Implementation Issues with a Guaranteed Annual Income

On Wed, January 15th there was a Twitter discussion about the idea of a basic income supplement with economists Stephen Gordon (Laval), Kevin Milligan (UBC), Mike Moffatt (Ivey) and myself facilitated by the twitter account, @bis_pilot. To kick off that discussion, I had a blog post that provided some background on the concept. The main items of the twitter discussion itself are featured in this storify and the full discussion can be read by searching #bispilot on Twitter. I would like to follow up on this discussion with some more fulsome thoughts on some key issues.

As an economist working in a School of Public Administration I have now been well schooled by my public admin colleagues about the importance of implementation. Before moving to the School, I was your typical economist who did not consider such issues. But now implementation is so important to me that it has spurred a whole research agenda. Implementation is the reason why good policy ideas fail (e.g. HST, congestion charges, tax reform). Implementation has been dubbed the ‘missing link’ of policy analysis yet this is what transforms a policy idea into successful delivery. Implementation is not just what happens after an idea or policy design, implementation overshadows policy from an embryonic idea all the way through to evaluation.

With my implementation hat on, what are some key things that really need to be outlined as part of this BIS proposal?

1. Objectives: What objective(s) or goal(s) is/are to be achieved by the policy? What exact problem is the policy attempting to resolve? Objectives outline the desired result to be achieved and give a sense of direction by providing guidelines for activities. Proponents need to be clear about why they are advocating a BIS and to reflect this in clear and transparent objectives.  The primary objective(s) need to be paramount during design and implementation and communicated to the public to make sure they fully understand what is being proposed. There are two caveats to this. First, too many objectives and the policy becomes confused. Second, lots of people agree with an idea because they insert their own objectives into it. Their objectives, however, may not be yours. A lot of mistrust and confusion can arise when objectives are not stated in advance of implementation.

How much a BIS might cost is very dependent on setting out these objectives. Typically, the objectives of a BIS is to help lift at risk groups out of poverty. This requires us to agree on a definition of poverty, who are poor, and to what level to lift them. Do you target only the working poor or all poor? What about those on reserves? What about kids? What about the homeless? What about regional differences? How fair is it to provide $18k to everyone when the costs to live vary by regions and family conditions? Lots of questions need to be answered in order to be able to operationalize this objective.

2. Jurisdiction: Once it has been determined that BIS is the preferred policy instrument for the identified policy problem, the next step is to consider jurisdictional authorities. A BIS either needs to work  in conjunction with or replace our existing social assistance and tax benefit programs. Such programs are delivered by NGOs, municipalities, provinces, and the federal government. Any structure would need to have unprecedented cooperation between all these players. While many people point to the fed-prov agreements for delivering child income support, I would like to point out that these agreements have not done much to reduce child poverty in Canada. In addition, if the notion of  a BIS is delivered as a Negative Income Tax (NIT) by a province rather than Ottawa, than all due respect must be paid to the tax collection agreements.

The key considerations of cost and size of benefit is dependent on knowing how the BIS would work with our existing programs and tax system. It is also key to understanding where the funding for a BIS might come from.

3. Tax considerations: The BIS being advocated is that which takes the form of a Negative Income Tax (NIT). The advantage of delivery the program through the tax system is that it avoids the social stigma that many of our social programs have. In addition, it would force people to file a tax return in order to benefit. But in that lies the rub. For many, particularly the poor, having the knowledge and the trust to fill out a tax return is no small hurdle to over come. It makes me wonder what companion policies need to go hand in hand with the BIS? An expansion of the CVITP? Perhaps even pre-populated tax forms?

We also need to think about what incentive the BIS would have on our old nemesis, tax evasion. The BIS creates very real incentives for those with the ability to under-report their income to maximize their BIS payment. Consider even just those with tip income or those with self-employment income. Tax evasion in the presence of a BIS creates real distortions and challenges with achieving the stated objectives yet overcoming tax evasion is not easy.

The last tax item I will consider for this blog post is those who earn income from non traditional sources like personal corporations and windfalls. WRT to personal corporations these individuals can choose how much they pay themselves and how to pay themselves. How would dividends and other payments factor into the calculation of the BIS? Further, many of these individuals can choose to leave their money in the corporation, extracting it only from the share sale of the business which, low and behold, is tax free assuming a CCPC structure. These individuals can benefit both from the BIS and from the income earned from the corporation. How do you deal with these very real realities of our existing tax system?

In Canada, windfall gains (also prescribed prizes which will be a topic of a future blog post) are not considered income. A windfall gain is a type of payment that is unexpected or unplanned and usually does not recur, such as a lottery. So someone could win a lottery or take home the pot from gambling and still get the BIS unless we explicitly change our treatment of this gain such that it is reported, and perhaps even taxed. But again, this requires tax reform for which there seems to be little political appetite.

These are just some of my concerns, not with the idea of a BIS, but with the implementation of a BIS. You cannot implement a BIS without breaking some eggs and changing a lot of other policies. Other concerns were raised during our discussion like cost, benefit levels, affordability, labour market attachment, and revenue sources. Settling these and other issues matter a great deal for my acceptance of any particular proposed program.  I worry that the good folk pushing the Basic Income System (BIS) do not fully understand that by leaving “such details to the experts” the idea is likely to once again not get off the ground. After all, the devil is indeed in the details.


6 thoughts on “Implementation Issues with a Guaranteed Annual Income

  1. Just give the exact same BIS amount to every person and stop wasting resources trying to claw it back. Raise sales taxes and income taxes to compensate.

    • Thanks for the comment. What you propose is a universal demogrant (UD). Such a form of GAI is less targeted on the poor than existing programs and would mean billion given to middle and high income earners that don’t need it. Even with tax increase (which BTW assume a lot about elasticities that we know to be untrue) a UD that actually address proverty would far to expensive to be implementable.

      • I think if you leave the benefit levels at about 165% of the absolute poverty threshold which is in use in the United States for a single person, you’ll definitely find a Basic Income to be too expensive, though the expense of ensuring a BI cheque goes to the top quintile will only increase the cost of a Basic Income by 20%.

        Frankly, the neoliberalized quasi Basic Income being floated here seems determined to repeat all the problems of the welfare state.

        Canada could afford a BI that covers 25% of per capita GDP, that’s about $1065 for a single adult, or about 150% of what they’re getting now, unless they’re over 65, in which case it’s 85% of what they’re getting now. (Apparently we all magically became just as able to work our 66th year as our 26th the other day, never mind the life-expectancy gap by income). To do that, at the federal level we’d need a 30% marginal rate (or a little less for incomes under $180,000 if we were to take Piketty-Saez seriously), an end to GST exemptions, a carbon tax of $50/tonne, no tax expenditures, clawing back the 3.5% of GDP that the provinces currently spend devising some of the most Kafkaesque income support programmes in existence, the 5.5% spent on interpersonal transfers by the federal government, somewhat higher corporate tax rates (~25%), using a portion of U-18 BI to replace education funding (3.6% of GDP) and a change to the tax code that treats inheritances as income.

        The issues are all jurisdictional and political. But to claim that we can’t give everyone the capacity to live out of poverty, instead of out of less-than-my-neighbours, presumes a purely relative poverty line, that grows at the same rate as mean income, and always has. This is bunk. A study of national poverty lines shows that as income increases, the national poverty line increases, not at a 1:1 rate, but rather, logarithmically.

        This, by the way, is where the programme becomes highly affordable.

        The present slope for a country with our kind of per capita income is 0.18, but let’s be radical and say 0.5, which means for every percentage point increase in real per capita GDP, the real value of the Basic Income should increase by half a percentage point. Every moment of growth and progress brings those with the lowest earning capacity among us further from poverty, but the size of the programme, as a percentage of the economy, shrinks.

        Now should we find ourselves unable in the short-run to prevent the kind of crises that plunged Canada into a better than three-point output gap for a time, and which still persists as a very mild depression. (And indeed, an output gap this size, sustained for 7 years, I believe we can safely call a mild depression), we will be able to move through the most expensive years of Basic Income’s introduction via fiscal stimulus. When you need a deficit of 3% of GDP, suddenly you need only find 22% in spending… and if you don’t think an economy about to swallow the largest housing bubble in Canadian history, larger in rent-price ratio terms than the American housing bubble, isn’t likely to have some liquidity-trap-level weakness in the coming decades, I must express significant surprise.

        So yes, I wouldn’t support a guarantee that would simultaneously mean a real benefits increase for someone like me, and the massive increase in marginal tax rates that would come with it. Are you trying to discourage bottom-decile work as badly here as the United States has done? Frankly, I think we need to recognize that to the extent that the laffer curve exists, the peak moves right as income increases, and thus returning to the high-clawback system which income support recipients already deal with will produce results which will be used to discredit claims of work-effort once the threat of penury is removed.

        Honestly, I begin to wonder if the people throwing $19,000 a year around have lived on so little. It must seem a terribly bleak number to them. To me, it’s 90% of my 2011 after tax income. And that was a good year for my bank balance. Perhaps they also want to solidify the idea that child subsidies are really for adults, and thus propose not extending BI payments in any form to U-18s, in which case, they have made being single and having a third child a sentence of actual poverty.

        While I respect logistical issues, the affordability claim (and when you’re trying to find 36% of per capita GDP instead of 25% going on 24.8%, I can see how you would find a universal payment, i.e. a far more politically secure payment,) is thrown in with no back-up, no calculations, and frankly, I think the generally eroneous source poisoned Sarlo’s somewhat more accurate (but still, riddled with problems) poverty threshold calculations.

        Ask people who share a three-bedroom house what rent costs, ask your local municipality what transportation costs, and you can check with the USDA, but they still think $365 is lots for a single person for a months’ worth of food, and as someone who does the shopping, so do I, unless you’re eating two pounds of steak daily.

        Your threshold number is in left-field, you replicate all the marginal rate on those with highest marginal propensity to produce (i.e. lowest income) problems one would hope that economists had learned to avoid, and you use, instead of an entitlement-level benefit, unaffected by income, you use the Tax Return, and a tax rate plus a claw back rate is still a marginal tax rate, whether you put it in the tax code or not… to which I will only ask this one logistical question:

        What if I lose my job on January 2nd?

        Frankly, I’m stunned. It’s all the worst ideas in one place.

      • I am not at all clear on how your comments relate to my concerns outlined in the my post. You seem to be overly focused on material completely unrelated to my blog post. Affordability was discussed in the twitter chat and not something that I wanted to rehash. It was noted clearly in the chat that a UD approach to GAI would be very unaffordable as would a means tested NIT payment at $19k.

  2. Great blog and helpful posts on BIS/GAI/NIT. Thank you.
    Your post put forth two caveats:
    1. too many objectives and the policy becomes confused;
    2. lots of people agree with an idea because they insert their own objectives into it.
    Even so, consider …
    1. the notion “constructive ambiguity,” introduced in an address to an ORSA session by Harlan Cleveland. Challenged to explain how ambiguity could ever be constructive, Cleveland replied: “Who in your family REALLY makes the decisions?” The response: “It depends.”
    2. the notion of multiple and/or mixed motives, routinely rejected by the “must not merely do the right thing, but must also do it for the right reasons” crowd. Among others, pedagogue Clive Beck, in his text Moral Values, defended multiple motives as undeniably helpful when purposing contentious issues in complex social environments.

    • Thanks for the comments. I do understand your point and the point of the folks behind this proposal, however ambiguity and mixed objectives are the death knell of public policy.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s