**6.1 Fixed effects model estimates**

Empirical results are presented in **Table 5**. As mentioned above, the estimation methodology accounts for both country and time effects. Thus, the table provides estimates by country and year.



*Notes: Robust standard errors in parentheses. a, b and c indicate significance at the 1%, 5%, and 10% levels, respectively.*

#### **Table 5.**

*Fixed effects model estimates (LSDV estimation).*

The results confirm the negative relationship between productivity-pay and youth unemployment observed in **Figures 2**–**4**. In the three columns, the *PP* coefficients are negative suggesting a negative influence of the productivity-pay link on the three youth unemployment rates. However, the absolute coefficient value is biggest for the female youth unemployment rate. The coefficient indicates that a one percent improvement in the strength of the pay-productivity link reduces the total, male, and female youth unemployment rates by 0.13 percentage point, 0.12 percentage point, and about 0.22 percentage point, respectively. These coefficients suggest therefore that the influence on the female youth unemployment rate is almost double the influence on the male youth unemployment rate. The payproductivity link plays a relatively more significant role in reducing female youth unemployment in the GCC countries.

*GROWTH* also has a negative influence, as we would normally expect. It is statistically significant in the total and female youth unemployment rate models. Interestingly, both *PP* and *GROWTH* have equal coefficients: A one percentage point increase in GDP per capita growth rate reduces the female youth unemployment rate by 0.22 percentage point. In contrast to [1], *GFINANCE* does not have a statistically significant influence on the different unemployment rates. The degree of urbanization, *URBAN*, reduces the total and male youth unemployment rates but not the female youth unemployment rate.

The coefficients of the country dummies – the country-specific effects - suggest that the country effects do not reduce total or male youth unemployment. This is obvious in Kuwait, Saudi Arabia, UAE and Qatar. Only in Kuwait and Qatar, the country-specific effects reduce the female youth unemployment rate.

The coefficients of the time dummies are negative and statistically significant in the total youth unemployment rate model. The time dummies are statistically significant at least at the 5 percent level in 2007–2009 and 2012. The increase in the world oil prices in 2006–2016, as **Figure 5** shows, may provide an explanation for this outcome. The oil price increase possibly feeds through GDP growth rate reducing the total youth unemployment rate.

*Youth Unemployment and Productivity-Pay in the GCC Countries DOI: http://dx.doi.org/10.5772/intechopen.99975*

**Figure 5.** *World crude (Brent) oil prices (1987–2020; \$/barrel).*

#### **6.2 Feasible generalized least squares estimates**

Although the GCC countries seem alike, they may be different from each other in how the youth unemployment rate is explained by the empirical model. Thus, I test for panel-level heteroscedasticity. I also test for autocorrelation in the error term. If either or both empirical issues are present, I plan to adopt a feasible generalized least squares (FGLS) estimation methodology.

Since the likelihood ratio χ2 test rejects the null hypothesis of panel-level homoscedasticity at the one percent level for the three youth unemployment rates, as **Table 6** shows, I adopt FGLS accounting for panel heteroskedasticity. The Wooldridge *F* test for autocorrelation fail to reject the null hypothesis of no panel-level autocorrelation in the case of total youth unemployment rate. However, the test rejects the null hypothesis at the five percent level in the case of male and female youth unemployment rates. I account for serial correlation in FGLS estimation of these two youth unemployment rates.

FGLS estimation results are presented in **Table 7**. Results confirm the negative influence that linking pay to productivity has on the total and female youth unemployment rates but not on the male youth unemployment rate. An improvement in the pay-productivity link strength by one percent reduces the total and female youth unemployment rate by 0.23 percentage point and 0.21 percentage point, respectively.


*Notes: The null hypothesis for the heteroscedasticity test is panel-level homoscedasticity. The null hypothesis for the autocorrelation test is the absence of (across) panel autocorrelation. a, b and c indicate significance at the 1%, 5%, and 10% levels, respectively.*

#### **Table 6.**

*Panel heteroscedasticity and autocorrelation tests.*


#### **Table 7.**

*FGLS estimation results.*

In contrast to the results of **Table 5**, empirical evidence shows that government expenditures increase the youth unemployment rates. A one percentage point increase in the share of government consumption expenditures in GDP increases the total and female youth unemployment rate by 0.44 percentage point and 0.4 percentage point, respectively. This result is in line with [1], who finds that an increase in government consumption expenditures (percentage of GDP) by one percentage point increases the total youth unemployment rate.

It might be the case that through consumption expenditures that the GCC governments put incomes in the hands of households, which discourage youth from searching for and taking jobs. From labor economics theory perspective, this is the negative (positive) income effect on work (leisure) hours. An increase in incomes discourage youth from increasing their labor supply. Alternatively, an increase in incomes encourage youth to take more leisure hours.

In contrast to the positive influence of government finance and like the result of **Table 5** for total youth unemployment, the degree of urbanization reduces the total youth unemployment rate. It also reduces the female youth unemployment rate. In addition, the GDP per capita growth rate reduces only the female youth unemployment rate but only marginally statistically.

### **7. Conclusion and policy recommendation**

The empirical analysis of the two estimation methodologies supports the beneficial influence of linking pay to productivity on the total and female youth unemployment rates. Labor market efficiency can help reduce youth unemployment. This suggests that having an equal link strength in both labor market segments rewards the national youth, especially females.

The adoption of policies that increase labor market flexibility and efficiency to reduce (youth) unemployment is not unusual. Spain, for example, in 2010 and

#### *Youth Unemployment and Productivity-Pay in the GCC Countries DOI: http://dx.doi.org/10.5772/intechopen.99975*

2012 relaxed contract termination, reduced severance pay, notice period, and business social security contributions [1, 10].19 The strength of collective bargaining was reduced, and enterprise-level agreement increased [10]. Firms have been able to withdraw from collective bargaining agreements one year from the agreement expiry date. Despite increasing labor market flexibility, youth unemployment remained high due to the *downward rigidity* of nominal wages, among other reasons [1]. The implicit GCC social contracts result in downward nominal wage rigidity.

The chapter recommends the adoption of labor efficiency policies and practices. Paying labor its economic value establishes efficiency in both labor market segments and encourage national youth to search for and take jobs not just in the government and state-owned enterprises – the traditional employer of nationals, but also in the private sector. Estimation results suggest that the magnitude of influence on female youth unemployment is relatively strong. GCC policy makers should be well-versed about the importance of the pay-productivity link and its implications for human resource management at the organization level. The obtained results suggest that national female youth are motivated by pay rewards.

Besides, efficiency helps clear the negative perceptions about labor market segmentation and the ensuing pay unfairness, and the gender bias. Labor efficiency helps in the efficient use of available domestic labor resources and in less reliance on foreign labor. By encouraging nationals of both genders to seek opportunities in the private sector, the adoption of labor efficiency policies and practices can help nationals of both genders develop a gamut of work-related skills including entrepreneurship. Entrepreneurship can help in job creation and reducing youth unemployment.

Finally, although this chapter makes the case of linking pay to productivity to *reduce the youth unemployment challenge*, I should point out though there is no onesolution-fit-all if the goal of reduction of the youth unemployment rates is regarded as merely political. GCC policymakers may consider alternative solutions to youth unemployment in changing economic, political, social environment. The remaining reality, however, is that each adopted solution has pros and cons that should be considered.
