Gotta Take a Dollar to Give a Dollar

Tom Sgouros has penned another missive explaining why Rhode Island's fiscal conservatism has spelled its doom, and why more progressive spending and a bigger state government is the solution. I know. I know. Such are the indications that, though we all breathe the same air, we live in the different realities of our preconceptions.

One side's obvious conclusions are the other side's insidious illusions. My first reaction is to see in Tom's rhetoric and in his sources a deliberate deception founded in financial and ideological motivation. His sympathizers will see in my response a desperate attempt to spin away the incontrovertible at the behest of the puppet masters who control my financial well-being (and at whose suggestion I suffer through those staged photo shoots on construction sites).

I'll give Tom this: He bases his argument on a persuasive table, which Economy.com's Mark Zandi has gone so far as to present to Congress (PDF):

My understanding of this data — although specifics aren't easily available — is that it derives from a macroeconomic model into which Zandi plugged the various instances of government spending. If that's correct, then the exercise is fatally skewed based on the simple fact that the government must take a dollar in order to give a dollar.

Actually, the government must take more than a dollar in order to give a dollar. In 2004, for example, the Food Stamp Program spent about $0.20 for every dollar that it gave away (PDF). That's not the whole story, of course, because there were expenses associated with collecting, allocating, and processing the program's budget. According to Charity Navigator, 9 out of 10 charities spend no more than $0.54 per dollar given away on administrative costs, and 7 out of 10 spend no more than $0.33. For the sake of consideration, then, let's assume that it costs the federal government no more than another five cents per dollar handed out to get that money from the taxpayer to the Food Stamp Program, so the total cost per food stamp dollar would be $0.25.

That means that, for every food stamp dollar given, the government must take $1.25 out of the economy at some other point. According to Zandi, an across-the-board tax cut would add $1.03 to the next year's GDP, so it follows that taking a dollar costs $1.03. Based on an across-the-board increase in taxation, the cost of every food stamp dollar is therefore $1.29, making the actual amount that the whole process adds to the next year's GDP only $0.44. Inasmuch as it does not cost any money not to take a dollar from somebody, a broad tax cut would actually add $0.59 more to GDP than would the food stamp.

The intuitive sense, here, is that it doesn't make any difference, economically, whether I spend a cash dollar on groceries or a welfare recipient spends a food stamp dollar on the same items. If the government largess is extracted from everybody, then the working poor and middle class don't have those dollars to spend. In fact, it appears that funding food stamps by taking a dollar from a family that must therefore reduce its grocery bill in response winds up costing the GDP $0.43.

The appeal of Sgouros's argument comes in the fact that those who don't have to spend will tend to save; those who take in more than they could possibly spend will save even more. Indeed, looking at the numbers on the table, it's tempting to observe that a one-dollar increase in the corporate tax rate would appear to cost the GDP only $0.30, so reprocessing that dollar into food stamps would provide a net GDP gain of $1.43. That possibility is an illusion for two reasons. The first is that the $0.70 difference must come from some theoretical savings account (or untapped credit); thought through in reverse, the reason a dollar of corporate tax cuts only results in a GDP gain of $0.30 is that the rest goes somewhere unproductive. If that tax rate becomes confiscatory in order to alleviate the tax burden on the poor and middle class, corporations and the proverbial rich will not for long watch their reserves being depleted without reacting.

The second reason is that, when heavily taxing the rich, a marker of single dollars no longer applies. Wealthy entities (whether families or organizations) work in different amounts than do the the rest of us. It's true that a rich man may be less productive with a free dollar than would a poor man, but take from him ten million dollars, and he won't invest in a company, donate to charity, build a house, and so on, and those activities all filter down to folks who'll spend their money in the same fashion as welfare recipients, but without the government processing fee. If a few percent of the society is going to pay most of the cost of government, the taxation dollar amounts are exponential.

To be sure, Sgouros whistles an enticing tune when he writes:

People routinely misunderstand the important points of policy that stem from Keynes's findings. Government spending and progressive taxation aren't good things because they support government workers or "punish" rich people. They are good things because they are how a government can help the economy grow. (Up to a point, of course, a detail Keynes made clear.) Government workers with money to spend will spend it, and that drives the economy. Progressive taxation keeps more money in the hands of the poor and people in the middle, both of whom are more likely to spend their income than rich people are.

But he loses the thread of his earlier wisdom. Private-sector workers funded via mutual agreement will also spend their money, but they have more reason to earn it efficiently than public-sector workers funded via compulsory taxation. They also can't hide their wealth as thoroughly. Tom notes that a "dollar saved is not a dollar invested" and that "people have different preferences for how they hold their money," but he doesn't acknowledge that unionized government workers save their money in the form of perks, accumulating benefits, and defined-benefit pensions. There must be some form of savings — actual or theoretical against future taxation and revenue — in order for somebody in his '40s or '50s to retire for the rest of his life. A person who lives for thirty years on an annual pension of $35,000 had stored away over a million dollars in some nook of the system.

For its larger projects and continued growth, a society needs people with financial reserves, but those reserves have to be accessible for spending. A society also needs people motivated to create, and capable of creating, wealth, and those (as I've argued here, here, and here) are the families that Rhode Island has been chasing out. Sgouros and Co. can flash all the statistical pictures they want, but what Rhode Island must do is to maintain the number of wealthy taxpayers while relieving the burden of (and thereby attracting) the productive class.

Ultimately, that leaves only one broad category to which to turn to balance the state's budget: Those who represent a net cost to the government.