This paper considers what we argue was the first experiment of an exchangerate band. This experiment took place in Austria-Hungary between 1896 and1914. The rationale for introducing this policy rested on precisely thoseintuitions that modern target zone literature has recently emphasized: theband was designed to secure both exchange rate stability and monetarypolicy autonomy. However, unlike more recent experiences, such as theERM, this policy was not undermined by credibility problems. In other wordsthe episode provides us with an ideal testing ground for some importantideas in modern macroeconomics: specifically, can formal rules, whenfaithfully adhered to, provide policy makers with some advantages such asshort term flexibility? First, we find that a credible band has amicroeconomic influence on exchange rate stability. By reducinguncertainty, a credible fluctuation band improves the quality of expectations,a channel that has been neglected in the modern literature. Second, weshow that the standard test of the basic target zone model is flawed anddevelop an alternative methodology. This enables us to understand whyAustro-Hungarian policy makers were so upbeat about the merits ofexchange rate target zones. We believe that these findings shed a new lighton the economics of exchange rate bands.