> Greenspan, Kudlow, & others argue that it's wrong not to count
> software expenditures as depreciable investment rather than an
> expensed current cost, and that NIPA figures underestimate investment
> and growth. If investment is understated than profitability is
> overstated, even though Greenspan says it's probably understated, but
> let's forget about that for now. My impression is that if software is
> a capital expenditure - and why not, conceptually? - it depreciates
> very rapidly, at least by my personal experience. Maybe I upgrade too
> much. Any other thoughts?
I'd say a lot depends on what kind of software you're talking about. As you say, some software depreciates very rapidly - maybe not as rapidly in large corporates (where version control acts a brake) as in small offices, but pretty rapidly none the less.
'Structural software' is, however, I think another case entirely. By 'structural software', I mean software which is core to business process - e.g. systems like SAP R3. 'Big software', like 'big computers', generally works on a different business model to 'small software' like office suites. It is more analagous to a factory, than to a pen.
Any guesses on how investment in software breaks down into 'big software' vs. 'small software'? Figures I've read in trade journals on the amount of expenditure on items such as 'data warehouses' are quite stunning.
Peter -- Peter van Heusden : pvanheus at hgmp.mrc.ac.uk : PGP key available 'The demand to give up illusions about the existing state of affairs is the demand to give up a state of affairs which needs illusions.' - Karl Marx