This paper examines whether various hypotheses put forward to explain the downward trends in government capital spending are supported by the data. Using panel data for 22 OECD countries for 1980-1992, various hypotheses are tested in a model. The authors find support for three hypotheses: (1) capital spending is reduced during periods of fiscal stringency, since this category of government spending is politically an easier target for cuts than other spending categories; (2) myopic governments will cut investment spending more than governments which have a longer policy horizon; (3) private investment influences government investment spending, because both types of investment are complementary.