This paper presents the main issues involved in estimating potential output. Theobjective is to describe the alternative methods and analyze their application andimplications for growth forecasts and macroeconomic policy in Brazil. The textemphasizes the determinants of potential output under fixed and flexible coefficientsof production. Given the wide use of aggregate measures of Total Factor Productivityin growth accounting, and the sensitivity of such a variable to economic assumptionsand errors of measurement, the text also presents the main applied critiques andalternatives to aggregate growth-accounting exercises. The main conclusions are: i)the annual potential growth rate of Brazil?s Gross Domestic Product (GDP) variessubstantially depending on the method and hypotheses adopted and, what is mostimportant, potential GDP is not separable from effective GDP in the long-run; ii)growth-accounting and time-series studies of Brazil result in low potential-outputgrowth rates because they extrapolate the slow growth of 1981-2003 to the future;iii) capital seems to be the main constraint on growth in Brazil and, therefore, ademand-led increase in investment can raise both its effective and potential outputlevels; iv) however, because of the slow adjustment of the capital stock, an investmentboom can also hit a supply constraint before the stock of capital has time to adjust tothe growth rate of investment; and v) aggregate measures of potential output do notcarry much information about the economy and, therefore, they should becomplemented by sectoral estimates of capacity utilization to identify the bottlenecksin inter-industry flows and the corresponding demand pressures on inflation.