FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W4385664767> ?p ?o ?g. }

• W4385664767 endingPage "27" @default.
• W4385664767 startingPage "1" @default.
• W4385664767 abstract "AbstractWe investigate how local political corruption shapes corporate financial reporting conservatism. Using a large sample of U.S. public firms, we find that firms located in areas with higher levels of political corruption tend to adopt greater accounting conservatism. We also find that firms in more corrupt areas bear greater political expropriation costs. Further analysis reveals that the positive effect of corruption on conservatism is stronger for firms with less bargaining power against corrupt officials, firms with lower public visibility, and firms with weaker dependence on the government for sales. Overall, our findings support the expropriation hypothesis that corrupt officials have incentives to expropriate resources from local firms, which induces them to adopt more conservative reporting strategies to shield their assets.Keywords: Political corruptionAccounting ConservatismExpropriation costJEL Classifications: D22D72G38M41 AcknowledgementWe appreciate helpful comments from Beatriz García Osma (Editor) and two anonymous reviewers. We also thank Sudipta Basu, Liandong Zhang, Qianqian Huang, and Yaxuan Qi, Ke Wang, and Jingyu Yang for their constructive suggestions. All the data used in this study are publicly available.Disclosure StatementNo potential conflict of interest was reported by the author(s).Supplemental Research MaterialsSupplemental data for this article can be accessed on the Taylor & Francis website, doi:10.1080/09638180.2023.2242399.Notes1 See e.g., Bushman and Piotroski (Citation2006), Ma et al. (Citation2020), Qiang (Citation2007), Ruch and Taylor (Citation2015), Watts (Citation2003), and among others.2 In section 4.3.1, we show direct evidence that on average firms located in more corrupt areas bear greater political expropriation costs. We acknowledge that, in some circumstances, firms may benefit from a corrupt economy, e.g., paying bribes in exchange for government procurement contracts: see discussion in section 4.3.4.3 See https://en.wikipedia.org/wiki/Vince_Fumo.4 This paper examines conditional conservatism, but frequently refers simply to conservatism. In the Internet Appendix, we discuss the distinction between conditional conservatism and unconditional conservatism.5 Our expropriation hypothesis is slightly different from current political cost explanation for conservatism. While the political cost story focuses on the public pressure faced by regulators from excess profitability of public firms, our expropriation hypothesis instead centers around the political pressure imposed on firms by the corrupt government.6 In a contemporaneous paper, Chang et al. (Citation2022) also document a positive association between political corruption and firm accounting conservatism in China. They show that a one-standard-deviation increase in corruption enhances conditional conservatism by 16%, which is comparable to the magnitude reported in our study. However, they do not adopt an instrumental variable approach in their endogeneity tests and do not provide direct evidence on the effect of corruption on expropriation risk. Moreover, their empirical setups are mostly built upon the unique institutions in China, which could limit the generalization of their findings to other countries.7 Following the literature, we assume that a typical U.S. firm’s major operations are focused around its headquarters. To account for firms operating in multiple states, we calculate a weighted-average conviction rate in the Internet Appendix; the results are similar.8 Besides the Basu’s (Citation1997) asymmetric timeliness coefficient, in the Internet Appendix, we also use four alternative proxies for conditional conservatism and find similar results.9 In unreported analyses, we examine the association between political corruption and discretionary accruals and document a positive relationship between them. Similar to Xu, Dao, and Petkevich (Citation2019), we also find that the effect is mainly driven by the negative accruals, which effectively leads to a decline of total earnings. These findings are consistent with the negative effect of political corruption on accounting conservatism.10 See, e.g., Ahmed and Duellman (Citation2013), Ball et al. (Citation2000), Chung and Wynn (Citation2008), Kravet (Citation2014), Kim et al. (Citation2013), García Lara et al. (Citation2016), Martin and Roychowdhury (Citation2015), and Ramalingegowda and Yu (Citation2012).11 Bushman and Piotroski (Citation2006) focus on the expropriation risks of forced nationalization or state intervention due to poor firm performance. These risks are relatively low in developed economies like the U.S., compared to emerging markets. Our study investigates another type of expropriation risk arising from politicians’ and/or government officials’ corrupt behaviors (Smith, Citation2016).12 See https://www.justice.gov/archive/opa/pr/2006/May/06_crm_269.html13 See, for example, Ahmed and Duellman (Citation2013), Lafond and Roychowdhury (Citation2008), and Ramalingegowda and Yu (Citation2012).14 There is some delay between crime commission and conviction. For example, Cordis and Milyo (Citation2016) show that the average gap from the date of criminal referral to conviction is about two years. In our main analysis, we use corruption level at year t-1, which may capture corrupt practices occurred in year t-3. In unreported analysis, we also conduct analysis using corruption in year t+1 or t. The results are similar.15 In a very few district-year observations the conviction number is missing, we use the average of the conviction rates of the prior year and following year to impute missing values. In our robustness checks, we also scale the conviction number by local government size, producing similar results.16 One limitation of the data is that it does not provide information on specific types of crimes, disallowing us to explore the heterogeneous effects of different corruption types on firms.17 Since one county FIPS code may correspond to more than two zip codes, we remove these observations and only keep the unique-match pairs.18 COMPUSTAT only maintains current corporate headquarters locations. We use a smaller sample with available information on historical locations in our robustness tests; the results reported in the internet appendix remain unchanged.19 We exclude firms located in Guam, Puerto Rico, the Virgin Islands, and the Northern Mariana Islands, which are districts out of U.S. main territories.20 We could match with our financial variables during 1990 to 2010. The sample size is, thus, somewhat reduced, as shown in Table 4.21 It equals 1−∑i=13⁡ni/N, where N is total population in a district and ni is the population of Whites, Blacks, and Asians, separately.22 In the Internet Appendix, we test the exogeneity of our instrument variables directly. We show the two instruments are largely uncorrelated with firm covariates.23 We acknowledge that political cost and expropriation cost may be correlated. One advantage of using this variable as a proxy for political cost is that it is constructed based on keywords that are related to government and political officials in general. It thus captures the overall cost when firms interact with the politicians, instead of the unique cost relating to expropriation. The data could be downloaded at: https://www.firmlevelrisk.com/. We are grateful to the authors for making the data public.24 In unreported results, we find that firms in more corrupt districts generate more sales from the government, which is consistent with this argument.Additional informationFundingThis work was supported by National Natural Science Foundation of China [grant number: 72002097; 72132004; 72272076]; the Philosophy and Social Science Planning Project of Guangdong Province, China [grant number: GD22CGL40]; Natural Science Foundation of Guangdong Province, China [grant number 2019A1515011635]; China’s Ministry of Education Humanities and Social Science Research Youth Fund Project [grant number: 20YJC630167; 22YJC630072]." @default.
• W4385664767 created "2023-08-09" @default.
• W4385664767 creator A5019704910 @default.
• W4385664767 creator A5026865760 @default.
• W4385664767 creator A5061894131 @default.
• W4385664767 date "2023-08-08" @default.
• W4385664767 modified "2023-09-25" @default.
• W4385664767 title "Political Corruption and Accounting Conservatism" @default.
• W4385664767 cites W142124823 @default.
• W4385664767 cites W1577114345 @default.
• W4385664767 cites W1962746210 @default.
• W4385664767 cites W1984706499 @default.
• W4385664767 cites W1991128340 @default.
• W4385664767 cites W1991774331 @default.
• W4385664767 cites W1995973359 @default.
• W4385664767 cites W2006306238 @default.
• W4385664767 cites W2015327645 @default.
• W4385664767 cites W2033076206 @default.
• W4385664767 cites W2041499464 @default.
• W4385664767 cites W2084647474 @default.
• W4385664767 cites W2134472446 @default.
• W4385664767 cites W2197597133 @default.
• W4385664767 cites W2234458595 @default.
• W4385664767 cites W2493789806 @default.
• W4385664767 cites W2564268022 @default.
• W4385664767 cites W2623598673 @default.
• W4385664767 cites W2765870642 @default.
• W4385664767 cites W2770414254 @default.
• W4385664767 cites W2784687599 @default.
• W4385664767 cites W2795029922 @default.
• W4385664767 cites W2883887431 @default.
• W4385664767 cites W2888803042 @default.
• W4385664767 cites W2938249760 @default.
• W4385664767 cites W3007443035 @default.
• W4385664767 cites W3023050320 @default.
• W4385664767 cites W3023308195 @default.
• W4385664767 cites W3121252168 @default.
• W4385664767 cites W3121472198 @default.
• W4385664767 cites W3121951548 @default.
• W4385664767 cites W3122194337 @default.
• W4385664767 cites W3122480927 @default.
• W4385664767 cites W3122489725 @default.
• W4385664767 cites W3122908417 @default.
• W4385664767 cites W3123215895 @default.
• W4385664767 cites W3123393350 @default.
• W4385664767 cites W3123435704 @default.
• W4385664767 cites W3123455257 @default.
• W4385664767 cites W3123475946 @default.
• W4385664767 cites W3123595534 @default.
• W4385664767 cites W3123859855 @default.
• W4385664767 cites W3123937835 @default.
• W4385664767 cites W3124096147 @default.
• W4385664767 cites W3124337013 @default.
• W4385664767 cites W3124479856 @default.
• W4385664767 cites W3124572978 @default.
• W4385664767 cites W3124683019 @default.
• W4385664767 cites W3124826409 @default.
• W4385664767 cites W3124846795 @default.
• W4385664767 cites W3124862953 @default.
• W4385664767 cites W3124972540 @default.
• W4385664767 cites W3125236353 @default.
• W4385664767 cites W3125422807 @default.
• W4385664767 cites W3125688476 @default.
• W4385664767 cites W4244668517 @default.
• W4385664767 cites W4244992121 @default.
• W4385664767 cites W4285472592 @default.
• W4385664767 doi "https://doi.org/10.1080/09638180.2023.2242399" @default.
• W4385664767 hasPublicationYear "2023" @default.
• W4385664767 type Work @default.
• W4385664767 citedByCount "0" @default.
• W4385664767 crossrefType "journal-article" @default.
• W4385664767 hasAuthorship W4385664767A5019704910 @default.
• W4385664767 hasAuthorship W4385664767A5026865760 @default.
• W4385664767 hasAuthorship W4385664767A5061894131 @default.
• W4385664767 hasConcept C121955636 @default.
• W4385664767 hasConcept C124952713 @default.
• W4385664767 hasConcept C142362112 @default.
• W4385664767 hasConcept C144133560 @default.
• W4385664767 hasConcept C17744445 @default.
• W4385664767 hasConcept C199539241 @default.
• W4385664767 hasConcept C2780027415 @default.
• W4385664767 hasConcept C2781426361 @default.
• W4385664767 hasConcept C4577558 @default.
• W4385664767 hasConcept C94625758 @default.
• W4385664767 hasConcept C96640997 @default.
• W4385664767 hasConceptScore W4385664767C121955636 @default.
• W4385664767 hasConceptScore W4385664767C124952713 @default.
• W4385664767 hasConceptScore W4385664767C142362112 @default.
• W4385664767 hasConceptScore W4385664767C144133560 @default.
• W4385664767 hasConceptScore W4385664767C17744445 @default.
• W4385664767 hasConceptScore W4385664767C199539241 @default.
• W4385664767 hasConceptScore W4385664767C2780027415 @default.
• W4385664767 hasConceptScore W4385664767C2781426361 @default.
• W4385664767 hasConceptScore W4385664767C4577558 @default.
• W4385664767 hasConceptScore W4385664767C94625758 @default.
• W4385664767 hasConceptScore W4385664767C96640997 @default.
• W4385664767 hasFunder F4320321001 @default.
• W4385664767 hasFunder F4320321921 @default.

• next