AbstractThis paper examines the relationship between tax officers’ access to community information and the effectiveness of tax authorities in constraining tax avoidance in China. Numerous studies document that people can obtain valuable information in the community in which their daily activities take place (community information). Focusing on a valuable setting in China in which corporate taxes are collected by tax offices that are located very close to firms (with an average distance of 6.5 km), this study examines whether tax officers’ access to community information at a firm’s location can reduce the firm’s tax avoidance. I broadly define tax avoidance as the reduction of income taxes per dollar of pre-tax accounting earnings. Using a sample of publicly listed firms between 2007 and 2017, I first use geographic distance between tax offices and firms to measure the likelihood of tax officers’ having access to community information at a firm’s location and examine its effects on the listed firms’ level of tax avoidance. I find that the level of tax avoidance is higher when tax offices and firms are located far away from each other. I also find that the distance effect is stronger when firms’ tax officers and managers are socially connected.
Second, I use functional distance between tax offices and firms to measure the likelihood of tax officers’ having access to community information at a firm’s location, and examine its impacts on tax avoidance. I first proxy functional distance by the brightness of night light as captured by satellites. As brighter night light in an area indicates higher levels of economic and social activities in that area (Henderson et al. 2012), and consequently a greater likelihood of tax officers’ visiting the area and having access to community information in the area, I hypothesize that the distance effect on tax avoidance is stronger if firm location is brighter than tax office location. I find results consistent with these hypotheses. I further examine the effects of community information on tax avoidance by using the opening of a new shopping mall within a community. The opening of shopping malls will increase the probability of tax officers’ visiting the area where the shopping malls are located. The probability of tax officers’ visiting a firm’s location and having access to the community information in the firm’s location will increase (decrease) if shopping malls are located near to (far away from) the firm. Thus, I hypothesize that the level of tax avoidance will decrease for firms located nearer shopping malls and increase for firms located farther from shopping malls. Using a difference-in-difference research design, I find results consistent with my hypothesis. Specifically, firms located near (within 3 km of) shopping malls engage in less tax avoidance, while firms located far away (at least 3 km) from malls engage in more tax avoidance.
In addition to my major analyses, I conduct a host of additional tests to verify the robustness of my main results. Overall, my results consistently suggest that the availability of community information at a firm’s location to tax officers can significantly influence tax avoidance.
I contribute to the literature in several ways. First, my results have implications for the literature on determinants of tax avoidance. My study offers evidence that a new determinant—tax officers’ access to community information—can have a significant impact on the level of corporate tax avoidance. Second, I contribute to the strand of literature on the effects of community information in two ways. First, I extend the literature by demonstrating the effects of community information on regulatory effectiveness. Second, I create new proxies for the level of community information by using the brightness of night light as captured by satellites and the opening of new shopping malls. Lastly, my research findings have policy implications. My study suggests that the decentralization of tax collection systems can help to address the problem of information asymmetry between tax authorities and firms.
|Date of Award||19 Aug 2019|
|Supervisor||Man Lai Sonia WONG (Supervisor), Michael FIRTH (Co-supervisor) & Xiaofeng ZHAO (Co-supervisor)|