The rapid evolution of financial innovations has provided many opportunities for profit and in other side, more risk exposures for banks. Famous financial disasters in 1990s led bank supervisors to develop extensive risk disclosures requirements. From a regulatory perspective, preventing disruptions in financial systems due to crisis from overall economy posing a wider systematic risk is a rationale for government intervention in banking. The Basel Committee in the Bank for International Settlements modified the 1988 accord in 1996. There are two major changes in the 1996 version of Basel accord (its Taiwanese version is issued in 1998 which we refer to as the new regime) to modify the 1988 old version (its Taiwanese version is issued in 1992 which we refer to as the old regime). First, the new version requires banks to measure their market risk exposures. Second, specific loss reserves do not count as part of Tier 2 capital. Previous research shows that banks have incentives to manage accounting numbers for regulatory capital or earnings. Loan loss provisions and write-offs are relatively lager discretionary accruals for banks and have significant impact on banks' regulatory capital or earnings. This paper investigates the 1998 changes in capital adequacy regulations to construct more powerful tests of capital and earnings management effect on loan loss provisions or write-offs. The contribution of this paper lies in the examination of whether the changes in regulations caused changes in bank managers' estimate of the loan loss provisions and write-offs. This paper expects that the 1998 regulatory changes create (remove) incentives to use loan loss provisions for capital management (earnings management). Second, this paper expects these changes remove incentives to use write-offs for capital management. Overall, these changes substantially alter banks' incentives to manage capital via loan loss provisions/write-offs. After controlling the impact of the decrease in the sales tax rate for financial institutions in 1999, the main empirical findings are as follows. (1) These regulatory changes create stronger incentives to increase loan loss provisions for capital management under the new regime. (2) There is no evidence of earnings smoothing behavior via loan loss provisions under the new regime. (4) These changes remove incentives to depress write-offs before 1998. (5) Within the new regime, the loan loss provisions are larger for banks with higher specific loss reserves than those with lower specific loss reserves.