scholar
Pesaran, M.H. (2007). A Simple Panel Unit Root Test in the Presence of Cross-Section Dependence. Journal of Applied Economics, 22(2), pp. 265–312. Crossref, Google Scholar
Pesaran, M.H. and Y. Shin (1999) An Autoregressive Distributed Lag Modelling Approach to Cointegration Analysis. In: Strom, S. (ed.), Chapter 11 in Econometrics and Economic Theory in the 20th Century the Ragnar Frisch Centennial Symposium, Cambridge University Press, Cambridge, pp. 371–413. Crossref, Google Scholar
Pesaran, M.H., Y. Shin and R.J. Smith (2001). Bounds Testing Approaches to the Analysis of Level Relationships. Journal of Applied Economics, 16(3), pp. 289–326. Crossref, Google Scholar
Ran, J. and J.P. Voon (2012). Does Oil Price Shock Affect Small Open Economies? Evidence from Hong Kong, Singapore, South Korea and Taiwan. Applied Economics Letters, 19(16), pp. 1599–1602. Crossref, Google Scholar
Saint-Paul, G. (1999). The Political Consequences of Unemployment. Universitat Pompeu Fabra, Department of Economics & Business, Working Paper No. 343. Google Scholar
Senzangakhona, P.N. and I. Choga (2015). Crude Oil Prices and Unemployment in South Africa: 1990–2010. Mediterranean Journal of Social Sciences, 6(2), p. 407. Google Scholar
Shin, Y., B. Yu and M. Greenwood-Nimmo (2014). Modelling Asymmetric Cointegration and Dynamic Multipliers in a Nonlinear ARDL Framework. In R. Sickels and W. Horrace (eds.), Festschrift in Honor of Peter Schmidt: Econometric Methods and Applications. Springer, pp. 281–314. Crossref, Google Scholar
Trang, N., T.N. Tho and T.T.H Hong (2017). The Impact of Oil Price on the Growth, Inflation, Unemployment and Budget Deficit of Vietnam. International Journal of Energy Economics and Policy, Econjournals, 7(3), pp. 42–49. Google Scholar
Uri, N.D. (1996). Crude Oil Price Volatility and Unemployment in the United States. Energy, 21(1), pp. 29–38. Crossref, Google Scholar
World Bank (2022). World Development Indicators. https://databank.worldbank.org/source/world-development-indicators Google Scholar
Zumba, Y. (2022). Stock Prices and Economic Activities in Nigeria: Sector Level Evidence. Macroeconomics and Finance in Emerging Market Economies, 17, pp. 5–24, https://doi.org/10.1080/17520843.2022.2027619. Crossref, Google Scholar

Vol. 06, No. Supp01

Metrics

Downloaded 38 times

History

Received 16 February 2024

Revised 15 October 2024

Accepted 21 October 2024

Published: 27 January 2025

Information

This is an open access article published by World Scientific Publishing Company. It is distributed under the terms of the Creative Commons Attribution 4.0 (CC BY) License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.

Keywords

PDF download