>Assessing productivity on a farm or in a factory is relatively easy.
>The output is easily measured: so many bushels of wheat, so many cars
>off the assembly line.
I have no idea what the prevailing standard for measuring productivity among mainstream economists is, but doesn't this definition above seem rather physiocratic ? What matters is not physical output per worker-hour, nor even physical output per worker-hour per hourly rate of pay, but market price of good/service per worker-hour per hourly rate of pay. That is, productivity measures need a universal equivalent for all inputs and outputs in all sectors, and that universal equivalent is exchange value, not incommensurable physical outputs. (Moreover, we don't or can't know what the exchange value of a commodity is until it actually sells in the marketplace, however efficiently it is produced -- "the contradictory unity of production and realization of value").
>Nor is it clear that the
>Internet explosion has reached the critical mass to make it a really
>significant factor in the recent spectacular growth of productivity;
>e-commerce, totaling an estimated $150 billion in 1999, is still puny
>in the $9 trillion American economy.
I don't have the time or energy to figure out how the following amendment does or might affect Roach's overall argument, but data for revenues from e-commerce (which includes on-line retailing) is hardly a reliable stand-in for data on purchase and implementation of IT capital goods.
So are we to conclude from this article that the long expansion, especially in the late autumn of its life, is largely an artifact of highly-leveraged speculation in NASDAQ-type stocks (itself connected to post-97 and 98 capital flight from E. Asia) ? How come one sees so little in the left-liberal, mainstream, and business press about domestic job _loss_ as a consequence of the e-commerce boom (retail workers in the non-niche "bricks and mortar" world), w/attendant shifts in wage structure, surplus value distribution, etc.
John Gulick