This study exploits two institutional features of China to test the causal link between tax and capital structure. First, the central government exclusively determines the corporate tax rate in China, which results in 2,316 tax increases and 2,189 tax decreases affecting different firms in various years during the period of 2000 to 2011. Such mandatory tax shifts provide a quasinatural experimental setting for our difference-in-differences analysis investigating the impact of tax on leverage. We find evidence supporting the dynamic trade-off theory, namely that firms are unresponsive to tax cuts but increase long-term leverage when taxes rise (particularly those in low statutory tax regimes). Second, governmental intervention in capital allocation is common in China such that political connections are usually regarded as an asset for firms in accessing bank loans. Using anti-corruption events as shocks to the value of political connections over the sample period, our research is the first study to show that political connections become a liability that enables banks to recall loans from affected firms during the anti-corruption campaign periods. This change overturns the typical tax-leverage relationship observed, as we find anticorruption affected firms reduce long-term leverage when taxes are cut and they become insensitive to tax increases. Our results have implications not only for how firms have adjusted their capital structure to tax changes in the past, but also for what new feature of political ties that policy makers/regulators ought to consider in monitoring bank loan markets.