**6. Why Do Jordanian Firms Hold Cash?**

We continued to investigate why firms with government ownership in Jordan hold more cash by following Gao et al. [63], examining whether they use cash for investments, use capital expenditures as a proxy or for pay outs, or use dividends as a proxy. The results are reported in Appendix A Table A1 for the former and Appendix A Table A2 for the latter.

Other types of ownership did not have an impact on the level of cash holdings and were found to be statistically insignificant, which is very surprising due to the high average stakes held by the owners. Government ownership is more salient to the levels of cash, even when a high number of shares are held by other types of owners. In fact, a study by Tayem [64] found that ownership (by the largest shareholder) had no effect on the level of cash holdings in Jordan. It could be inferred that ownership structure as a governance mechanism in Jordan is not effective, except for the government for political reasons. However, this question will be left for future research.

According to our results, firms with higher levels of government ownership tend to hold higher levels of cash. On the one hand, this could be attributed to the agency problems associated with government ownership, as predicted by the free cash flow theory. While government owners are more interested in political and social goals rather than shareholder interests, the managers of such firms are more likely to be directed towards also exploiting other shareholders' interests [64]. The weakness of corporate governance and the moral hazards associated with government ownership also need to be considered [65–67]. Therefore, in such an environment of weak internal monitoring, managers find it more convenient to hold greater levels of cash and reduce the need for external financing and their associated monitoring and scrutiny.

On the other hand, government-owned firms might use cash to make investments in research and development, acquisitions, and capital expenditure or to pay dividends. Therefore, we extend our analysis to examine why such firms hold higher levels of cash. To further understand this question, we investigate the effect of government ownership on the relationship between cash holdings and pay Not necessarily-out policies, and cash holdings and investment decisions.

Specifically, we examine the interaction between government ownership and cash holdings in investment decisions and pay-out policies. Capital expenditure is employed as a proxy for investment policies, and dividends as a proxy for pay-out policies. We were unable to examine other proxies, for example, research and development and acquisitions, due to data unavailability.

As shown in Appendix A Table A1, the coefficient on Govt. × Ch is statistically insignificant; therefore, according to the results, as government ownership increases, firms are unlikely to increase capital expenditure. The coefficient on Govt × Ch, as shown in Appendix A Table A2, is also insignificant, indicating that as government ownership increases, firms are unlikely to use more cash to pay dividends.

The evidence suggests that, as government ownership increases, firms in Jordan do not hoard higher levels of cash for investments nor for paying dividends. Alternatively, we interpret our results to support the expectations of free cash flow theory, that higher

government ownership is associated with more agency problems, which is consistent with the work of Firth et al. [68] and Chen et al. [14]. As the most liquid asset, cash is expected to be the first and most vulnerable to be exploited by management [3]; therefore, in light of government political goals, the way firms manage their cash holdings can be a clear measure of the existence of agency problems. Moreover, according to our results, managers in Jordan are more likely to implement the convenience mechanism [68]; higher cash holdings keep firms at a convenient and significant distance from market scrutiny, so managers prefer the convenience of cash and hold higher levels of it.

Our results support the common understanding that government ownership suffers from poorly structured governance and moral hazards. It was found to increase agency problems, so it is recommended that government ownership policy is revised to clarify its roles as an owner of firms and to set the objectives and expectations of its business ownership to ensure that it acts as a diligent owner and maintains an arm's length relation with the client. Moreover, the Jordanian government is considering the privatization of more listed firms in an attempt to enhance the performance of the capital market; the results of this research provide support for this step to be evaluated.
