(Bloomberg View) —Perhaps the most common policy argument I see is that the government ought to channel more resources to lower-income Americans, and finance this transfer by raising taxes on the wealthy. This argument is seductive, because so often it seems that spending by the wealthy just isn’t that important — did you read about the recent luxury handbag auction at Christie’s? Yet the transfer argument rests on flawed premises. Namely, it doesn’t compare the proposed policies — such as a boost to food stamps, disability payments or lead abatement — to the best possible alternatives.

To see the problem, let’s say we ranked a bunch of general alternatives from best to worst:

More money for lead abatement More food stamps More money for disability insurance More money left in the hands of the wealthyYou might not agree with those rankings, because perhaps the rich people will use their money for investment and innovation. Still, this example certainly shouldn’t prejudice the analysis against redistribution.

Now, let’s say we are considering more money for disability insurance. It would be easy enough to claim the disability payments are better than more Birkin bags for the wealthy. But we also need to compare the disability program to the best possible alternative; under these assumptions,  that’s lead abatement. Viewed in that light, the increase in disability insurance no longer looks so good.

I would suggest that if you read an analysis comparing only No. 3 to No. 4, and without considering what broader ranking might exist, the writer is misleading you (and probably himself or herself).

When we make personal decisions, we usually compare a choice to the best possible alternative, not the worst. Imagine if you suggested to your spouse that you go out to the movies, and your spouse asked why that might be a good idea. It wouldn’t be much of an answer to say that the movie is better than the very worst show on television at home. Rather you should focus on comparing the movie to the next best thing you might do, like watching your favorite TV show or going to a new restaurant you want to try.

The upshot is that we should compare anti-poverty programs to other anti-poverty programs, and favor only the prioritized ones. But just how much of a priority does a program need to be?

One way to proceed is to ask: If we expand some programs, what is the most likely political response? It could be either lower spending in some other program or, in fact, raising taxes on the wealthy. But the evidence on the “fiscal gap” — the space between what the government owes and what it collects — suggests that the opportunity cost of expanding one transfer program is likely some government spending elsewhere, rather than expensive handbags for the wealthy.

If we look at the overall fiscal position of the U.S. federal government, we are spending beyond our means and the future will require some mix of spending cuts and tax increases. According to a report from the Government Accountability Office: “To close the gap solely by raising revenues would require a revenue increase of about 33 percent, and maintaining that level of revenue, on average, each year over the next 75 years.” I would submit that revenue increases of such magnitude are unlikely or perhaps even impossible, and thus any new spending will have to come out of other government spending. In other words, for better or worse, we’ve already committed to spending that tax increase on the wealthy that you were planning to use for other purposes.

The problems run deeper yet. You might want to boost disability payments, and in exchange cut spending on the very worst government program, whatever you think that might be. But the more realistic approach is to compare your proposed expenditure to spending on the most politically vulnerable programs — those are the ones most likely to be cut in response. Quite possibly those are the transfer programs Trump wishes to cut today, the ones that everyone is so scurrying to defend.

Furthermore, the keener you are to expand lots of government programs, the more likely the opportunity cost of those expansions is high. It’s actually the moderate fiscal conservative, and not the left-leaning progressive, who should believe most firmly in any single program expansion.

To be sure, there is no particular reason to believe that the current mix of government spending is optimized, and if you can see an improvement, go for it. But if you are comparing your favorite social programs to the frivolous consumption of the wealthy, that social dividend has already been spent, and politics will make some other sacrifice. The times you feel good and righteous about your policy advocacy are exactly when you are making the argument look too easy.

This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.