Thursday 20 December 2018

Sunk costs in politics

Sunk costs matter more in politics than in markets. They matter everywhere, but firms have better incentives to stop throwing good money after bad.

In politics, you have to keep throwing that money at things like the America's Cup because reversing would signal that you screwed up in the first place, and politicians expect voters to punish them more for admitting that than for just riding the damn thing out.

This was part of a longer thread from Auckland Councillor Richard Hills, some of which is now deleted.

Auckland Council voted to throw money at a yacht race - the America's Cup. And when it inevitably came back begging for more money, as everyone had to have known it would, "you can't stop half way... you're not going to stop it when you've spent so much."

I wish Councillors believed it politically possible to say 
"Know what? We should have foreseen this pitch for extra cash, but we believed them when they said that this time is different. Our bad. But we're not going to give them any more money. If they want to cancel their boat thing or take it elsewhere, that's their call. But we'll want all our money back if they do - and we will never ever again put a dime of public money into a boat race otherwise."
I'm not blaming Hills here - he notes that he'd tried taking more expensive options off the table. But it is a nice illustration of political thinking around sunk costs.

Wednesday 19 December 2018

Coalitions mean plausible deniability

Who knows what the true model here is. Whatever the true model, we simply don't know whether statements by New Zealand's Foreign Affairs Minister represent New Zealand foreign policy.

Here's Richard Harman. You should subscribe to his newsletter if you want to keep track of what's going on in NZ politics:
The Prime Minister, Jacinda Ardern, confirmed yesterday that she did not see Winston Peters’ major speech on NZ-US relations before he delivered it in Washington, DC, on Saturday.

The speech has raised questions about whether New Zealand is moving its foreign policy closer to the Trump administration and away from the independent stance it has pursued since the ANZUS breakdown in 1985.

It has surprised observers in Wellington that such a major speech could be delivered without at least the Prime Minister’s prior approval and probably that of the Cabinet as a whole as well.

And it has left unanswered, questions as to where New Zealand sits between the United States and China.
Harman has lots more at the link.

So it's impossible to tell whether Winston's speech represents the current view of the government, whether it's Winston off on a branch, or whether it's a deliberate coalition strategy to have someone who can be dismissed as the crazy uncle cozying up to the US to let other parts of the coalition stay friendlier with China: "Oh, that's just Winston. You have to forgive him. They're retail politicians. Retail politicians can get away with anything here. They just have to remind people that they're retail politicians when they're doing it."

The thing I hate most about MMP is the diffusion of responsibility. Given the limits of voter attention, the best we can hope for is that they dish out reward or punishment appropriately come the election, and that fear of that constrains government. But when everyone and no one is responsible for policy, that is a bit harder.

Tuesday 18 December 2018

Petition of the Printmakers

We come to offer you a wonderful opportunity for your — what shall we call it? Your theory? No, nothing is more deceptive than theory. Your doctrine? Your system? Your principle? But you dislike doctrines, you have a horror of systems, as for principles, you deny that there are any in political economy; therefore we shall call it your practice — your practice without theory and without principle.

We are suffering from the ruinous competition of a rival who apparently works under conditions so far superior to our own for the production of light that he is flooding the domestic market with it at an incredibly low price; for the moment he appears, our sales cease, all the consumers turn to him, and a branch of French industry whose ramifications are innumerable is all at once reduced to complete stagnation. This rival, which is none other than the sun, is waging war on us so mercilessly we suspect he is being stirred up against us by perfidious Albion (excellent diplomacy nowadays!), particularly because he has for that haughty island a respect that he does not show for us.

We ask you to be so good as to pass a law requiring the closing of all windows, dormers, skylights, inside and outside shutters, curtains, casements, bull's-eyes, deadlights, and blinds — in short, all openings, holes, chinks, and fissures through which the light of the sun is wont to enter houses, to the detriment of the fair industries with which, we are proud to say, we have endowed the country, a country that cannot, without betraying ingratitude, abandon us today to so unequal a combat.

Be good enough, honourable deputies, to take our request seriously, and do not reject it without at least hearing the reasons that we have to advance in its support.

First, if you shut off as much as possible all access to natural light, and thereby create a need for artificial light, what industry in France will not ultimately be encouraged?
Here's Fran O'Sullivan in the Herald:
Rod Sims' report on the impact of Google and Facebook on Australian media and advertising could just as easily have been written for New Zealand.

Pity it wasn't.

The preliminary findings from the Australian Competition and Consumer Commission's (ACCC) Digital Platforms Inquiry confirm what most of us in the New Zealand news business are so painfully aware — Alphabet's Google and Facebook have decimated the revenues of traditional news media (print and digital) as they siphoned up the bulk of the advertising market.


But despite the trend being obvious, Governments — current and prior — have not been troubled sufficiently by the behemoths predatory behaviour in the advertising market to do anything about it.
The Sun reduces demand for candles by being that much more efficient at providing light (in daytime) for those wishing to see; online platforms may reduce demand for advertising in traditional media by being that much more efficient at providing viewers for advertisements for those wishing to purchase access to those particular viewers.
What is admirable about the ACCC report is the forensic probing of the way Google and Facebook have effectively ripped off journalists' work.

"This reduces value for the news businesses that have invested a lot of money and time in creating the content. Journalists may work many days or weeks to break an exclusive online story and a competitor can quickly reproduce that story, post it on a rival site which, due to the reach of the digital platforms, may draw traffic away from the original source of the story," says Sims.
So when an outfit like the Daily Mail rips off an exclusive that another paper worked on, Google and Facebook are to blame? Was it never the case that stories broken by smaller newspapers were ripped off by larger ones with bigger reach before 1999? The public-good nature of this stuff is hardly new.

Wealth inequality - reader mailbag

An informed reader reminds me of a couple points on wealth inequality in response to this morning's post:
Just saw your piece on wealth. To add to the points you make:
  • Strongly agree with your point about the use of cross sectional data. This is particularly true in a country where a large number of people in the first ten to 15 years of adulthood go abroad. We would expect the distribution to be highly skewed if the most effective means of building capital is to go abroad.
  • There is likely to be a depressing effect on saving associated with the generosity of the state superannuation scheme. Or to be more precise: people have an implicit overlapping generations model in the minds that sees long run intergenerational transfers by government as credible. This means they see the tax and transfer system as an implicit saving scheme. When combined with home ownership and universal health care, it is a reasonable question to ask what purpose would it serve people in the lower half of the income distribution to build up assets? (there is a paper by Andrew Coleman somewhere on this)
Both points are good. The latter one especially should be better recognised. NZ Super guarantees you a decent basic retirement income. NZ Super currently pays $926 per fortnight if you're single, or $701 per fortnight per person in a couple. After tax, it's $617 per person in a couple. 

So a retired couple gets $1234 per fortnight after tax in NZ Super, so $32,000 per year. That's in the middle of the second decile of equivalised household income for a couple, before housing costs. So anyone earning less than that could easily be forgiven for not building any retirement savings: they would be reducing their current consumption in order to gift themselves higher consumption in retirement than they were ever able to afford when working. What does that mean? It is completely rational for the lowest income twenty percent of the population to not be building wealth. 

Or, to put it another and perhaps better way, we should view NZ Super eligibility as being equivalent to lump sum wealth holdings for each person, rich or poor, equivalent to the present discounted value of the stream of entitlement payments once you turn 65. But that's not all! Government provides other services as well. There's a retirement care subsidy (which draws against your Super entitlement, but also involves additional subsidy from the government). Government provided health care too. Entitlements to all of these should be viewed as part of the lump-sum entitlement that's fairly evenly distributed. 

And what happens to any inequality measure if you ignore a giant lump sum that's provided across-the-board? The inequality measure will overstate real inequality. 

Household net wealth

My take on the latest Stats NZ Household Wealth survey is a bit different from Max Rashbrooke's.

Here's Max:
The Household Net Worth survey by Statistics New Zealand shows the wealthiest one percent have 20 percent of all assets and the wealthiest tenth have 59 percent. The poorest half of adults - 1.8 million New Zealanders - have just 2 percent of all wealth. Māori net wealth is significantly lower than that of Pakeha.
These figures are broadly unchanged since the last such survey in 2015. But that's barely reason to celebrate.
Wealth - in the sense of things that people own, like houses, cars, financial investments and cash in the bank - provides security and stability. It is something people can draw on during tough times, a stake in the community, a base from which to plan for the future.

So it is hugely concerning that so many New Zealanders have so little wealth. And their position is not improving. Statistics New Zealand reports that the poorest 40 percent have seen no increase in their wealth in the last three years.
But look at the chart underlying this:
There was a half a percentage point increase in the number of people with net wealth of less than $100,000: those will be people with substantial debt. When I check the student loan tables, about 0.075% of the domestic population has student loan debt into those figures - the rest has to be other stuff. 

Once we get into the positive net wealth categories, every net wealth category up to and including the $400-500k band had a smaller proportion of the population falling into it. Most bands above that saw an increase in the proportion of the population in the band, barring the $700-900k group. And the proportion of people with more than $1.5m in net wealth increased from 8.3% of the population in 2015 to 12.5% of the population in 2018.

It would then be fair to characterise this as an overall upward shift in wealth, barring a small increase in the proportion of the population with substantial net debt.

The median person in 2015 would have been in the $200-$300k net wealth range. The median person in 2018 is in the $300-$400k net wealth range. But note that the figures are not inflation adjusted.

The overall curve looks to me to have shifted to the right, barring the increase in the proportion of people with substantial net debt of more than $100,000. All the focus on what proportion of the population has what proportion of overall wealth misses what looks like an overall increase in wealth.

And recall too that life-cycle stuff enters heavily here. These annual snapshots give a picture of the cross-section in any particular year. But people accrue wealth as they age, then start consuming from that wealth in retirement. Stats' press release has young people (age 15-24) with median net worth of $2k and older people (age 65-74) with net wealth of $416k. 

Every year, new people are born, existing people age, and some people die. Looking at annual snapshots doesn't tell us anything about how people move through the life cycle. Every person in 2015's bottom 40% could have increased in wealth, with no change to the wealth band of 2018's 40th percentile person, as the 2015 people move up through the age ranks and new people come in at the bottom. 

I'd also be pretty nervous about claims around the proportion of wealth held by different quintiles if substantial net debt held by a small proportion counts at the bottom. Some of it is student loan debt where the human capital generated by that education is not counted in the wealth tables - but only a small amount. I wonder if some of it is mortgage debt where the corresponding housing asset is held in a family trust not included in the wealth. Does it really seem likely that anybody would lend over $100,000 to those unlikely to be in a position to be able to pay it off? 

Monday 17 December 2018

Where's my hockey rink?

Treasury has updated its CBAx cost-benefit analysis tool to bring in some of the measures from its new Living Standards Framework.

And I think I can get a proper* hockey rink out of it.

Let's walk it through.

According to the CBAx tool, being able to express your cultural identity is worth $9,563 per year per point of increase on a four-point scale. Canadians living in the Wellington region would enjoy at least a one-point increase in their ability to express their cultural identity if there were a hockey rink here. It's plausible that it could even increase from zero to four. Let's call it only a two-point increase, to err on the side of being conservative here. So that's then about $19,000 per Canadian per year. Over a 20 year lifespan of a rink, at a 6% discount rate, that's worth $218,000 per Canadian.

Winnipeg's MTS Place cost $134m CAD - let's call it $150 million NZD.

But wait! I've forgotten the deadweight cost of taxes needed to fund the thing. That $150m rink really costs $180m. And projects like these always have cost blow-outs. Let's say $250m all-up to be conservative.

That really isn't a problem that cannot be surmounted by CBAx and the World-Leading Living Standards Framework™.

You see, all season ticket holders, and anyone going to more than three games per year, would also be a member of the Canadian Hockey Club. Being a member of a club, per membership, is worth $2,536 per year in Treasury's new framework. And gaining a friend is worth $592 per friend gained. Everyone who joins Hockey Club would be making at least two friends. So we're already at an additional $3,720 per year - if we conservatively estimate only two new friends. Friends!

And let's say that one Canadian in ten would report a one point reduction in their feelings of loneliness. A one point reduction in loneliness, on a 0-4 scale, is worth $17,633 per year. Since we are being very conservative here and assuming that only one Canadian in ten would experience that reduction in loneliness, that's $1,763 per Canadian on average.

So we've already added another $5,483 in benefits per person per year - in addition to the $19,000 in cultural identity benefits. So $24,500 in annual benefits - capitalised over 20 years, that's $281,000 per Canadian.

If there are at least 890 Canadians in the Wellington area, (let's round up to 900 Canadians - again to err on the side of being very conservative here), then building us a hockey rink totally makes sense. All of my assumptions have been conservative here too. And I've not considered the benefits to others in Wellington. For example:

  • Hockey improves mental health. Every one point change in mental health (on a 100-point scale) is worth $4,608 per year. Really, claiming a point or two increase on this one across a big enough population can justify just about anything. 
  • People could go skating and become more fit. Every one point improvement in physical health (on a 100-point scale) is worth $1,158 per year. 
I wanted to check whether there were enough Canadians in Wellington to justify the programme, but all the circuits at Stats are occupied in calculating our spiritual health - "just coping with that problem right now, and wow, it's a biggy! Be with you in a while." 

Anyway, I think we can easily conclude that Treasury's World-Leading Living Standards Framework™ completely overturns what every other economist in the world has ever concluded on the economics of stadiums. We've just shown it with a series of very conservative assumptions. 

Oh - and the memberships thing also means that the government should pay my membership fees at the Wellington Club if I would otherwise let my membership lapse because of the cost. The benefits outweigh the club fees. 

* Hockey is played on ice with skates. It shouldn't need an 'ice' prefix to distinguish it from that NZ game played on grass with comically tiny sticks.

Friday 14 December 2018


The murder of Grace Millane is a tragedy.

As much as a number of commentators and one justice minister like to paint it thus, the apparent flouting of a suppression order that has revealed the identity of her alleged killer is not.

Frustrating? Yes. A challenge to our slow-moving justice system? Maybe. But certainly no tragedy.


Many will struggle to remember both by the time of the  trial, probably at least 12 months away. Some may even have forgotten by the time the 26-year-old alleged killer makes his next appearance, when suppression is likely to be dropped.

It's worth remembering, too, that the murder of tourists in this country, and associated overseas interest, is still rare. Suppression is observed in the great majority of cases that make their way through our courts.

Those relaxed about the impact of such indiscretions on justice also have evidence to back their ambivalence.

Law expert Warren Young and others researched such influences on juries in 2001, on behalf of the Law Commission.

They concluded that "publicity both before and during the trial currently has little, if any, effect on jurors".

So there is every reason to believe that, despite the level of interest in this case, justice will be served.

Another Law Commission report, 2009's Suppressing Names and Evidence, suggested the issuing of orders to force internet providers to remove information in breach of suppression orders.

But if the internet is a new frontier, then social media is the wild, wild west: once the horse has bolted, it's next to impossible to bring it under control. Even after a person is named, officially, people may be able to track their footprints through Facebook, Google and other sites.

So the challenge is significant, perhaps even insurmountable.

One that our justice system may have to live with, but one we are confident it will survive.
Not only does the editorial make sense, it also links through to the cited work. Many kudos.

Law and regulation always has to be able to respond to large cost shifts in the underlying environment. Suppression orders were pretty easy in the 80s. Anyone who might report on the trial would know that the order was in place. The number of media outlets was limited. And when you needed to get a permission note from Reserve Bank to get the foreign currency to subscribe to a foreign newspaper or magazine that might show up a few weeks after publication - risks that way were pretty trivial.

All of that would lean toward relatively liberal use of suppression orders. If the judge thought that there was at least some benefit in it, enforcement costs weren't much worth worrying about. Enforcement was easy. So the orders could be used in a broader range of cases.

Susie Ferguson's interview this morning with Bar Association's Jonathan Eaton QC had the Bar Association wanting strong enforcement of the existing rules without regard to the tech change that's happened. Susie's questions were great. But Eaton seemed to be expecting the impossible. Google's said that they were never notified about the order; Eaton imagined a world in which Google would somehow back-check, in every jurisdiction in the world, for each and every court case ever as they came up and ongoing in case the situation changed, whether there were a suppression order in place so they could make take the appropriate measures. That seems ...nuts.

Maybe there's some tech way around it, where courts would put suppression orders up into a central repository that were machine readable and Google (and others) could have a running check on that list.

Australia currently has a suppression order out on the verdict in the trial of an Australian high-level Catholic official (he's guilty).  New Zealand media's reported broadly on it; I even got a push notification on it from the Washington Post. There'll always be a way for Australians to read that stuff. And are newspaper apps supposed to run a GPS check on where the phone's owner currently is located before running a push notification? It's just dumb to expect it. It would be completely unreasonable to expect Google's Blogger to be able to tell what the trial is at the start of this paragraph and block it for Australian readers too. And given that mess, it is an absolute nonsense that Australian media has to censor the verdict. Some folks just aren't living in the real world.

Some bottom lines then:
  • Tech change means the costs of implementing a suppression order in high profile cases are very high. Courts should then be more reluctant to issue them than they were in a prior era when those costs were lower.
  • If they want these things to have half a chance of working, they need to figure out the tech of getting a repository of decisions rather than expecting every platform to be watching every court case for every change in whether a suppression order is in place or not. And that's something that should be set through international cooperation so there's one repository for the things using a common standard rather than a pile of them. 
  • And even if the order doesn't work, it's not much of a worry given LC's work on whether jurors are prejudiced by it. That's good, because it is impossible for an order to really work unless it is enforced globally. VPNs exist. And even if platforms like Twitter or Facebook tried to geoblock particular key words, there are a billion ways around it. China has an army of internet censors trying to keep up with the ways that social media users develop euphemisms for things they're not supposed to say.

Micro costs of macro prudence

A few months ago, I was at one of the typical Productivity Hub roundtable sessions where Wellington-types mull over why something in markets isn't working the way they'd like and how they could nudge markets around to improve things.

That time, it was about why there seemed to be problems in small business access to credit. After folks went round for a while about the purported failures, I asked whether anybody had looked at whether the macroprudential rules that came in during the GFC had killed off the financing tier that some small businesses had relied on. Remember how, before the SCF collapse and bailout, there used to be lots of ads for finance companies in the papers inviting retail investment? They'd take the money, invest it across a range of things, and pay a return to the bond investors so long as they didn't tank in the interim. Doesn't much happen now.

There was an awkward silence while folks considered whether Wellington might have caused the problem they were mulling over. Then after a "Nah, nobody's checked that", folks went back to more comfortable worthy suggestions for Making Markets Work Better Through Government.

I was reminded of it in this week's VOX CEPR Policy Portal piece. And here's the underlying IMF paper:
Combining balance sheet data on 900,000 firms from 48 countries with information on the adoption of macroprudential policies during 2003-2011, we find that these policies are associated with lower credit growth. These effects are especially significant for micro, small and medium enterprises (MSMEs) and young firms that, according to the literature, are more financially constrained and bank dependent. Among MSMEs and young firms, those with weaker balance sheets exhibit lower credit growth in conjunction with the adoption of macroprudential policies, suggesting that these policies can enhance financial stability. Finally, our results show that macroprudential policies have real effects, as they are associated with lower investment and sales growth.
So macroprudential regs reduce systematic risk but also make it tough for weaker firms to get credit.

Update: And remember that RBNZ is looking for funding for more people to do prudential regulation. They need more expertise in that area. Hopefully they're looking to improve the quality rather than the quantity margin on prudential regulation.

Thursday 13 December 2018

Back to the sweet sweet bog

For the past several years, the public health crowd has brushed off John Gibson's work on sugar taxes by saying that they don't worry about things that aren't in refereed journals.

It takes a lot longer to get things published in economics journals than in public health. Inaccuracies in public health work can then go around the world's newspapers several times before economics starts testing the more robust work in series of departmental seminars before sending the revised and improved draft off to a journal.

Gibson's paper with Bonggeun Kim is now up in the Journal of Development Economics and is ungated for the next month via this link

Here's the abstract, which won't be unfamiliar to those who've here been following the debate.
Estimating potential effects of price reforms is a key issue for many developing countries. Demand studies increasingly use household survey data on budget shares, which vary with quantity, price, and quality. If quality response to price is ignored, estimated price elasticities of quantity demand conflate responses on quantity and quality margins. Our review finds over 80% of published studies using budget shares from household survey data have this error. We use survey data from Vietnam, with prices and qualities observed over space, to directly estimate the price elasticity of quality. This is much larger than what is derived from the income elasticity of quality, based on the Deaton (1988) separability restrictions. Across the 45 items we study, the own-price elasticity of quantity demand is overstated by a factor of four, on average, if the response of quality to price is ignored.
The paper covers complicated technical issues in simple language. Economists are sometimes stuck with household expenditure data that only says how much a household spent in total on a product category over the past week, fortnight or month. If they want to know how demand responds to price changes, they have a problem because that data does not say how much was purchased or the price at which things were purchased. The data only says what total expenditure was. Total spending could change because quantities changed, or because
These issues were recognized, and potentially solved, thirty years ago in a set of papers by Angus Deaton (198719881990). Deaton derived the response of quality to price so as to isolate quantity demand elasticities, without needing price data. He assumed weakly separable preferences so that unobserved effects of price on quality could be derived from income elasticities of quality and quantity. Intuitively, by forcing the effect of price on quality to operate as an income effect, Deaton leveraged off what household surveys are good at – measuring incomes or expenditures – to get at what they are bad at or rarely do, which is measuring local prices.1 If one had a household survey with good measures of local prices, and with the usual data on food group expenditures and quantities, one could directly estimate the effect of price on quality by using unit values to indicate consumer quality choice (because the unit value is the product of price and quality). Indeed, Deaton (1990, p.302) concluded that it “would be extremely desirable to have direct measures of market prices against which this method could be tested.”
Gibson and Kim have data allowing this testing. And the testing shows us that weak separability fails, so back to the bog:
Our main title deliberately copies Deaton (1988) because in our view unidentified quality responses are still biasing price elasticities of quantity demand estimated from survey data. Our sub-title is from Gordon Tullock (1985, p.262) describing his role in an intellectual debate:
“… my role in this controversy is to watch people trying to get out of the swamp and then push them back in. Clearly, my role is not a constructive one, but nevertheless, I feel it necessary.”

Our contribution may be viewed similarly; before Deaton, economists used unit values as if they were prices when estimating elasticities of quantity demand. They were in a bog where quality and quantity effects intermixed. Deaton found a way out, pulling himself up just by the bootstraps of separability restrictions, with no firm ground (good price data) in sight.7Standing on firm ground now, with good data on local prices and on consumer's choice of quality, we are pushing people back into the bog by showing that these separability restrictions do not hold. More generally, any model that assumes that the price and income elasticity of quality are closely related is unlikely to hold. The necessary role we play, even if not a constructive one, is to show that we are still bogged down; many estimates of the effect of price on quantity are instead some murky mix of quality and quantity responses. Our defence for our role is that it is only by realizing that we are still bogged that the value of firm ground (good data on local prices and on quality) becomes clear. In our opinion, there will be little headway in using household survey data to accurately estimate quantity responses to spatial price variation until better data on local prices and qualities are collected, so that responses on both the quantity and quality margins can be directly estimated.
I love Tullock references - this one's to Tullock's pushing people back into the bog by showing that explanations around the 'why so little rent-seeking' problem were wanting. Gibson and Kim show that Deaton's path out of the problem doesn't work.

Public Health critiques around hierarchies of evidence and that this is just one paper compared to dozens of other papers so we should look to metastudies - they completely misunderstand the nature of Gibson's contribution here, whether through ignorance or stubbornness, I don't know. Gibson and Kim show that the dozens of other papers that use the Deaton method for estimating price elasticities out of household survey measures are systematically incorrect. Doing a metastudy of papers that have a systematic error in method will not get you a correct answer.
These large biases, from either ignoring quality response to price, or from restricting that response to be what weak separability allows, are in line with the few prior studies on the quality response issue. In the first of these, quantity demand elasticities were inflated to an average of 250% of their unrestricted value, if quality response is ignored (McKelvey, 2011). While that evidence was just for six broad food groups, our results are much the same if based on broad consumption groups or narrower price survey items, so this magnitude of bias may hold more widely. A similar level of bias is seen in studies of soft drinks demand using the unrestricted method and the standard price method. In Melanesia, where spatial price variation is high and product differentiation of soft drinks more limited, the quantity demand elasticities are overstated two-to three-fold if quality response to price is ignored (Gibson and Romeo, 2017). In Mexico, where there is less spatial price variation and more quality variation the bias is four-fold (Andalón and Gibson, 2017).

These quality responses may undermine price policies that aim to reduce consumption of unhealthy items. For example, quantity demand elasticities that ignore the quality response to price are used by Grogger (2017), to forecast that Mexico's peso per liter tax on soft drinks will reduce steady state body-mass index (BMI) of Mexicans by up to 1.8%, which is enough to provide some health benefit. However, if quality downgrading in response to tax-induced price rises is accounted for, the average BMI will fall by just 1/200th (Andalón and Gibson, 2017). This is salient to Vietnam, where a special consumption tax of 10% on soft drinks, instant tea and flavored milk has been proposed by the finance ministry as a way to reduce the health burden of high sugar intakes. However, if adjustment to higher prices is on the quality margin and quantity consumed falls only a little, as in Mexico, this proposed tax will be largely ineffective in achieving health objectives.
The implications are rather broader than sugar tax.
Extrapolating from income effects to price effects may be unwise, as seen with failure of the weak separability assumptions, but a prior debate in development economics about effects of income on nutrition is germane to this discussion. Early studies derived indirect estimates of the income elasticity of calories from food expenditure data, assuming that higher spending on a food group meant proportionately more nutrients. This ignored within-group quality substitution, and later studies showed that extra spending on food as incomes rose went on attributes other than food quantity. Writing about poor people with rising income, Behrman et al. (1988, p.308) noted that:
“… at the margin they concentrate on food attributes other than nutrients – taste, appearance, odor, degree of processing, variety, status – that are not necessarily highly positively correlated with nutritive value.”

In perhaps the same way that policy makers learnt that effects of income changes on nutrients are mediated by within-group quality substitution, so too may they need to learn that effects of price changes on quantities can likewise be mediated by responses on the quality margin.
While Gibson and Kim don't cover it here, because it is obvious, the analysis applies whether a soda tax or sugar tax is ad valorem or a per-unit excise. Whatever you thought the demand response to your price increase was going to be, it'll be smaller to the extent that your elasticity estimates conflate quantity and quality adjustments.

The only case where this won't be true is for people who are already only consuming product that is at the lowest quality point - and you'd probably then want to have an elasticity estimate for that specific cohort anyway rather than assuming that the estimates that apply elsewhere also apply to that cohort. 

Wednesday 12 December 2018

Hard Spirits

I write a lot of columns. I rather like how this one turned out. It was up at Newsroom Pro yesterday; it's ungated here, and copied below.
Hard Spirits

It was during the discussions of measuring spiritual capital that the ghost of Sir John James Cowperthwaite hovered near. The shade whispered in my ear, “When I was Financial Secretary of Hong Kong, I refused to collect economic statistics for London. Why? For fear that I might be forced to do something about them.”

I wish more people at last week’s Indicators Aotearoa New Zealand indicator selection event had been able to hear him.

Statistics New Zealand’s Indicators Aotearoa project aims to build a comprehensive suite of (you guessed it) indicators measuring wellbeing. Where SNZ’s 2014 Progress Indicators Tool, now discontinued, included 16 measures ranging from adult educational attainment to distribution of selected native species, the new framework aims to include about 100.

The project is linked to Treasury’s Living Standards Framework and to the Government’s desire for more measures of wellbeing. Treasury has been developing its own suite of wellbeing indicators as part of the Living Standards Framework; whose indicators will reign supreme remains a bit up in the air. If Treasury’s framework and indicators wind up satisfying the Government’s wellbeing needs, then Statistics New Zealand’s indicators project – or at least those indicators that do not make it into Treasury’s framework – might yet follow its predecessor into the bin of discontinued data series.

So the bureaucratic stakes are high, as far as these things go.

Statistics New Zealand took a reasonable approach in trying to figure out which measures to include: they asked New Zealanders what wellbeing means to them in a series of workshops around the country, along with an online campaign. Then data experts tried to figure out which of those might possibly be meaningfully quantified among those that are not already measured.

That led to a long list of indicators to be pared down at last week’s indicator selection event, where I was visited by the ghost of Sir Cowperthwaite.

Inputs, outcomes and A3 posters

There’s a management truism that what gets measured gets managed, and what doesn’t get measured doesn’t get managed. Choosing the right measures then matters, especially if these kinds of indicators get targeted by a Government desperate to be seen to be improving our wellbeing.

The big headline measures should reflect final outcomes (like healthy life expectancy) rather than intermediate outcomes (like cancer rates) or inputs (like the proportion of people using sunscreen appropriately).

If there are lots of intermediate outcomes that are all captured by the final outcome, promoting intermediate outcomes into being final outcomes is a very bad idea. It invites paying more attention to those selected measures rather than to other ones that Government might more usefully target. In the example above, if an intervention targeting diabetes did more to improve healthy life expectancy than an intervention targeting cancer, the latter might nevertheless be preferred if cancer is a headline statistic and diabetes is not.

Turning inputs into headline statistics is at least as bad, for similar reasons. Inputs might be detrimental to some measures of wellbeing, but useful for others. An index of health-related behaviours, counting exercise, diet, and various consumption choices would be a poor one to include in headline statistics on health. To the extent that those behaviours improve health, they are already captured in the healthy life expectancy statistic. But including them as a headline measure risks targeting them for improvement in ways that may hurt other measures of wellbeing.

And, finally, the outcome measures chosen should be things that the Government might plausibly have any business doing something about. We might wish to be careful about what gets measured lest we be managed into improving our wellbeing in ways we might not welcome.

To take an obvious example, I have yet to see a survey of wellbeing or happiness that, if it included satisfaction with one’s sex life as a measure, failed to find that it mattered considerably. If any participant in Statistics New Zealand’s workshops had been honest and reported that sexual fulfilment mattered greatly in their wellbeing, their input did not make it into the workshop. Really, if the number proved to be a bit softer than we might have wanted, what on earth should policy seek to do about it? All options seem atrocious.

Unfortunately, the process of outcome selection at the Statistics New Zealand event was less fruitful than it should have been.

About 150 people attended at the Michael Fowler Centre.

Two hours of introductory remarks included a welcome from the Government statistician, a hand-clapping game, an invitation to check our privilege with the help of a bingo-sheet of privilege indicators like not being red-haired, and an extensive discussion from a PhD candidate who warned (among other things) about consulting with Ngāi Tahu on issues Māori because they are too corporate.

The two hours of introductory remarks did not include any discussion of the difference between final outcomes, intermediate outcomes, and inputs. Nor did it point out that current measures of inputs and intermediate outcomes would continue to be collected regardless of whether they became headline statistics.

After being encouraged to reflect on my privileges in being able-bodied, not hard of hearing, having a loud speaking voice, and not being red-headed, I joined a group of about 50 people who all had to stand for an hour around an A3 poster to discuss the proposed health indicators – before moving on to stand and talk around other A3 posters. Those unable to stand for extended periods were denied effective participation, as were those even of normal hearing if they were not very close to some of the very soft-spoken participants. Microphones and chairs may not have gone amiss, but I am probably in too privileged a position to have standing to comment on that. Our facilitator did his best but had a difficult task.

Because nobody had explained the difference between inputs and outcomes, or that failure to select particular measures did not mean they would cease to be measured, conversations had to keep coming back to those basic points. Every worthy-sounding input measure had its proponent for inclusion in the mix.

And many of the indicators seemed, well, problematic. Some outcomes were listed as “contributing unambiguously to progress” when they seemed manifestly contestable. For example, the response to the Initiative’s report last year lauding the merits of diversity and immigration suggested not everyone agrees diversity is a good thing. But appreciation of diversity was included as a potential outcome variable deemed to contribute unambiguously to progress. It seems a political argument to win rather than an outcome variable to technocratically maximise. If it were unambiguously associated with progress, we would have received less hate mail insisting on the demerits of diversity and that New Zealand needs fewer immigrants from culturally dissimilar places.

A dangerous thing to attempt to measure

Most contentious in the health discussion was the inclusion of spiritual health as a desirable outcome. And it is there that the spirit of Sir John James Cowperthwaite visited me – as I am a very spiritual person in my own utterly non-spiritual way.

Sir John is generally credited with Hong Kong’s transformation from a desolate place at the end of the Second World War into about the richest and most economically free city in the world. He credited some of his success to his refusal to provide Whitehall with the statistics they might attempt to use in applying Atlee-style management to the Hong Kong economy. His policy of positive non-interventionism certainly outperformed the UK’s more scientific-looking approach built on a mountain of economic statistics.

His spirit warned me that spiritual health is a dangerous thing to even attempt to measure as a headline wellbeing outcome, even if it is deeply meaningful to many communities. There is the obvious problem that atheists like me would not have a clue how to begin answering a survey question on whether we are spiritually fulfilled as the area, to those who share my views, is meaningless. But there is the far worse problem that what gets measured may be managed. If a headline outcome measure of spiritual fulfilment were half of what people expected, or double(!), what on earth should the Government ever do about it? The measure invites management. And any attempt at management would be worse than the measured problem.

I do not envy Statistics New Zealand the task ahead of them. The workshop will not have been as useful as it could have been in selecting an appropriate set of indicators. Will they weigh more heavily what makes sense as outcome measures, or what participants claimed to want? Either way, they will be stuck with trying to measure the things – while being unsure whether much of the project will be superseded by Treasury’s work. All this while battling with what seems an ever-worsening problem in getting the Census out.

I wish Statistics New Zealand good spiritual health in the months ahead, as they will need it.
I feel bad for Stats.

They took a fair bit of stick about having us fill out a Privilege Bingo form. I think that we wound up doing the privilege bingo exercise by accident. They had time to fill when a video screwed up, and rather than fill the time with something useful (see the column), the team scheduled to give that presentation scrambled for something tangentially related to their having asked lots of different communities for input into their data exercise.

They were utterly utterly oblivious to that they were stepping in very firmly on one side of the culture wars, perhaps because those involved were so ensconced in one side of it that they couldn't anticipate that it was problematic. If it hadn't been a late fill-in, maybe somebody with some nous would have headed it off. But it wasn't completely last minute. They had had time to have photocopied Privilege Bingo sheets on everyone's seats along with the day's agenda.

And today, the Census problems alluded to in my penultimate paragraph hit the front page of the Dom and Radio NZ.

I wonder about the usefulness of the community consultation exercise. It is very very easy for those things to make it very hard for people with potentially dissenting views to say anything. Remember the kinds of preference falsification exercises that Timur Kuran talked about, and information cascades? I wonder if anybody at Stats reads this kind of stuff.

If you're in a room where a couple people have just made very clear that the inclusion of spiritual wellbeing is more than just important to them, that they'd take it as a slight to their entire way of being and sense of self if anyone challenged that measure's inclusion (like, not explicitly, but in the way they make the argument), how many would be happy to step in and challenge it? At the poster session, we could put stickers on particular measures. Before discussion, there were about 10 red stickers on spiritual health, suggesting ten people wanted it demoted, and 2 green stickers, suggesting two people disagreed with the red stickers. I spoke against the inclusion of the measure, then a couple of people spoke about how sacred the measure was to them, and not a single other person who had put up a red sticker was willing to say a word.

If there's just been strong discussion of the importance of kaitiakitanga in land management, would you be the one willing to say "You know, my family settled our farm a hundred years ago and has worked and loved that land for four generations. Shouldn't we count too?" Like, the two can work together, but it is hard to get to that kind of discussion.

A friend gave an analogy that seems to fit. Suppose you've got a juggler who's really really good at juggling five balls. The business-as-usual stuff at Census is the five balls - there is a pile of it. In a Census year, they have to juggle six or seven balls. And they do a great job of that too, because they can manage six or seven for a short period. But throw another ball in the mix and everything starts falling down because eight is way harder than five. I wonder whether the Indicators project isn't that deadly eighth ball.

Census woes

I don't know what normal people talk about at drinks. For the past few months, whenever I'd catch up with other economists over drinks, it's been rumours about just how bad things are at Census. 

The mess hit the front page of the Dom Post today.
Documents released under the Official Information Act show the department is planning for an August 2019 deadline, and continues to stare down a high risk it may not produce viable results.

A low turnout during Census 2018 has caused considerable pain for Stats NZ, which last week announced it would not make the third deadline it set - July 2019 - and promised to announce a release date in April.

Stats NZ failed to count an estimated one in 10 New Zealanders in Census 2018, previously providing an interim response rate of at least 90 per cent of individuals providing full or partial information.
Some tales have it that the incomplete census responses are on top of the 10% non-response rate rather than part of the 10%.
It was announced in June that other sources of Government data would be used to fill gaps in the results.

Both the actual number of responses received and the number of additional data for 2018 were redacted from documents, and Stats NZ declined to release them.

The documents show "high risks" Stats NZ continues to face include: the failure of methods to patch census data; census data failing to be fit for purpose, leading to "less than ideal" decisions being made; and a failure to provide information for the re-drawing of electorate boundaries.
I really hope that they're flagging interpolated data in the back-end systems so that analysts know whether they're dealing with generated regressors. So, for example, suppose you're trying to estimate the relationship between a bunch of admin back-end variables and some Census result. If the "Census" result was really generated by those back-end variables in the first place for some of the respondents, you're going to botch your analysis unless you know that that's gone on. You could just be rediscovering the algorithm used to generate the data rather than any real relationship.
Also described are a "severe incident", where some data fields on the census forms failed to cross over to another system and contributed to a month-long setback.
... BERL chief economist Dr Ganesh Nana said Stats NZ needed to "draw a line" under Census 2018.

"We all might have to put our hands together and say, 'Okay, the data won't be as rigorous and robust as previous census'. We might just have to admit that."

A standard census release would have BERL using the data intensively over a six-month period from about now, he said.

"That's clearly been put out of kilter this time around. The biggest concern for us is the timing of the data … the data is already starting to become stale."
And drinks chatter has speculation over whether the whole thing will yet require a do-over.

It isn't the first year that they've combined online and paper - I filled in my Census online in Christchurch last time. Quite why everything's gone wrong this time - I'd love to know.

I'd made brief allusion to the problems at Census in my column at NewsRoom yesterday ($). That column wasn't about Census but about the fun they're having with the Indicators Aotearoa project. Will post on that one soon.

Update: Ganesh Nana talks with Kathryn Ryan about it on Nine to Noon.

Update 2: Hoisted from the comments (on the mobile version, where I've not been able to get Disqus to work):
One big difference between this Census and the last one was that last time the online form was opt-in, but this time the paper form was opt-in. In 2013 everyone received a paper form, but people who registered could ignore that and do it online instead. In 2018 you had to register to receive a paper form. If you didn't get one in time then it was do it online or not at all. That's a big difference in practice and could be part of the reason for the problems. 

Tuesday 11 December 2018

The Globlish solution

A couple weeks ago, I'd suggested NCEA ought to make things simpler for students who do not know basic words like trivial by restricting itself to the 3000 most commonly used words.

Globlish takes it one step further by reducing the language to the 1500 most commonly used words and simplified grammatical structures. It is intended to be the minimal vocabulary necessary to conduct global business among people who are not native English speakers.

Just think about how New Zealand's measured literacy scores could improve if we restricted assessment to only those words that the students really need to know, instead of all those extra ones.

HT: Sam Hammond, who imagines a fun movie based on it. You should follow him on twitter.

Thursday 6 December 2018

Morning roundup

The morning worthies:

Wednesday 5 December 2018

Cannabis reform

Russell Brown's take on cannabis law reform over at RNZ:
Yet, the CRC may still learn lessons from what's happened abroad. Alison Holcomb, the criminal justice director of American Civil Liberties Union in Washington state, who helped design the successful initiative there seven years ago, says that "voters responded strongly to messages that reassured them about tight control of this novel policy experiment. Messages about freedom and individual rights fell flat. I continue to believe that acknowledging basic human nervousness about change is always important."

Tamar Todd, legal director of the Drug Policy Alliance, who jointly authored California's Proposition 64, says voters in her state wanted reassurance in detail. That's what they got: the full text of Proposition 64 ran to more than 100,000 words of legal and technical definitions and proposed amendments to laws and regulations, together making up the Adult Use of Marijuana Act.
This is one reason I think we should be modelling cannabis legalisation on our existing framework for supply and sale of spirits. There is no time before the referendum to write up the kind of detail provided in Prop 64. But we could adapt our existing framework.

Tuesday 4 December 2018

Stadium logic

Sporting events would create around 20 big occasions with any of the options. The covered stadium options enable many smaller events: music, conferences, other community events. These are forecast to produce small 'profits', as opposed to the uncovered option that would generate small 'losses'. A massive caveat is that when the 25-year life cycle of ALL of the options is considered, (big venues need substantial reviews every 25 years or so), all of them lose about $7-8m a year.

Hence the report favours the bigger spend, arguing it would create more activity, and lose about the same amount of money over the long term. It is worth stressing this again; even with the most optimistic predictions for stadium use in the report, the best-case scenario still forecasts it costing us $7-8m year; on top of the $474m construction cost.
They also hit all the points that should be hit around the difference between creating new events and shifting events from one place to another.

Every time a NIMBY cries...

The Court of Appeal has quashed resource consents granted for a $500 million development in Wellington at Shelly Bay.

The Court found the Wellington City Council made an error of law when determining whether or not to grant resource consent.

"As a result of the error, matters such as the environmental effects of the proposed development were not given appropriate consideration and weight by the Council," the judgement said.

The Wellington Company Limited applied for consent in September 2016.

The original proposal at Shelly Bay would have involved the construction of 12 apartment buildings hosting 280 apartments, 58 townhouses and 14 individual homes.

A 50-room boutique hotel, an aged care facility and buildings for commercial and community activities were also planned.
The original consent application was over two years ago. Everything since then has been consent process and litigation. 

Dom Post covers it here:
The Court found that the council had misinterpreted the Housing Accords and Special Housing Areas Act. The act had more permissive rules for housing consents, and was intended to be used to increase the supply of houses, and housing affordability, in areas with a housing shortage.

The city council did not notify the application for public consultation and did not hold a hearing.

The court said the council had used the housing accord law to effectively "neutralise" all other considerations. Had the correct approach been adopted there might have been a different outcome, the court said.

Usual planning considerations were still relevant and the Court of Appeal found the council failed to properly consider matters such as the preservation of the natural character of that part of the coast and the protection of historic heritage from inappropriate use and development.

The council had also used the need to increase housing supply to make a finding that the environment effects of the development were "no more than minor".

The aim of increasing housing was not logically relevant to deciding whether an environmental effect was more than minor, the court said.
 The SHAs were supposed to get around this kind of red tape.