Against the backdrop of growing institutional interest in digital assets, this study examines the implications of incorporating Bitcoin into a diversified multi-asset portfolio within a dynamic asset allocation framework that extends beyond static optimization. Rather than treating asset classes as constant components, the study explores how the integration of cryptocurrency interacts with traditional instruments such as equities, fixed-income securities, and commodities under time-varying market conditions using monthly excess return data spanning from January 2014 to October 2025. The study evaluates whether the return-risk profile of Bitcoin can improve portfolio efficiency and robustness under institutional investment constraints and whether dynamic asset allocation strategies offer advantages over their static counterparts in a Korean institutional investment context. To achieve a more stable and practical allocation, this research utilizes the Black–Litterman model as the primary optimization tool, specifically designed to address the limitations of traditional mean–variance optimization, such as the "corner solution" problem and the extreme sensitivity of portfolio weights to minor changes in input data. By incorporating market equilibrium and specific investor views into the optimization process, the Black–Litterman model produces portfolio allocations that are more stable and economically interpretable. This approach is particularly relevant when dealing with Bitcoin, as its high volatility could otherwise lead to impractical and overly concentrated positions. The model is implemented under realistic investment constraints, including the prohibition of short-selling and borrowing, to reflect the practical conditions faced by domestic institutional fund managers. The framework mitigates estimation error by reducing extreme portfolio weights, thereby improving the robustness of the resulting allocations and providing a more stable basis for investment decisions. This feature is particularly important in the context of digital assets, as the high volatility and non-normal return characteristics of Bitcoin may otherwise lead to unstable portfolio allocations and excessive concentration in optimization-based strategies. At the same time, this study incorporates the time-varying nature of financial markets by employing a Markov Regime-Switching Model (MSM). Unlike static models that assume constant return distributions, the MSM allows expected returns, volatilities, and correlations to vary across different market regimes, thereby capturing time-varying market conditions more effectively. The model identifies market regimes based on three distinct criteria: macroeconomic business cycles derived from the Composite Leading Index (CLI) published by Statistics Korea, equity market volatility measured by the V-KOSPI200, and the internal return dynamics of Bitcoin itself, capturing both macroeconomic and cryptocurrency-specific cycles. This framework enables regime-dependent portfolio adjustments as market conditions evolve over time. Such an approach is particularly relevant in financial markets characterized by structural breaks, asymmetric volatility, and rapidly changing cross-asset correlations, where static allocation strategies may fail to respond effectively to shifts in market conditions. The empirical findings suggest that the proposed framework enhances risk-adjusted portfolio performance. Even with relatively small allocations to Bitcoin, portfolios incorporating the digital asset exhibit higher Sharpe, Sortino, and Omega ratios compared to those composed solely of traditional assets, indicating enhanced portfolio efficiency across multiple performance dimensions. The improvement is most pronounced in downside-risksensitive measures such as the Sortino and Omega ratios, suggesting that Bitcoin inclusion improves downside-risk-adjusted performance. This enhanced efficiency stems not only from Bitcoin's high return potential but also from its diversification benefits, as evidenced by improvements in the Diversification Ratio (DR). Bitcoin maintains relatively low correlations with traditional asset classes, particularly sovereign bonds and commodities, highlighting the importance of accounting for cross-asset interactions when evaluating portfolio performance. While Bitcoin inclusion tends to increase tail-risk measures such as Maximum Drawdown (MDD) and Conditional Value-at-Risk (CVaR) under static allocation, the dynamic MSM framework helps mitigate these risks by tactically reducing Bitcoin exposure during high-volatility or bear-market regimes, including major Crypto Winter episodes, while maintaining exposure during recovery periods. The benefit of Bitcoin inclusion is not uniform across all regimes: during high-volatility or bear-market phases, dynamic strategies help preserve capital by reducing exposure, whereas maintaining Bitcoin exposure during low-volatility or recovery phases contributes to improved risk-adjusted returns. These regime-contingent results underscore the importance of treating Bitcoin not as a static allocation but as a dynamically managed component whose optimal weight varies with market conditions. Notably, even conservative investors with a high level of risk aversion (λ=10) can benefit from allocating as little as 1–2% of their portfolio to Bitcoin, as such a modest inclusion is sufficient to improve risk-adjusted performance. This finding has direct practical implications for institutional fund managers who face strict investment constraints yet seek incremental portfolio efficiency gains. Robustness checks using alternative estimation windows, rebalancing frequencies, and risk-aversion levels produce qualitatively similar results, suggesting that the findings are not overly sensitive to model specifications. During periods of significant market downturn, including major Crypto Winter episodes, dynamic strategies demonstrate lower cumulative losses compared to static allocation approaches. Overall, the results suggest that the observed performance improvement is driven not solely by higher returns but also by a dynamic risk adjustment process that responds to regime shifts. The role of Bitcoin in a portfolio is therefore not constant but varies with market conditions, and its contribution is closely linked to regime-dependent interactions with other asset classes. From a policy perspective, the results highlight the importance of establishing a clearer regulatory and institutional framework for digital asset investment in Korea, including standardized legal definitions, accounting treatment, and capital market guidelines, to support the responsible integration of Bitcoin into institutional portfolios. However, these findings should be interpreted within the context of the sample period and model specifications employed in this study, and caution is warranted when extrapolating the results to different market environments. Taken together, this study provides an empirically grounded framework for understanding the role of digital assets in dynamic portfolio management.