A Public Policy Keltner List

In the 1985 edition of his Baseball Abstract, Bill James published his Keltner List, a series of 15 questions designed to evaluate a baseball player's worthiness for the Hall of Fame.

I have a similar, albeit shorter, list of questions I like to ask when analyzing public policy proposals.

1. What is the policy?

We need to know what we're talking about.

2. What is the policy meant to accomplish?

Very straightforward.  Is the policy meant to reduce unemployment? Increase productivity? Reduce hospital wait times?

This question weeds out the policy-for-policy-sake type proposals:  "Let's build a high-speed train from Saskatoon to Regina.  Trains are cool!"

3. Will the policy accomplish what it intends to accomplish?

In other words: Will it work?

4. What will the policy cost?

Where cost can be used in a broad opportunity cost sense.

5. Does the policy have any unintended consequences (either positive or negative)?

What side-effects does the policy have? How important are those side effects?  Are they positive and negative?

6. Is this the best policy for the job?

Are there other policies that would accomplish the same thing (or better) at a lower cost and/or with less negative side-effects?

Those are my six questions.  You can add others.  I did not add any questions about political feasibility.  Partly because I do not feel particularly qualified to answer those questions and partly because my position on political feasibility has always been I understand but I don't care.

Let's apply this Keltner List to a recent proposal - Belinda Stronach's term limit proposal:

1. What is the policy?

In her words:

My proposal would be to limit the number of consecutive terms that an MP could serve to two, and have a roster by lottery, to ensure a turnover of 50 per cent of members in each election cycle.

I'm curiously if this would apply only to majority terms.  Otherwise you may have a generation of politicians with very short careers.

2. What is the policy meant to accomplish?

Reducing partisanship.  At least, I think that's what it's for:

The role of parliamentarians should be to work for the best interests of the country in a constant and postpartisan search for solutions. In the way our current electoral system is structured, the focus is almost entirely on the mechanics of getting re-elected once elected, and the time horizon for assessing the impact of decisions tends to be the immediate electoral cycle rather than what’s good for future generations of Canadians. The emphasis must be on public service where politics isn’t seen as a career but rather as a tour of duty. Terms limits could encourage this fundamental shift.

3. Will the policy accomplish what it intends to accomplish?

I personally do not see how.  Presumably many of these politicians would want to have jobs with their parties (party brass, pundit, consultant, etc.) after office, so they would still want to be seen as 'team' players.

4. What will the policy cost?

As far as a monetary cost?  Very little:

For example, to help cope with the challenge of having higher-than-normal turnover and numbers of new members, much better orientation, training and mentoring programs for new MPs could be put in place. 

Direct costs seem to be quite low with this proposal, so that is one thing it has going for it.

5. Does the policy have any unintended consequences (either positive or negative)?

A raft load.  From scaring good people out of politics (why enter if you know you can only stay for a maximum of 8 years?) to having a host of unaccountable politicians in their 2nd terms.  I would want to see these issues addressed in a systematic way before even considering such a proposal.

6. Is this the best policy for the job?

Hard to analyze given that we don't know all the unintended consequences of the proposal.  I suspect Mark Jarvis could come up with a number of proposals that would have far fewer unintended consequences.  I would want to know what the alternatives are before implementing this proposal. (Edited to add: Mark kindly pointed me to his Macleans.ca piece - The House: Three Part Reform)

I hope you find this way of thinking helpful.  I find it clarifies issues for me if I think about them systematically.

Comments

[Stephen Gordon]

This is going to be very handy. Maybe if these questions are asked often enough, politicians will start thinking about the answers *before* they make the policy proposal, and not trying to do it off the tops of their heads *after* the proposal.

[Ralph Musgrave]

Question No 7:

Are there other extant policy instruments designed to achieve the same policy objective, and if so, which policy instrument should be abolished so as to cut duplication of effort?

E.g. the policy objective “support pensioners” can be achieved by the instrument: “pension”. The latter does not need to be duplicated with instruments like “cut price public transport for pensioners” or “local tax concessions for pensioners”.

Hat tip to the economics Nobel laureate, Jan Tinbergen who formulated the principle that (in my words) “the number of policy instruments must be at least equal to the number of policy objectives”.

I prefer a slightly different wording: “each policy objective requires one instrument and one only”.

[Mike Moffatt]

Ralph: That's *terrific*. I will amend with your question. I may want to change the ordering of the questions in order to reflect that.

[Shangwen]

I definitely appreciate the demand for clarity in that it might lead to better government. But if you are a politician, senior bureaucrat, or a lobbyist with a hand in designing a policy, vagueness is an asset, not a weakness. The type of clarity you're asking for is the kind of interest to taxpayers; the kind of clarity desired by the rent-seeking parties are the kind usually spelled in soporific detail in law or left advantageously open in ways undetectable by the public, e.g.:

1. Will this protect me from competition? (Supply management of dairy products)
2. Will this propose disproportionate burden on my competitors/opponents? ("buy Canadian")
3. Will this boost my partisan allies while leaving my own stakeholders off the hook? (Waivers in the Affordable Care Act ["Obamacare"])
4. Will this convey the impression that we are doing something, and conceal our real powerlessness? (National wait times intiative)
5. Does this give us much larger access to public funds that would at first appear? (veterans benefits in the US; R&D tax breaks and subsidies, etc.)

The people who support these policies are getting very clear answers on the impacts and merits. What taxpayer with a real job is going to sit down and read the legislation?

[Determinant]

Hmm. Universal Pharmacare.

(1) Universal provision of a list of drugs that are free or have trivial dispensing fees at the point of use.

(2) Universal, equitable access to drug therapies not administered in a hospital or clinical setting. Insulin for diabetics is a poster-child example of an expensive, widely used and utterly necessary therapy that is not paid for out of Medicare.

(3) Yes. See the NHS in Britain. Universal coverage for drugs. Not a complicated proposal on its face.

(4) It will cost at least as much as our present drug costs, probably 10% more to ensure equitable access to currently uncovered individuals. Large potential savings to be had in mass volume discounts and rigorous clinical evaluation of new (read expensive) drugs as was done in BC. Elimination of current employer drug plan costs as well as tax incentives to provide same.

This will eliminate the employer based medical benefits brokerage industry but this same industry already provides disability insurance benefits, often in the same plans. DI benefits are unaffected by this proposal and the insurance market in these benefits will remain.

(5) Only in coverage of dubiously therapeutic drugs (see BC evaluation policy) and elimination of drug deductibles and copayments. The NHS in the UK is a good example of the use of low copayments to control costs. NHS drug benefits have been in place since 1948.

(6) Employer based benefits suffer from poor plan design, poor cost control, (both related to poor drug evaluation or inflexible labour contracts). Benefit waiting periods of six months or more are common and cause significant gaps in coverage.

Government policy reliance on employer-based benefits ignores the significant segment of the population that is employed in small businesses, contract, self-employment non-standard labour and those who are unemployed for significant amounts of time.

The private, individual, voluntary community rated insurance market has outright disappeared in Ontario and Quebec, the former has not replaced it and the latter has taken the common policy step of mandating coverage.

Pharmacare is indicated due to demonstrated failure of all previous existing proposals.

[Bob Smith]

I suspect the problem isn't so much with the questions, but with the answers. I'm pretty sure that policy makers do ask those questions (or similar questions) with respect to most policy proposals. I'm alo pretty sure that every proponent of a particular policy answers questions 3, 4, 5 and 6 as follows "yes", "not much", "no" and "yes" without too much regard for the truth of those statments. (In fact, both the questions and my suggested answers would not be out of place in the typical regulatory impact analysis that the feds do when they introduce new regulations - which analysis typically gives new meaning to the word "superficial").

The problem is that if politicians, bureaucrat, etc., aren't well placed to recognize those answers for what they are, the fact that they're asking sensible questions doesn't do much for them. Asking good questions isn't of much use if you can't tell sh*t from shinola when it comes to the answers.