This paper examines the appropriation of audit fees for order-taking industries on discretionary accruals. After the ‘accounting cliff’ incident of the so-called Daewoo Shipbuilding & Marine Engineering (DSME) in 2015, the issue of accounting transparency of the order-taking industry has once again attracted attention. Daewoo Shipbuilding & Marine Engineering is an industry that produces low-demand, high-priced products, including construction, shipbuilding, computer programming and systems integration. The order-taking industry is characterized by a long- term industry, from product completion to delivery. The percentage-of-completion method is based on the completion of construction contracts, in calculating the revenue for subcontracting works such as bridges, dams, roads, etc. It is a revenue recognition method by considering the construction progress ratio between each profit and loss account and explaining the actual costs incurred in response to the construction revenues. However, recently controversial accounting disputes of the order-taking industry have been continuing, due to the accounting process based on this construction progress criterion. In other words, it can reduces accounting transparency by arbitrarily reducing the total redetermined costs and carrying out high construction progress ratio, as in the case of Daewoo Shipbuilding & Marine Engineering. However, companies can estimate the construction costs when the construction progress rate is released. As the results, the industrial competitiveness will be degraded. Accordingly, authorities and related organizations have proposed various measures and regulations in order to prevent mismanagement of accounts receivable and accounting operations. In 2016, the Korea Certified Public Accountant Society firstly introduced “Key Audit Matter, KAM” for Financial Statements of the Order-taking Industries established various practical guidelines, to establish accounting transparency. In 2017, the regulators announce the guideline to broaden the scope of sanctions for auditors that neglected such fraudulent accounting. On the other hand, the auditing is carried out through rigorous criteria for the order-taking industry. It is not possible to erase the accountability for external accounting if the auditors have been faced the lower audit fees and audit efforts. That is, in the accounting industry, it is argued that audit quality may be deteriorated due to low audit fees(Lee and Lee 2004; Rho and Bae 2004; Kim and Ko 2017). In other words, due to the uncertainty of the long-term construction progress estimate and lack of disclosure on the change in the construction progress ratio, audit quality is deteriorating(Shin 2005; Kang et al. 2011; Sohn et al 2014; Hyun et al. 2016). The main factor is low audit fees which cannot compensate the high litigation risks(Pratt and Stice 1994; Gul et al. 1998; Abbott et al. 2006; Lee and Kim 2017). In addition, the level of increase in audit fees for listed companies in Korea has been lowered compared to the increase in audit time since the adoption of K-IFRS in 2011. It is also resulted in deterioration of auditing quality. However, there is a lack of empirical research on the audit quality of the order-taking industry, which has raised many controversies. Therefore, it is meaningful to examine the relationship between the audit fees and the discretionary accruals in order-taking industries. In this study, we examine the effect of audit fees and abnormal audit fees on the absolute value of discretionary accruals in the order-taking industries such as constructing industry, shipbuilding industry, computer programming and system integration industry. The results of this study, which analyzed the effect of audit fees and abnormal audit fees on the quality of financial information by dividing total samples into order-taking and non-order- taking industries from 2009 to 2015, showed that audit fees are a negatively correlated with the absolute value of discretionary accruals for order-taking industries. But, there are normally negative correlation between audit fees and abnormal audit fees and discretionary accruals for non-order-taking industries, but not significant. This implies that the increase of audit fees in the order-taking industry have higher the accounting credibility, whereas the increase in audit fees can not affect the discretionary accruals of non-order-taking industries. It implies that the incentives of managers to manage their earnings for order-taking firms is higher than for non-order-taking firms due to the uncertain progress rate. Thus, the higher audit quality is required in the order-taking industries due to the unclear standard. In addition, these results indirectly show the differential application of audit fees according to industrial characteristics. The contribution of this study is as follows. First, the relationship between audit fees and abnormal audit fees and discretionary accruals according to industry characteristics was demonstrated through empirical analysis. This study presents the empirical analysis on the effect of audit fees on the quality of financial information by dividing the industry into the order-taking industry and the non-order-taking industry. Second, the results of this paper are systematically meaningful in that the audit fees can differently affect the audit qualities by the industries. Currently, the controversial accounting of the construction and shipbuilding industries, which are the typical order-taking industries, continues to arise. Therefore, this study indirectly suggests ways to improve the accounting credibility in these industries. It also provides implications for the authorities and firms on the appropriate level of audit fees for these controversial industries.