Technically, this isn't quite true, assuming consumers can have access to the same credit markets as the government (a reasonable assumption). In fact, consumers are indifferent between policies: they are no better off if taxed now for spending today, or taxed tomorrow for spending today (government borrowing).
This indifference condition is a fundamental premise to contracyclical government stimulus.
But your point is spot on: too much spending, whether financed by taxation or borrowing, is a bad thing. The difficulty is in quantifying "too much". I actually wrote a paper on this topic, where I proposed using election outcomes as a proxy for consumer willingness to pay. Rational consumers of government spending will stop consuming at the point where the marginal gains are maximized. I think we can look at election outcomes over changes in government expenditures and/or effective tax rates as a proxy for this; consumers express a desire for more government consumption by electing progressives, and less by electing conservatives.
Since the end of WW2, the effective tax rate has been approximately 18% of GDP. Efforts to substantially increase it have been political failures, and efforts to substantially decrease it have been political failures (resulting in the election of the opposition party, whether the effort to change spending came through taxes or borrowing doesn't really matter). The current dire straights of the Democratic party are largely due, in my opinion, to the present vast disequilibrium: spending is too far out of whack with the optimal rate. Ironically, this was a big part of the Republican party's problems in 2006 and 2008; I think Democrats simply misunderstood voter intent. Spending has been slowly outpacing revenues for years, but voters were relatively tolerant of the phenomenon because of the combined recession, foreign wars, and slow rate. They've now reached a breaking point.