Three Quick Notes on the Basic Income

Wonkblog had a forum yesterday about the basic income. I did not go, but I did follow along on Twitter. In the course of following along, I noticed three lines of discussion that are common in these discussions, but also irritatingly confused.

Against the Basic Income or a Specific Amount?

A basic income is just a flat dollar amount that gets paid out to each person. It does not imply any particular dollar amount. The dollar amount could be $1 or it could be $20,000. So if we are trying to have a clear debate about installing a basic income policy, we need to distinguish between arguments against a basic income itself or arguments against having a basic income that is too high.

So if someone says, for instance, that a basic income of $12,000 per person each year would have all of these negative consequences, they have not yet offered an argument against having a basic income. They have only offered an argument against having a $12,000 basic income. This tells us nothing about their views on, say, a $3,000 basic income as I have proposed.

Do not allow arguments against high basic incomes pass as arguments against all basic incomes. They are totally different points.

The Fiscal Cost of a Negative Income Tax

A negative income tax is when the government taxes you a negative amount for a specific increment of income. So for instance, if the first $5,000 you received was taxed at negative 10%, you would receive $500 from the negative income tax for a grand total of $5,500. Generally, a negative income tax is phased out somewhat quickly, and because of this phase out, it is deemed by those who look at budget spreadsheets without contemplating their deeper meaning as having a lower fiscal cost than a basic income. But this is confused.

The Earned Income Tax Credit, our country's existing negative income tax scheme, helps make my point.

This graph shows how much EITC benefit you receive at a given income level. It also perfectly encapsulates the conceptual way those who say it has a lower "fiscal cost" think about the program. What you see in the graph is people at each income level getting a specific benefit amount with those above a certain income amount getting no benefit. Since those above the amount get no benefit (and there are those phase in and phase out periods where people get lower benefits), it would seem like this does have a lower fiscal cost, right?

Wrong. When you march through the phase out period, you can think of that as a benefit reduction, or you can think of it as you marching through a much higher effective marginal tax rate.

You can see the effective marginal tax rate phenomenon in these graphs. The tax rates are negative for a while, and they shoot up during the EITC phase out period. When the phase out period ends, they fall back down (hence the big dip you see in each of them).

In reality then, what's going on is not that the EITC has lower costs. It is that there is a phase out period where taxes are rapidly jacked up. Everyone gets the EITC. It is just that some have incomes that walk them through the big phase out period and so they pay taxes that totally offset their EITC benefit. But when analysts calculate the fiscal cost of the EITC, they don't include the EITC amount that people beyond the phase out period did essentially receive. They reason that because that EITC was effectively taxed back that there was no real fiscal cost there.

Now we can quibble on whether you want to call that phase out period a benefit reduction or an income band with a fairly high marginal tax rate. But however you want to think about it, you need to also apply that way of thinking to the basic income.

So if you want to say that the fiscal cost of the EITC is actually just the sum total of the net benefit amount of each positive net beneficiary (which is how it is reflected in budgets), then you have to say the same thing for the basic income. For everyone who pays more in the basic income tax than they get in the basic income benefit, you cannot count their basic income benefit as a fiscal cost. And for those who pay less in the basic income tax than they get in the basic income benefit, you have to reduce the benefit amount by the tax amount to get their fiscal cost.

The only way you end up thinking that a negative income tax inherently has a lower fiscal cost than a basic income is if you apply totally different accounting standards for how to calculate their fiscal cost. This happens when budget wonkery gets too into the spreadsheet to remember what all the numbers actually mean.

Rehash of Basic Income "Costs"

I wrote before about the deep confusion that goes into discussions about how much a basic income would cost. Adding up all of the check amounts and taking the tally and calling it the cost is all well and good if you are just trying to communicate that figure. But when you are using that figure and calling it a cost as an argument against a basic income, that's where you really need to be careful. As I explained in my three-bucket model of government costs, it is obviously the case that spending a dollar on a basic income costs some other way of spending it, e.g. on a road. But a policy of not taxing a dollar that you could tax equally costs some other way of spending it, e.g. on a road.

If you are committed to objecting to a basic income purely because it is costly in the above sense of the word, then you also must object to a policy of not levying some possible tax purely because it is costly in this sense of the word. Of course, nobody ever does that because ideology is strong, but that's a point for another day.

Now, if by costs you mean macroeconomic costs, that's actually a far more legitimate and coherent line of criticism. This argument observes that both the tax and the transfer part of a basic income could reduce labor supply or have some other effect that reduces overall productivity. Moving dollars around is not a real macroeconomic cost, but the way it alters incentives might be. However this more serious analysis of cost has very little to do with the size of the budget line item for the basic income, and discussions of it should not proceed by gawking at that budget line item figure. They should proceed by modeling labor supply effects and tallying up that total cost, which is of course totally different than what they actually do when they talk about the cost of the basic income.