In the standard representation of technology, productivity is built into the production function as a given relationship between inputs and output. This representation needs an equilibrium framework, in which the ratios between the factors and output are constant and corresponding to those dictated by the production function coefficients. In this framework, technological advances should be instantaneously mapped into increases in productivity, and the only way to explain the ‘productivity paradox’ is to assume adoption delays. We propose a different approach, by which productivity is the outcome of an out-of-equilibrium process triggered by a technological shock. The potential gains of a superior technology may only be appropriated if agents succeed in reshaping the productive capacity (whose distinguishing feature is to be temporally articulated), and in recovering the intertemporal coordination disrupted by the introduction of the new technique. Physical, human, and financial capital are complementary in this process of reshaping, and may constrain each other. The outcome of the disequilibrium process depends then on the interaction of accumulation choices, learning, and money supply rules. We argue that the different performances of the US and Europe in the last two decades may be explained along these lines.