**5. Explaining youth unemployment in the GCC countries - the empirical model**

The negative relationship between the different youth unemployment rates and productivity-pay observed in the above graphs is investigated empirically. Building on [9] who examines the relationship between labor market flexibility and youth unemployment, the empirical model includes the pay-productivity link as an explanatory variable.17 In this chapter, I explain the different youth unemployment rates (total, male, or female). The empirical model is now expressed as:

$$\text{UIR}\_{\text{\tiny{\tiny{\tiny{\tiny{\tiny{\tiny{\tiny{\tiny{\Gamma}}}}}}}} \beta\_0 + \beta\_1 \text{PP}\_{\text{\tiny{\tiny{\text{H}}}}} + \beta\_2 \text{GROV} \text{UTH}\_{\text{\tiny{\text{H}}}} + \beta\_3 \text{GFINANNCE}\_{\text{\tiny{\text{H}}}} + \beta\_4 \text{URBANN}\_{\text{\tiny{\text{H}}}} + \varepsilon\_{\text{tr}}$$

where *UR* is youth unemployment rate. *PP* is the pay-productivity link indicator (log). *GROWTH* is economic growth, a proxy for the expansion in

<sup>17</sup> In [6], control variables include an output gap measure to control for business cycle fluctuations, the size of government, the degree of trade openness, the rate of urbanization, population density, a crisis dummy, and the lagged level of unemployment rate(s).

job opportunities. It is measured by the annual GDP per capita growth rate. *GFINANCE* is government fiscal expenditures, which account for the major employment role the GCC governments play in hiring nationals. *GFINANCE* is measured by the general government final consumption expenditures (as a percentage of GDP). *URBAN* is the degree of urbanization in the economy. It accounts for the geographic concentration of business in urban areas and is measured by the percentage of urban to total population. ε is the error term. The subscripts *i* and *t* are country and time indexes.

Data on *PP* are obtained from GCI. Data on the other variables are obtained from WDI. The data covers the period 2007–2017.

Panel data models are used in estimation. I account for both the GCC country- and time-specific effects. Country-specific effects may include factors such as culture and traditions, which do not change by time at least in the short and medium terms. Time-specific effects are related to global factors, such as oil prices or foreign direct investment, which are not country-specific and may impact the GCC countries at specific years.
