Conventional wisdom holds that in order to evade electoral punishment governments obfuscate welfare state retrenchment. However, governments do not uniformly lose votes in elections after they cut back on welfare benefits or services. Recent evidence indicates that some of these unpopular reforms are in fact vote-winners for the government. Our study of eight Danish labor marked related reforms uses insights from experimental framing studies to evaluate the impact of welfare state retrenchment on government popularity. We hypothesize that communicating retrenchment is a better strategy than obfuscating retrenchment measures. In addition, we hypothesize that the opposition’s choice between arguing against the retrenchment measure, or staying silent on the issue, affects the government’s popularity. Thus, the study presents a novel theoretical model of the popularity effects of welfare state retrenchment. In order to evaluate our propositions, we move beyond the standard measure in the literature and use monthly opinion polls to reduce the number of other factors that might affect government popularity. We demonstrate that governments can evade popular punishment by communication. They can even gain popularity if the opposition chooses not to attack. On the other hand, government popularity declines if the government obfuscates – and the decline is even larger if the opposition chooses to attack.