Abstract The paper aims to present for the first time the methodology for building the stress-testing at the Bank of Albania, as well as to evaluate the quality of its forecasts. Through the forward-looking stress-testing analysis the financial system stability and capital adequacy in the banking sector are estimated for a period of up to two years. Regardless of the purpose of stress-testing to assess whether in the event of large losses the banking sector has the ability to absorb them and not to accurately predict the indicators of the banking sector, it is still important to assess whether the attitude toward risk is sufficiently conservative. The results suggest that the forecast of the capital adequacy ratio is quite close to its actual values. However, disintegrating these developments by the contribution that comes from the underestimation of the regulatory capital, in order to preserve the conservative trend of the exercise, has eased the underestimation of the risk-weighted assets, mainly reflecting the changes in the regulatory framework of the Bank of Albania. Following the results of the analysis, the paper proposes several ways to further improve the quality of forecasting. They relate mainly to the transition towards a dynamic stress test forecasting, the consolidation of the conservative trend for the forecasting of the regulatory capital and a preliminary assessment of the regulatory changes and their inclusion in the stress test.