Monday, 22 January 2018

Hard to get good advice

A friend in one of the Ministries sends me this, after reading the post on Treasury's interesting problem definition in the RIS for the ban on foreign house buyers:
A lot of the messaging at [REDACTED MINISTRY] at the time it started appearing more likely that there’d be a change of government, was about how giving Helen’s government in 1999 frank advice on the wisdom of its election promises undermined the relationship and caused distrust. So it was made fairly clear upfront that ministries were not going to be telling ministers that their policies didn’t make sense.

Thinking back, that was reported to me at a meeting as having been discussed at a senior officials meeting - interagency senior officials - at which Treasury would've been represented.
Weight it as you will.

I wish Ministries were less risk averse on this stuff and more willing to consider that their Ministers just might want to get their officials' actual views on stuff rather than pandering. And that Ministries also have a responsibility to Parliament and voters in providing accurate assessments of the likely effects of policies.

I'm sure I've seen a paper on this, likely from Ami Glazer, where an agent who's revised her views on something has incentive to preference falsify if the principal will interpret the flip as the agent's having become disloyal.

Adapting to the current scenario: if the Minister will interpret the Ministry's "Your policy will not achieve its intended effect and/or will have these bad side-effects and/or is way less effective than this other policy" advice as the Ministry not sharing the Minister's values, then the Ministry will want to dissemble when there isn't as much at stake so it won't be written off as disloyal or as providers of 'ideological burps'.

It's a fun problem though. Imagine that you're an incoming Minister, and that your Ministry expects to be punished for offering frank advice, and that you actually want frank advice.

Your Ministry is second-guessing you and trying to avoid saying anything that would make them appear disloyal on margins they think are less important, so that they might be believed if a sufficiently important issue comes up and they have to give advice contrary to what the Minister would like to believe. They, and every other Ministry, have had successions of Ministers that have claimed to want honest advice, but who've burned them for providing it. Inevitably, somebody OIAs the advice, the Minister takes flack for it in the press and in the House, and the Minister's gotten mad at the Ministry for the advice or for writing it down.

How can you credibly signal to your incredibly risk averse Ministry that you actually want frank advice? Even Stalin sometimes wanted the real deal rather than pandering.*

Can a Minister require the Chief Exec to make bets on policy outcomes?

* I really like Xavier's post on this. An excerpt:
Yet Pollock makes a good case that Stalin really wanted some genuine discussion and criticism as a way of furthering the progress of science, at least in some fields (though he underplays the connection of Stalin’s views on linguistics with his interest in strengthening national identities and making use of patriotic fervor), and goes on to make the more (speculative) claim that Stalin’s repeated assurances that science only progresses via discussion, and that it is not necessarily class-based, account at least in part for “science’s rising prestige in the post-Stalin decades”. Stalin really needed (some) science to work well in the coming competition of the Cold War, and dimly understood that this could not happen if dogmatism reigned everywhere. Yet as long as he was alive, no such discussion could take place. His influence was like that of an enormous gravitational body; once he intervened (or was even suspected of intervening), the space of discussion became completely warped.

Dogmatism was safety: one needed to know where to stand in order to get on with life. Wherever the orthodoxy was unclear, best not to tread.

Friday, 19 January 2018

Afternoon roundup

A selection of the worthies in the weekly closing of the browser tabs:

A Treasury without opportunity costs

Sometimes, I hear the criticism that the Treasury’s focus on living standards implies a lack of focus on the fundamentals.  That what we should do is train our sights on the basics: sound public finances, value for money, stable macroeconomic frameworks, and robust microeconomic analysis.

I both agree and disagree with this sentiment.

I wholeheartedly agree that those fundamentals remain critical, for both New Zealand and the Treasury.  As I’ve said before, it remains a fundamental truth that successful economies need, among other things, a stable and sustainable macroeconomic framework, sound monetary policy that delivers stable and predictable prices, a prudent fiscal policy and debt that’s under control.  For a small economy such as ours, fiscal control is critical and rumours of its death have been greatly exaggerated (and I’m happy to nominate any of my Public Sector Chief Executive colleagues as a reference on this point!).

I disagree that these fundamentals are all that we ought to focus on.  The fundamentals are necessary but they’re not sufficient.

Good public policy needs a wide-angle lens and, at its heart, the LSF is about what we mean by ‘value’.
Emphasis added.

One of the fundamental lessons of economics is the concept of opportunity costs. If you want to start a large body of work in a new area, you either need to staff up to do it, or stop doing other stuff to make room for the new project. This should be obvious to anybody whose boss has ever said "Oh, could you also do this by the end of next week?" The correct answer is almost always "Sure, but here's a list of the three other things I was going to get done - which should I push out to do this new thing?"

Hiring a pile of non-economists to do living standards stuff has an opportunity cost for rigorous analysis on other margins.

Case in point, Treasury's Regulatory Impact Statement on the policy banning foreigners from buying houses in New Zealand.

The problem definition section, required in establishing exactly what problem is out there that the policy is meant to solve, is to the point.
This proposal seeks to implement the Government's 100 day commitment to "ban overseas speculators from buying existing houses." 
What a marvel of problem definition! Empowered by the Living Standards Agenda, Treasury will soon be able to move on to even more concise definitions of New Zealand's pressing problems, like "The Government wants to do something about foreigners", or perhaps even "A Minister heard something on the radio that made him angry."

Section B's Summary of impacts is at least as stunning. It notes that costs will fall on regulated parties, the Overseas Investment Office, and third party agents involved in property transactions. Ignored are the costs falling on owners of New Zealand property who might like the opportunity to sell to a broader range of buyers. But those kinds of effects only matter in old-style cost-benefit assessment and not in the new Hexagon of Happiness, Pentagon of Pleasures, Dodecahedron of Delight - or whatever the latest variation of it now might be. In a world where more economists were available at Treasury to hit this stuff, maybe these things could be more rigorous.

I shouldn't be too harsh on Treasury here - they can't do much rigorous work on stuff that parties commit to in a 100-day-plan. An alternative and plausible reading of that problem definition is "We see absolutely no problem here that we can specify so we're just going to say the government made us do it."

But I'd have far preferred that Treasury instead work backward to the real problem Labour was trying to solve around housing shortages, then in the RIS note that banning foreigners is a dumb solution relative to plausible alternatives like getting on with the infrastructure financing changes that Phil Twyford's supported. I guess it's always a balance between pandering and providing free and frank advice. But if it's always the teaspoon of sugar and never the medicine, there'll be problems.

Anyway, I'll expect more of this kind of thing if, in the real world, there are opportunity costs to Treasury shifting away from the fundamentals.
Treasury has apologised for an error which could see fewer children projected to be lifted out of poverty as a result of Government families packages.

It's not yet clear whether heads will roll over the coding error, but officials have confirmed that child poverty reductions were over-estimated when the previous National Government delivered its Budget in May last year and the error appears to have carried through to affect the current Government.

It is understood one of Prime Minister Jacinda Ardern's first big speeches of the year would include a child poverty focus, but the Treasury blunder will likely throw early plans into doubt with the Government unable to be certain of its figures.

The exact number of children expected to be lifted out of poverty is not likely to be known until the end of February.
Again - really don't want to be too hard on Treasury for the coding error - it's hard not to get the occasional screw up in this stuff. Lumley's take is good. [Update: So's this at RNZ] But it points to the problem of opportunity cost. Getting the fundamentals right isn't easy. The more they're pushed to run Living Standards stuff, the more likely we'll get screw-ups on what I'd thought was more fundamental. And the more annoyed I am about how hard Treasury pushed to be stuck in this spot.

I chatted a bit with a Newshub reporter on this stuff yesterday afternoon.

Postscript: National's carrying on during the election campaign about the number of kids its tax change would lift out of poverty bugged me. At least some of it was an effect of finally inflation-adjusting the tax brackets. If inflation-adjusting the bottom tax brackets brings kids out of poverty, then it has to be the case that failing to adjust them in the prior years drove those same kids into poverty.

You can imagine a particularly stupid political ratchet where governments don't inflation-adjust tax brackets, more households hit a disposable income measure of household poverty because inflation pushes up the amount of tax they pay, and then every government in its third term announces an inflation adjustment that brings those same households back above the defined income line. Everybody celebrates the magnanimity of the government that saved those children from poverty while ignoring that failure to annually adjust the damned thresholds is what caused those kids to have fallen below the line in the first place.

Thursday, 18 January 2018

Rental rationing

Louis Houlbrooke has an eye for interesting questions.
Why wouldn't rentals simply clear on price?

The credit rationing literature points to one potential answer. In that literature, the market for loans doesn't clear just on interest rates because the people willing to borrow at the highest rates know that they're very risky types. So instead you get interest rates a bit below what would otherwise clear markets and lenders choosing among applicants to get borrowers who are good bets.

Application to rental markets? Suppose it's hard to evict a bad tenant. It'll take a long time, it'll be a hassle, and the tenants might destroy the place while you're going through the tenancy tribunal. If you set a high price and if it's hard to monitor and police what's going on in the flat, you might have problems. The high bidder might be the one expecting this to be a short-term game.

Landlords would want to evaluate a potential tenant's bid across a pile of hard-to-specify and possibly illegal-to-specify (but impossible to police unless you're dumb enough to write it in the ad) non-price margins. If you want that, you want to have excess demand at the posted money price so that you can clear on the other margins.*

  • The harder it is for landlords to evict problem tenants, the more we should see this kind of non-price rationing;
  • The harder it is for landlords to specify their actual requirements in a for-let ad, the more we should see queuing. Otherwise, you'd see less queuing and more clearing on prices among those who meet the landlord's other requirements;
  • Non-price rationing should be more common for flats where landlords are more worried about there being problems. 
  • The burden of non-price rationing will be felt hardest by people who would be good tenants but who landlords can't distinguish from risky tenants. They'll keep bidding on flats and keep failing to get one;
  • If you're in the market for renting, and you're of the type that landlords would prefer to rent to, you'll do better by finding a way to credibly signal as much. 
Meanwhile, Dan Rowe over at The Spinoff complains simultaneously that rents in Wellington are too high and that there's too much queuing and non-price rationing and that tenancy protection laws are too weak. This is what happens when you don't structure your thinking around an underlying economic model. Price increases and non-price rationing through queuing are both outcomes of demand exceeding supply at the prior price. 

Unless you address the underlying shortage, all you can do is choose among various ways of rationing scarce supply - whether prices or queues. Doing it through prices at least provides investors with incentive to build new rental stock and provides homeowners with a reason to put up with the hassle of having a tenant in the spare bedroom. Doing it the other way doesn't. 

It is depressing how many people still like rent controls, despite the evidence against it. 

* Please remember that this is a positive rather than a normative analysis. Not saying that any of this is good, just that it's how we should expect incentives to play out. 

Wednesday, 17 January 2018

Tertiary access isn't about tertiary fees

If you want to improve university enrollment rates among Maori and Pasifika kids, you should look at what's going on earlier in the education system.

Lisa Meehan, Gail Pacheco and Zoe Pushon find that ethnic gaps in school performance are the largest contributors to ethnic gaps in university enrollment rates. Those gaps matter far more than differences in socioeconomic status or parental education.

They use administrative data held in the Integrated Data Infrastructure to control for meshblock-level deprivation index scores rather than school decile rankings, providing a finer grained measure of background characteristics. They also have parent's highest degree from the 2013 Census, student ethnicity, school characteristics, migrant status, and distance to the nearest bachelor-granting institution.

Table 4, copied below, provides a decomposition of the relative contributions of the different variables. So the total difference in bachelor's level enrollment between Maori and European is 19.84 percentage points. If Maori students had the same characteristics as European students on things like frequency of switching schools, migrant origins, neighbourhood deprivation index, school characteristics, NCEA Level 1 performance, number of school notifications and parents' education, that gap would drop to 2.64 percentage points. Providing Maori students with the same NCEA Level One results as European students would increase their enrollment rates by 13.39 percentage points. Everything else is rats and mice. Swapping Maori students' neighbourhood deprivation index figures with Europeans' would increase Maori enrollment by 1.45 percentage points.

When they disaggregate school performance, most of the work's being done by the number of merit and excellence credits; it would be interesting to have further work comparing differences across types of credits - an excellence in calculus might be different than an excellence in gym.

In Table A3, when they run school-level fixed effects, the school fixed effect does more than the neighbourhood deprivation index in explaining the Maori-European participation gap.*

The big upshot: neighbourhood background characteristics are important in explaining ethnic gaps in student progression to Bachelor's level study, but how students perform at NCEA Level 1 is what matters most. You could argue for a multilevel model where deprivation characteristics work through two channels and also affect NCEA Level 1 performance. But all of it suggests that if you want to increase enrollment in Bachelor's level study among Maori and Pasifika students, you shouldn't be looking at zero-fee first year university policies. You should instead be working to improve performance at secondary school.

And that's exactly what I'd argued in our report a couple years ago on the zero percent student loans programme. I there argued that the $600 million per year that's blown on subsidies through the zero percent loans scheme should instead be directed to means-tested funding for tertiary students in need, and toward better preparation for tertiary study in high schools with poor track records in advancing kids through to tertiary study.

Instead, the government's looking to compound the problem of zero percent loans with a fees-free policy for first-year study. The inevitable critique is "Why not do both! Fees-free study and big increases in secondary school funding!" But the government is trying to work within a budget constraint that binds. Everything has an opportunity cost.

Meehan et al conclude:
Overall, our results suggest that ethnic-based policies aimed at encouraging entrance to bachelor’s degrees are likely to have a limited effect if used in isolation. Rather, our findings highlight the need for policy intervention earlier in the education system to help lift the NCEA performance of Māori and Pasifika, and in doing so improve the likelihood of their participation in higher education qualifications, such as bachelor’s degrees.
* Now, imagine a world in which it were not illegal for Meehan et al to simply array schools by their fixed effect and to announce the league table. Stats NZ won't allow it because anything individually identifying, including things that identify individual schools, is forbidden. But imagine. Wouldn't it be interesting to know, correcting for everything else about the kids, which schools have the best track record in progressing students through to Bachelor's level study?

Update: Dave Guerin weighs in, from his excellent Ed Insider newsletter:
2. Access Economist Eric Crampton looked at an academic paper on ethnic disparities in bachelor’s degree participation (it’s basically the same research as released last year by the Productivity Commission). He argued that the best value for money would come from targeted support to tertiary students, and better preparation for tertiary study (not interest free loans or Fees Free). He’s quite right too.

Tuesday, 16 January 2018

Avoiding supply management

Canadians wanting to run a dairy farm have little choice but to deal with Canada's crappy supply management system. 

Well, unless you're the Canadian Government, and you're running a public sector pension scheme, and you want to make investments in dairy, and you know that supply management is crap.

In that case, you have the resource to jump through New Zealand's Overseas Investment Act hurdles and buy dairy farms here, where there's no supply management and you don't have to pay tens of thousands of dollars in quota fees to buy the right to milk a cow.

Here's Radio New Zealand:
The sale of a Canterbury dairy farm for more than $17 million to a company owned by the Canadian government has been approved by the Overseas Investment Office (OIO).

OIO approval was given in November for the purchase in the latest round of decisions for overseas investment of sensitive New Zealand land.

The transaction includes a medium sized dairy farm of 335 hectares and a neighbouring dairy support block of 72ha, also on freehold land at Hororata. They will be combined to create a larger dairy farm.

Applying for OIO consent was Ramsay Dairy Farm Ltd, wholly owned by the Canadian government and linked to a public pension investment scheme.
It isn't the first investment in NZ dairy by the Canadian Government, via Ramsay Dairy Farm Limited. Here's an OIA decision from 2015:

DecisionConsent granted
Section 12(a) Overseas Investment Act 2005
Decision Date12 November 2015
An overseas investment in sensitive land, being Ramsay Dairy Farm Limited’s acquisition of:
  • a freehold interest in approximately 322.8995 hectares of land at 249 Domain Road, Oxford, Canterbury; and
  • a freehold interest in approximately 34.6068 hectares of land at 282 Domain Road, Oxford, Canterbury.
ApplicantRamsay Dairy Farm Limited
Canadian Government (100%)
Oxford Pastures Limited
New Zealand (100%)
Farm Partners Limited
New Zealand (100%)
The Applicant is ultimately, indirectly, owned by the Public Sector Pension Investment Board.
The land is currently being used as a dairy farm. The Applicant intends to acquire the land for the purpose of dairy farming and milk production and will engage FarmRight to manage the land for that purpose.
I suppose it saves the Canadian Government from having to buy dairy quota. Would that Canadian farmers could do the same in Canada.

SOOBs and disamenities

Housing within Amsterdam's red light district trades at a discount. Here's Erasmo Giambona and Rafael Ribas.
We measure the externalities of prostitution by quantifying the discount that households require to live next to a brothel. In our tests, we exploit a unique feature of Amsterdam's Red Light District (RLD), area inside a perimeter naturally delimited by canals where private homes are located next to prostitution windows. Using a novel two-dimensional difference-in-discontinuity (DiD) estimator, we find that households require a discount as high as 24% on homes inside the RLD. We also find that this discount disappears when prostitution windows are forcibly closed by local authorities. By incorporating the exact coordinates of brothel closings, our empirical design allows us to establish a direct link between these closings and changes in price discontinuities. To estimate the economic impact on households outside the RLD, we look at the closings of all brothels in Utrecht (the fourth largest city in the Netherlands) in 2013. Households are found to have paid up to 12% of the value of their home to be some distance from prostitution. In both cities, the contraction of the paid-sex industry is also associated with a drastic reduction in crime rates. Overall, our findings suggest that the nuisances prostitution creates do more harm than good to residents. 
I wonder whether there's New Zealand data available on the location of brothels. Prostitution law reform and the emergence of suburban owner-occupied brothels, and varied council approaches to zoning and regulation, would provide plenty of room for diff-in-diff study. Red light districts could be rather different than smaller owner-occupied facilities. 

Monday, 15 January 2018

Vogons vs Backpackers

Backpacker hostels face big fines if they continue the long held tradition of casually offering travellers free beds in exchange for unpaid labour.

Some industry stalwarts say travellers can't be bothered with the paperwork now required to legally work for accommodation, and they worry it will destroy hostel culture.

But the Labour Inspectorate is preparing to get heavy with those who do flout the law and is monitoring job advertisements on backpacker and work exchange websites.
There is a real tax and distortion issue here, but there might be a simpler solution. Backpackers will have an incentive to barter for cleaning services with backpacking patrons because PAYE and GST won't be imposed. But that could potentially be solved by having the hostels report the value of accommodation-nights provided under that kind of arrangement and charge FBT on it. That would get rid of the tax distortion while allowing mutually advantageous trades between backpacking tourists and hostels.

YIMBY strategies

How can YIMBYs overcome the NIMBYs? Local action and state-level policy reform says Kenneth Stahl.

Stahl argues that NIMBYs block development because of worries that increased supply will hurt the value of owners' biggest asset - their homes. I'm not sure that's right though. Upzoning will reduce the cost of housing while increasing the value of the newly upzoned land. And this should be especially true for neighbourhood-level upzonings as compared to broader upzonings, and it's the local ones that seem to draw the most local opposition.

But the problems with incentives facing councils have a lot of parallels with New Zealand's problems. 

Where local government there is debt constrained due to unfunded pension liabilities, they're debt constrained here because of debt limits, poor prior spending decisions, and voter reluctance to authorise debt because of sensible worries brought on by poor prior spending decisions. Where councils there get a share of sales tax revenue that distorts council decisions in favour of zoning for business rather than housing, here councils get the costs of new development offset by rates revenue while central government gets the bulk of the benefit of growth. 

And does this sound like the RMA?
Streamline Environmental Reviews.  Some states require all new development to go through an extensive and costly environmental review process.  This process can easily be manipulated by neighbors to fight development they dislike for reasons having little to do with environmental protection. The California Environmental  Quality Act (CEQA), for example, has become a preferred tool of homeowners looking to squash new development, as they can force developers to spend additional time and money on environmental reports. Previous efforts to streamline CEQA in order to lower the cost of housing have failed due to opposition from environmental groups, and it is notable that amid the flurry of housing bills passed by the California legislature in 2017, not a single one touched the CEQA process.
The article doesn't explain why California's such a mess while Houston and Atlanta aren't. Better mechanisms for financing urban expansion is part of the answer.  

Sunday, 14 January 2018

Food fights

This is a slightly nuanced situation. The newspaper report seems to say that one part of the DHB shut down the operation of another part of the DHB and are using the new legislation as a reason. To be honest there is nothing new about caution over freezing food and re-heating it for later service. So I read it as the DHB auditing all of their food handling practices (spurred on by the change in legislation) and finding a practice they would have been unhappy with at any time had they known about it.

Although the new Food Act is largely a re-packaging of the old Food Act there are two relevant new inclusions in the legislation. The first is a specific reference to applying appropriate standards to food prepared for "vulnerable populations". This provision clearly applies here. The second is the duty of regulatory authorities to "... promote standards and control mechanisms that are, as far as practicable, risk-based and science-based...".

The new Act was sold to the industry as liberating because compliance requirements would now reflect the actual risk of each operation. Market stallholders would be allowed to prepare food in their home kitchen. We could do away with the nasty old "one-size fits all" regulations. Unfortunately the likely outcome will most likely be the complete opposite. From what I have seen so far a multitude of standards promulgated by Foodsafety NZ will simply be enforced on everyone with no reference to actual risk (let alone science).

This story in Tokoroa is just the beginning of a general shutting down of the food industry rather than a liberation.

Friday, 12 January 2018

Costly discrimination

Danish kids are happy to pay to avoid having to work with someone of a different ethnicity. In this clever field experiment, Danish kids with traditionally Danish-sounding names were willing to forego expected earnings in order to avoid being paired with someone with a Muslim-sounding name - and vice-versa. 

It's a great experimental design. Kids do a first round stuffing envelopes on their own, paid a piece rate. For the second round, they have to choose a day to come in, and they'll be partnered with another kid with a joint payoff for how much the team gets done. They're given information on what the other kid achieved in the first round. Because the framing is choice of day to come in for the task rather than choice of partner (though the day determines the partner), there's less chance that the participants would expect the experimenter to be inferring racial preference. 

Team productivity in this task depends on the productivity of each of the workers in the prior round - there's no diversity benefit or penalty. Of course it's just envelope stuffing.

Preference-based discrimination is symmetric. Participants with Danish-sounding names were willing to pay to avoid being partnered with participants with Muslim-sounding names, and vice-versa. 
The insignificant estimate on Danish-sounding indicates that the tendency to discriminate is not different across ethnic types, after controlling for differences in prices. We think that this is a remarkable result for two reasons. First, attention both in the literature and policy debates usually focuses on discrimination of the minority group by the majority group because members of the majority group are more often in the position to discriminate, and workers from the minority group tend to be disadvantaged. However, our results suggest that observing more frequent discrimination of minorities may simply be due to the fact that majority decision makers have more opportunities to discriminate rather than a stronger ethnic animus.
This too was interesting:
Second, this result highlights the importance of controlling for prices when measuring discrimination. From simply looking at discrimination percentages, a layperson may be misled to conclude that decision makers with Danish-sounding names are more likely to discriminate. In fact, decision makers with Danish-sounding names discriminate in 44 percent of the cases, while those with Muslim-sounding names do so in only 33 percent of the cases (however, p = 0.517, χ2 test). Yet, these differences do not reflect differences in animus because decision makers with Danish-sounding names face a lower price on average than decision makers with Muslim-sounding names (€5.2 versus €7.8, p = 0.078, KS). The reason is that workers with Danish-sounding names are systematically more productive (116 letters) in round 1 than participants with Muslim-sounding names (100 letters). According to regressions (2) and (4) in Table 3, these price differences explain the observed differences in taste-based discrimination across ethnic types (Danish-sounding is insignificant, but Price is significant).
They conclude:
Using a sample from Denmark, we find that discrimination is common even at a substantial price, that majority and minority groups are equally likely to discriminate for given prices, and that the demand for discrimination is highly elastic. Our best estimate is that the probability to discriminate falls by about 9 percent if the price of discrimination goes up by 10 percent.
Denmark may be more polarised than other places; would be very interesting to see this repeated elsewhere.

There are a host of interesting implications for policy, which the cautious blogger leaves as exercise for the reader. The elasticity of immigration restrictions with respect to the stringency of workplace diversity mandates or the legal status of repugnant homeowner covenants is one fun one to think about. Second best policy in a world of crooked timber....

Oh: Jean-Robert Tyran is one of the paper's authors. He also wrote my most favourite ever experimental economics paper.