Discussion of ‘Government ownership, corporate governance and tax aggressiveness: evidence from China’

1. Introduction

Using a sample of Chinese A-share listed firms over 2003–2009, Chan et al. (2013) (hereafter CMZ) find that the government ownership and board characteristics affect firms' tax avoidance behaviour. In particular, they find that compared with government-controlled firms, non-government-controlled firms have more aggressive tax strategies. In addition, they find firms with higher percentages of the board shareholdings and with the CEOs also serving as the board chairman are more aggressive in tax planning and these relationships are more pronounced for the non-government-controlled firms. Moreover, local government-controlled firms in less developed regions are found to be more tax aggressive than the other government-controlled firms.

There is a substantial and growing literature on the role of firms' government ownership in affecting their tax avoidance behaviour in China. For example, Wu (2009) finds that firms with higher state ownership tend to maintain a closer relationship with the government and shoulder more social responsibilities by paying more corporate taxes. Evidences reported by Zeng (2010) and Wu et al. (2013) confirm this finding. In addition, Wu et al. (2013) find that the preferential taxation of private firms is associated with local governments' incentives to promote local economic growth. CMZ (2013) hence adds on and extends this stream of research by introducing corporate governance elements into the current literature.

2. The links between government ownership and tax aggressiveness

CMZ predict that government-controlled firms in China will have more aggressive tax strategies than the non-government-controlled counterparts by making three arguments. The first argument is that the government ownership acts similar to long-term institutional ownership in the United States which can reduce tax aggressiveness in US firms because institutional owners are more concerned about the long-term consequences of aggressive tax strategy. To me, this argument is rather loose as the non-government-controlled firms in China are also tightly controlled by the controlling shareholders who can play the same roles as other long-term institutional shareholders. Given both government-controlled firms and non-government-controlled firms having long-term institutional shareholders, this argument is problematic. However, I do agree with their second argument that the government-controlled firms are required to achieve political and social objectives by paying more corporate taxes compared with the non-government-controlled firms. Their third argument, by contrast, states that managers in the government-controlled firms have incentives to collude with insiders and divert corporate resources for their own private benefits. Therefore, they would like to reduce the tax and retain more cash within the firms. My concern about this argument is that the typical agency problem suggested here also exists in privately held firms, as managers in non-government-controlled firms have the same incentives to conduct their diversionary practices by saving more cash in the firms via tax planning. This argument indeed cannot be used as a strong argument to explain the difference between government-controlled firms and non-government-controlled firms. Alternatively, a more interesting argument CMZ can use is that to reduce managers' agency and free cash flow problem in the government-controlled firms, the controlling shareholders (i.e. government) have incentives to cash out the financial resources of the firms in the form of tax collections. This may help to explain the higher tax rates paid by the government-controlled firms in China.

3. The links between corporate governance and tax aggressiveness

Studies using the US data have found contradictory evidence on the relationship between corporate governance and tax aggressiveness. Hanlon et al. (2005) and Rego and Wilson (2012) find various equity incentive measures, which are regarded as mechanisms to align managers' interests with shareholders, are positively correlated with proposed IRS deficiencies. In contrast, Desai and Dharmapala (2006) find the opposite results, and they argue that the negative correlation is due to managerial rent extraction story, which is similar to the third argument CMZ makes in developing their H1. Another study by Armstrong et al. (2010) finds mixed results.

Actually, in terms of the broader agency framework, there are two types of different agency problems in corporate governance studies. The first one is the type I agency problem, which stems from the conflict of interests between outside shareholders and managers; the second one is the type II agency problem, which roots in the conflict of interests between the controlling shareholders and minority shareholders. The former one is prevalent in the US firms, and the latter one is prevalent in the firms in emerging markets. In China, the type II agency problem is more severe and prevalent (Firth et al., 2007).

CMZ (2013) firstly expect that there is a difference in tax aggressiveness between companies with different percentages of independent directors on board in their H2a. But, in fact, researchers find that evidence on the independent directors' role in improving corporate governance of Chinese firms is very weak (Clarke, 2003; Lin et al., 2012). In addition, I wonder to what extent the independent directors will be involved in the tax planning process in China. As we know, the independent directors in China usually have two or three meetings per year to discuss various significant corporate issues (Lin et al., 2012). Tax planning might not be considered as the priority task by the independent directors. CMZ actually find no significant results for the relationship between the percentage of independent directors and tax aggressiveness, which is not surprising given the above reasons.

Regarding the second hypothesis (H2b), the authors expect that for non-government-controlled firms, companies with the same person serving as the CEO and the board chairman will be more tax aggressive because when leadership is concentrated in one decision-maker, there is a higher risk of having irregularities. The concern that I have is there may be some omitted variables affecting the relationship between the CEO duality and tax aggressiveness that CMZ document, for example, the controlling shareholders' shareholding percentages. The more shares the controlling shareholders hold, the more influential they are in firms' daily management. They can appoint the same person to be both the CEO and chairman of the board, and at the same time, they have more influence in managing taxes. As CMZ do not include the controlling shareholders' shareholding percentages in their regression models, the omitted variables concerns cannot be ignored, and the results should be interpreted with caution. A more vigorous research design could be conducted via a change test. CMZ should observe that the firm becomes more aggressive in tax savings after the chairman of the board also becomes CEO, or becomes less tax aggressive when the shared role is split between two different persons.

In developing their H2c, CMZ make the prediction that the shareholdings of board directors should be positively correlated with firms' tax aggressiveness. As mentioned above, it is still unclear how the board members' compensation is related to firms' tax planning as the board members are not directly responsible for the routine business of the firm. Following this criticism, I wonder whether it is necessary to consider the executives' incentives for tax aggressiveness by the firm as controlling variables in the regressions. For instance, one may expect that the increased shareholdings of CEOs and CFOs who are responsible for the detailed tax planning will be positively correlated with firms' tax aggressiveness because their rights to the firm's residual income will increase. Given the fact that in most Chinese firms managers' shareholding data are not easy to obtain, controlling for managers' cash compensation is becoming essential.

4. Institutions, government ownership, corporate governance and tax planning

CMZ find that institutional factors play different roles in different Chinese firms' tax saving behaviour. Firstly, they find the government-controlled firms in well-developed regions tend to pay more taxes than government-controlled firms in less developed regions, and this result is mainly driven by the local government-controlled firms in well-developed regions. However, these results are not robust when CMZ use alternative tax avoidance measure and when they use alternative controlling shareholders' definition. These mixed results leave open plenty of opportunities for future research. CMZ interpret the results they find in as being due to the less effective implementation of corporate governance measures in less developed regions. But an alternative explanation could be that the local government in these less developed regions is keener for saving the tax in their own regions as they are more likely to face a fiscal deficit than the governments in more developed regions (i.e. only 40 per cent of the corporate tax from local government-controlled firms are taken by the local government, whereas 60 per cent are taken by the central government).

In fact, Wu et al. (2013) find that the non-government-controlled firms in the less developed regions in China receive more preferential tax treatment. This is contradictory to CMZ's findings that the non-government-controlled firms are more tax aggressive in the more developed regions. These two contradictory empirical findings need to be reconciled in the future research. In addition, to what extent institutional factors affect tax avoidance by government/non-government firms differently needs to be further explored by researchers.

Another interesting angle to study, which CMZ fail to capture, is how institutional factors affect the role of corporate governance in firms' tax savings. CMZ actually do not interact the institutional factors with the corporate governance variables in their regression models. One may expect the role of corporate governance in affecting firms' tax savings would become more pronounced when the regional institutional factors to implement the firm-level corporate governance measures are stronger.

5. Potential tax research topics in China's setting

The unique setting in China does provide researchers a lot of interesting research topics in the tax area. For example, as mentioned above, governments acting as the controlling shareholders of Chinese government-controlled firms could use tax as a tool to reduce the free cash flow, which is under mangers' discretions and thus reduce the potential agency costs. This indeed extends the current tax literature by documenting another role of taxation is to mitigate the typical agency and free cash flow problem.

It is also worth exploring the economic consequences of paying more taxes in the government-controlled firms. For example, if the government-controlled firms are willing to pay more taxes, will the executives in these firms benefit from these generous tax payments? Consider the pay-off function of the individual managers in government-controlled firms under the assumption that they have large bargaining power when they negotiate with the governments. The managers in the government-controlled firms have to trade-off the marginal benefits against the marginal costs of managing taxes. If they choose to save more taxes, they can keep the cash resources within the firm and divert these cash resources for their individual gains such as investing in their own pet projects. If in equilibrium we observe the government-controlled firms do choose to pay more taxes, we can conjecture that the individual marginal benefits for the managers outweigh the individual marginal costs. More specifically, researchers may expect and find empirical evidence that the CEOs receive personal benefits from these additional tax payments. As Ke et al. (2012) find that the Chinese government-controlled firms are unlikely to grant large amounts of stock options or performance-driven compensation to their executives, one can expect that these CEOs would be more likely to get promoted in the future if they pay more taxes to the government.

Moreover, the variations within Chinese private firms' tax saving behaviours also remain unexplored. Wu et al. (2012) find evidence that the politically connected managers in Chinese private firms enjoy the tax benefits. Based on their findings, it is worth investigating how this personal connection with the government helps the firms to save taxes. In particular, one may expect the private firms in China to hire some CFOs or other executives who are previously connected to the local tax authority (proxied by previous working experience in the tax authority) to help them to pay less taxes.

As previously suggested in Li and Cai (2011), it would be also interesting to conduct research on how Chinese firms negotiate tax liabilities with the government because, in China, taxation is sometimes not dependent on firm's annual income, but a predetermined increase percentage set at the beginning of the year regardless of a firm's performance.

The minority shareholders' views on tax savings in Chinese firms also could be another interesting research topic. As discussed above, most of the non-government Chinese listed firms suffer from the type II agency problems, which are the conflict of interests between the controlling shareholders and minority shareholders. Will the minority shareholders of the non-government-controlled firms care so much about firms' tax saving as the remaining cash within the firms could be transferred directly to the controlling shareholders via various related party transactions or tunnelling activities? They could be indifferent on this issue as they will not benefit from this aggressive tax saving anyway. Put it differently, it will not be so value-enhancing for the minority shareholders in China if the type II agency problems are severe.

Another perspective to study the effects of ownership structure on Chinese firms' tax aggressiveness is to use the internal capital market framework. One important reason established in the current finance literature for why firms – especially those firms in emerging markets– build up conglomerates is to make better use of the internal capital market when the external capital market is less efficient (Desai et al., 2004). If Chinese non-government-controlled firms do make better use of the internal capital market by controlling a group of firms, one would expect that when there is a sudden shock in the financing needs of some firms in the conglomerate, the controlling shareholders of the unaffected firms would make the unaffected firms more aggressive in tax saving to generate more cash to feed the financing needs of the affected firms in the conglomerate.

6. Conclusions

This discussion reviews and discusses CMZ (2013) and makes several suggestions for future taxation research in China, a country that already has the second largest economy in the world. Hopefully, it will become more attractive to do China-based tax research, and I expect to see more research papers on Chinese taxation issues published in the top academic journals in the future.