You know, there’s a reason no one’s put in a guaranteed annual income yet

Posted on November 20, 2017 in Social Security Policy Context – Opinion – All the things promised by GAI advocates are good and worth having, but they are also impossible to combine in one package. Unless you’re willing to advocate for a drastic increase in taxes, that is
November 20, 2017.   STEPHEN GORDON

The Quebec government asked a committee of three economists to report on the feasibility of a guaranteed minimum income, and they presented their report last week. The basic thrust of the report — recommending some modest reinforcements of the existing income support system, and advising against grand schemes — was met with disappointment from the minister, and harsh words from anti-poverty groups for not coming up with a solution to poverty. This criticism is misplaced: no-one could have met those expectations.

There are three phases of the evolution of opinions about proposals for a guaranteed annual income (GAI). At first, you suppose it’s a joke. Then, after mulling the idea over for a bit, you come to the conclusion that a GAI is such a brilliant idea that you can’t believe it’s not already in place everywhere. The third and final stage is reached when you get to trying to figure out the mechanics of how it might actually work. It is at this point that you realise that while the GAI as it is usually imagined is not actually a joke, it’s not a proposal to be taken seriously, either.

At first, the GAI looks like Alan Frankel’s 1992 article in the satirical Journal of Irreproducible Results, which I quote here in its entirety: “Poor people do not have enough money. Give them enough money. They will no longer be poor.” Some people never get past GAI Stage I.

But once you get past the initial snort of derision, the idea of a guaranteed income starts to actually make sense. The idea of a “guarantee” is uncontroversial enough: we’ve already accepted that Canadians are entitled to a certain minimal standard of living. Why not make that implicit guarantee an explicit one?

There’s a lot good to be said for the idea of a flat, universal payment if you’re concerned about economic efficiency, A perverse result of means-tested benefits is that they effectively impose an extra level of income tax on low-income households: benefits are clawed back as incomes increase. Some low-income people face marginal tax rates in excess of 100 per cent: working more ends up in an actual decrease in their disposable income. A payment that doesn’t diminish as income increases would dismantle the “welfare wall” that blocks the path of many low-income people to economic independence.

In addition to these considerations of incentives, there is already a large body of convincing evidence documenting the benefits of direct cash payments. Some — if not all — of public money used to hire people to monitor and deliver services to low-income households could be simply sent out as cheques.

This is GAI Stage II, home to a small but vocal number of GAI enthusiasts. It’s a very comfortable place to be: you can spend your career campaigning for a GAI and writing about the benefits of a guaranteed income for low-income households (it’s almost always about low-income households) and for society as a whole. And fairly convincingly, too: the benefits are certainly real enough.

Finally, if you take the idea of a GAI seriously enough, you start working through the mechanics of how it would work. For example, suppose we want to send $2,000 a month to every Canadian — an amount approximately equal to Statistics Canada’s Low Income Measure — so that no-one would have to live in poverty. How much would that cost? The answer is $800 billion a year.

You’ve entered GAI Stage III when you grasp the implications of that sort of number: $800 billion represents roughly 40 per cent of GDP, or the entire take of all levels of government, including the public pension plans. (You’re still in GAI Stage II if I could have written “800 parsecs” instead of “$800 billion” without changing the effect or indeed the meaning of that estimate.) Unless you’re willing to try and persuade Canadians to accept a doubling of their current tax burden, you’re going to have to back well away from the sort of GAI that lights up everyone’s imaginations.

Some compromises are in order. There are three elements we’d like from a GAI: moderate cost, generous payments, and a low rate at which benefits are clawed back as income increases. (That last point refers to the welfare wall and the disincentives to work.) The UBC economist Kevin Milligan poses the problem as a trilemma: you can only have two of these three wishes. As we have seen, generous benefits with a zero clawback rate are hideously costly. You can have generous benefits at low cost, but only if they are focused on those with low incomes and are rapidly clawed back from those with higher incomes. And finally, you can have universal benefits with low cost, but the payments will be very small.

The committee mandated by the Quebec government could not avoid the implacable arithmetic of the GAI trilemma; all they could do was document it and identify the limits the trilemma imposes on the choices that are available. For example, if universal payments for people between the ages of 18 and 64 were to be financed with the existing income support budget, then they’d only receive $864 a year. Other scenarios for annual payments of $5,832 — an amount that was denounced as scandalously low — would cost around $30 billion, which is the equivalent to one-third of what the Quebec government currently raises on its own.

You’ve reached the third and final stage of opinions about the GAI when you conclude that all the things promised by GAI advocates are good and worth having, but that they are also impossible to combine in one package. Unless you’re willing to advocate for a drastic increase in taxes, the responsible thing to do is abandon the impossible GAI dream and focus on what is possible with current levels of tax revenues.

Tags: , , , , ,

This entry was posted on Monday, November 20th, 2017 at 9:14 am and is filed under Social Security Policy Context. You can follow any responses to this entry through the RSS 2.0 feed. You can skip to the end and leave a response. Pinging is currently not allowed.

Leave a Reply