Graph 1 shows a conventional PPP measure of the real exchange rate in Colombia and an index of the relative price of tradable goods in terms of non tradeable. The common conclusion is that, since 1990-92 there has been a medium term process by which the RER has appreciated by as much as 26%. The analysis of this process is the first objetive of this paper. Graph 2 shows the evolution of the nominal exchange rate during the same time period, measured as the percentage deviation from the exchange rate band's central parity. The conclusion there is the presence of an important degree of volatility in the latter part of the sample. The analysis of this aspect is the second objective of the paper. In abstract terms, the case for a nominal exchange rate band is based upon the argument that the regime allows a combination wherein the espectational effects associated with anchoring are obtained, without sacrificing the requiered degree of flexibility. At its best, a band system can obtain these effects, anchoring expectations whilw absorbing (and even smoothing) external schoks and thus achieve all of what one wants from an exchange regime. At its worst, the band operates much like a fixed parity under increasing pressure. The argument of this paper is that the recent experience of Colombia offers interesting evidence in favor o f the idea that the exchange rate band is very helpful in anchoring substantial swings of expectations. First, economic fundamentals are the main driving force of the real appreciation. Second, the devaluation which occured in mid 1995 was limited, in size , by the presence of the band. Moreover, it would have proven extremely difficult to handle had the system been a crawling peg, given its discretinary features. Thrid, the nominal appreciation which occured in 1996 is a return to fundamentals which, once again, might have been more difficult under a managed regime.