Republican advocates claim that their party is interested in growth, while the Democrats are only interested in redistribution. Well, the Republican Party may be interested in growth, but they are remarkably inept at delivering growth. From 1948 until 2008, there were 15 administrations.
Those with the highest rates of growth were all Democratic.
Those with the lowest rates of growth were all Republican.
The Democratic presidential terms all had higher growth than the Republican average.
The Republican presidential terms all had lower growth than the Democratic average.
The growth is measured by the change in Real Gross Domestic Product (Real GDP or RGDP) over four years. There are a variety of starting/ending points that might be taken, and the one you choose has some influence on the exact numbers, and even on the rankings. No reasonable choice, however, negates any of the four statements above.
I've done the calculations for three periods so that you can see these claims verified in particular numbers.
1) From inaugural quarter to inaugural quarter This shows most precisely the period of actual presidential authority. It has the weakness of being a quarter, and of being "annualized." That means that it is the production during that quarter multiplied by some factor (near 4, but not necessarily precisely 4) to represent the annual RGDP. The factor, however, remains constant from one administration to the next, so the ratio is the ratio of actual product in those quarters.
2) From election year to election year. This gives annual GDP, which is more stable and more reliable than the quarterly data. Unfortunately, it credits -- or blames -- the incoming administration with some economic activity before inauguration day.
3) With one year lag. This measures the change between the year the president is inaugurated with the year that his successor is inaugurated. It measures the named president's administration's effectiveness only in so far as the performance during the first year of an administration is credited with (or blamed on) the preceding administration.
Clearly, none of these is perfect. Even if we had measures of any period we wished -- and our sources do not provide anything less coarse than a quarter -- some of an administrations policies may be effective at some point, and others not until later. For example, the tax policies of an administration must be passed through Congress, and then will be implemented in the following year. (Much of the theoretical writing about tax policies, however, argues that behavior is based on what people expect their taxes to be.)
The consistency of the results, however, implies that if no period perfectly measures the influence of an administration, nonetheless the influence of party (of the administration) is real.