Around the traps

Blogging has been light as other commitments have pressed, but you may have caught me around the traps:

• I pointed to Stats NZ's disaggregated CPI figures in this piece by Susan Edmunds over at Stuff. 

  "Statistics New Zealand recently began releasing cost-of-living statistics for different groups to take account of differences in spending patterns. Since they started doing that, they have found that increases in the cost of living have been most sharply felt by the poorest because tobacco excise increases, petrol price increases and housing costs there have the worst effects. And these are areas directly under the government's control. Ceasing punitive tobacco excise increases and fixing the regulatory settings to allow new housing be built would substantially affect living costs for the poorest."

  You can check out Stats' shiny Living Cost Explorer right here.


• I also pointed to the consequences of failing to adjust NZ Superannuation in recognition of rising life expectancy in another article over at Stuff:

  Crampton said, if things were left as they were, it would not mean fiscal collapse.
  
  But it would mean government budgets were increasingly skewed to helping those who are older rather than those in greater need.
  
  "And, the longer we wait to make changes, the harder it will be to make changes because of changes in voter demographics. I view that as inequitable, but others can reach different conclusions."

• At the Herald, I suggested that reductions in the projected number of superyachts attending the America's Cup would worsen the already dubious case for government funding big sporting events. Tom Dillane there was only able to use a smaller part of my more verbose commentary, so here's the full text:

  “The case for government funding of the America’s Cup was always rather weak. After correction to some errors in the initial estimates, MBIE reported an estimated benefit-to-cost ratio only slightly higher than 1:1, with a range from 0.997 to 1.14. So every dollar of estimated benefit was matched, nearly one-to-one, for a dollar of cost. We can worry that these kinds of estimates are often optimistic about the benefits of these kinds of events. But the estimates would have been based on an expected number of visitors. If fewer superyachts are coming in for the event, then the benefits of the event will be a bit lower than expected. If we think that events like the America’s Cup do more to change the timing of just when tourists and superyachts come to New Zealand than to affect whether they ever come to New Zealand, then we should perhaps be less worried about things – except that that would also mean that the initial benefit estimates were always overstated.”
  
  “We should be sceptical that the path to national riches lies through public subsidy of large sporting events. If New Zealand taxpayers are happy for the government to spend a lot of money on what is effectively a big party in Auckland, then the event should be funded on that basis. But we should not delude ourselves that big parties attended by some foreign tourists are really investments.”




• Finally, I had a chat with Heather du Plessis-Allan about New Zealand's mess of regulations around heritage-listed buildings, that too often make it just too hard to own the things.

Low low prices

TV prices are so low that StatsNZ has to rebase the CPI. They normalise things in the base year to have an index figure of 1000. So the price of TVs and of cell phones in 2006 was 1000. Cell phones have dropped to 66; TVs have dropped to 60. Why? Quality adjustment.

The TV that cost thousands in the 2000s now costs maybe a couple hundred. Sure, you can still spend $5000 on a TV. But the TV you'd have spent $3000 on in 2006 is a hundred-dollar model now.

The CPI rebase
This section explains the reasons for the upcoming rebase of the CPI and FPI indexes and when it will be applied. For those of you interested in quality adjustment, or who use the index numbers in your models – read on.

We have been reviewing and reweighting the CPI basket of goods and services over the last six months. Part of the review also involves determining whether there is a need to rebase the indexes. Low inflation has led to relatively  small differences from the expression base of 1000 in the past decade, so we have kept the 2006 base period. However, due to increasing quality in some areas, we have been reporting falling prices in these areas of the index and some are now approaching zero. To adjust for this, we need to reset all of the CPI indexes back to 1000.

Technological advancements affect CPI index numbers
An important part of the CPI review is to ensure we account for improvements in technology. As it improves things get better, faster, flashier, and cheaper. The prices of mobile phones, cameras, and videos, for example, have fallen steadily over the past decade. The average price of a flat panel TV in 2006 was a staggering $3,382. If that same TV was still available today it would likely cost around $200. Similarly, a digital camera bought in 2006 would have cost about $500, now you don’t even have to buy one, they’re built in to your mobile phone and they have better specs. In the day of the ‘brick’, circa 1983, the luxury of being able to call someone while out and about would have cost roughly US$3,995 (see 40 years of the mobile phone: Top 20 facts). By 2006 the average price of a mobile phone was $249; this phone in today’s market would slip into our pocket for a mere $16.33.

When consumers get a better quality product or one with more features for the same price, this counts as a price drop in the CPI. We fix the quality of these types of items in the CPI so that we can price the same item through time.

Figure 1 shows the indexes for TVs and mobile phones for 2006–17.Figure 2 shows the indexes for the recreation and culture subgroup for the period 2006–17.Keep your eyes peeled for the updated CPI basket in January We update the CPI basket every three years, with additions and removals reflecting changes to consumer spending patterns. For example, DVD players were state of the art at the turn of the century and were added to the basket in 2002, but have been replaced by computers in the ‘cloud’ and were removed from the basket in 2008. Video-cassette recorders nearly held on that long and were only removed in 2006, signalling the end of an era of heading to the video shop on a Friday night. Mobile phones were added to the basket in 2002 and because they are still popular, will likely be there for a long time.

New CPI base period Due to all these technological advances, some of our indexes are now almost zero. From 2006, price falls recorded in the CPI for cellphones and accessories have moved the index for this item from 1000 to 66. The index for TVs is now at 60. Because these indexes are so close to zero we are losing accuracy in the CPI.

To adjust for these large differences, we will rebase the CPI by resetting all the indexes to a base period of June 2017 = 1000.

HT:  Newsroom Pro's 8 things at 8am

Health CPI

So Labour and National are scrapping over whether National's increased or decreased the health budget. There's no question that health budgets are well up since National took office, whether in per capita terms, real terms, nominal terms, or real per capita terms. Population's up about 10% since 2008; health budgets in CPI-adjusted terms are up about 29% since 2008. So real per capital spending has to be up.

But the more interesting question's on how to adjust health costs for CPI.

StatsNZ has a sub-index on health costs. And that's shown substantial cost inflation - well above CPI in some categories. But should health spending be adjusted for that sub-index? Let's look at what's in it.

Position in the CPI structure

The health group of the New Zealand Household Expenditure Classification represented 5.09 percent of the CPI at the June 2008 quarter.

Table 1

Expenditure weight for healthJune 2008 quarter
Group, subgroup, or class	Level	Weight (percent)	Examples of items within class
Health	Group	5.09
Medical products, appliances, and equipment	Subgroup	0.98
Pharmaceutical products	Class	0.61	Prescription medicines, oral contraceptives, and over-the-counter products such as painkillers, cough and cold preparations, sunscreen, and vitamins
Other medical products	Class	0.03	Bandages and contraceptive supplies (other than oral contraceptives)
Therapeutic appliances and equipment	Class	0.34	Corrective glasses and contact lenses
Out-patient services	Subgroup	3.32
Medical services	Class	1.97	General practitioner, specialist, and optometrist consultation services
Dental services	Class	0.94	Dental examination, filling, and denture services
Paramedical services	Class	0.41	Medical laboratory (eg scanning) services
Hospital services	Subgroup	0.78
Hospital services	Class	0.78	Private hospital services

Let's assume that this is all still the same in terms of weightings and definitions as it was in 2008.

If you flip over to the figures, Infoshare has 2006Q2 as base set at 1000 for the level 2 subgroups. Medical products, appliances and equipment had dropped to 890 by 2008Q3 (National comes in) and is now at 1030. Basically, no inflation there. This group covers pharmaceuticals, medical products, supplies and equipment.

Out-patient services were at 1035 in 2008 and are now 1356. This covers things like GP fees, specialist fees, optometrist fees, dentist fees, and medical lab fees. It's mostly GP, specialist and optometrist fees. Suppose that the government wanted to concentrate health spending on lower income cohorts and so reduced the capitation fee paid to GPs while increasing the range and subsidy on the Community Services Card - with the whole thing being revenue neutral if clinics upped the GP consultation fee. That would increase measured out-patient service cost inflation but it would be silly to increase health budgets on the back of that inflation.

Similarly, if specialists decided to charge private clients more, that affects measured inflation but only affects government health provision costs if they also change how they bill DHBs.

Hospital services rose to 1187 by 2008 and to 1527 by 2017. This reflects costs at private hospitals. If richer people demanded nicer services at private hospitals that would show up as cost inflation but wouldn't affect costs at public hospitals - but could reflect that public hospitals are declining in quality of amenities relative to private ones.

Bottom line: the health component of the CPI has increased far more quickly than other parts of the CPI, driven by increases in private hospital costs and out-patient services. It's very likely that the cost of providing health services in the public system has outpaced CPI inflation, but I'd be reluctant to just use the health services subcomponent of CPI for a weighting there.

Tobacco tax, CPI, and living costs at the bottom

Stats NZ has been putting together new price indices that track the prices of the bundles of goods commonly bought by people in different income and expenditure cohorts: Household Living-Costs Price Indexes. The basic method and background's here. 

The latest release found that higher cigarette and tobacco costs are hitting beneficiary households. The Stats NZ release makes it pretty obvious.

Beneficiaries experienced the highest inflation in the March 2017 quarter, Stats NZ said today. Their overall costs rose 1.4 percent, almost three times the rate of inflation experienced by the biggest spenders group (up 0.5 percent).

"Higher costs for cigarettes and tobacco had a greater effect on beneficiaries. About 5 percent of their spending went up in smoke, proportionally more than most other types of households spent," consumer prices acting manager Nicola Growden said.

Higher rents, which make up one-third of their total spending, also had a greater effect on beneficiaries.

Government's been hiking tobacco taxes. That it would be highly regressive is no surprise. Government restrictions on new building drive housing affordability problems too. Government transfers a lot of money to the poor, but also makes things pretty expensive for that cohort.

I rather like the chart we had in Jenesa's report, Health of the State.

Update: Reader mailbag brings me this rather nice chart from the Stats release. Happy coincidence as the email came in two minutes prior to this morning's 7am queued post.

Informed Reader's conclusion was similar to mine:

The most interesting are beneficiaries. The reason they have higher inflation is:

• Cigarettes and tobacco that is mostly because of the tax regime

• Rents whose cost is mostly covered by accommodation supplement and Temporary Additional Support (despite the name it is generally not temporary)

• Energy that is high because of the policy to subsidise aluminium smelting.

Most of the real difference is down to government. This makes sense if you think about the economics. People with a tighter budget constraint are going to be more price inelastic, so the only place you would see prices rise more for the poor are where prices are not determined by a market.
 Income quintile 1 is more complicated, not least because it includes many people who have income that statistics New Zealand surveys don’t pick up e.g. people starting a business who have “no income”. If you look at Bryan Perry’s work on incomes, you will find he explicitly warns against using lowest decile data. But even here, if you take out the ones for beneficiaries, you are left with property rates from local government and interest which you could reasonably suspect is being paid by those who lenders are prepared to lend money to (e.g. people starting a business who have “no income”...)

CPI visualizer

Keith Ng rightly won an award for this one.

I can't embed it, but it lets you easily check out component increases in the CPI back through June 2006. So if you're a data visualization geek, or you want to better understand the components of price level changes over time, hit the link.

Here's one example. Just check the changes in the prices of tobacco products since the anti-Nanny State National Party took over from those horrible "I'll force you to do what's good for you" Labourites (hint - the first bar on the right to pass the 4% line is Q1 2009):

The image doesn't do it justice; the link above gets all the interactive stuff.

I'm particularly recommending it to our Econ 104 and Econ 105 lecturers for teaching our students about the CPI. Thanks Keith!