This paper presents some conventional and new measures of market, credit, and liquidity risks for government bonds. These measures are analyzed from the perspective of a sovereign's debt manager. In particular, it examines duration, convexity, M-square, skewness, kurtosis, and VaR statistics as measures of interest rate exposure; a VaR statistic as the prominent measure of exchange rate exposure; the balance sheet approach (or contingent claims approach), and its consequent probability of default as the most promising measure of credit risk exposure; and an elasticity approach and a VaR statistic to measure liquidity risk. Along with the formulas for the various statistics proposed, we provide simple examples of their application to some common risk valuation cases. Finally, we present an integrated approach for the simultaneous estimation of a portfolio's interest rate and exchange rate risk using the VaR methodology. The integrated approach is then extended to also include N risk factors. This approach allows us to measure the total risk of a portfolio, provided that the volatilities and correlations among the risk factors can be estimated.