Ladies and Gentlemen... the Greens' would-be Finance Minister

Russel Norman, recently ruled out as Finance Minister in any Labour-Green coalition by Labour, tweets from the Finance Committee:

Sitting in Finance Cttee listening to Treasury officials describe tax as a 'burden'. Right wing political ideology runs deep in Treasury!
— Russel Norman (@RusselNorman) September 17, 2013

Tax is price we pay to live in a civilised society. If Treasury thinks it's a 'burden', perhaps they'd suggest abolition of public health.
— Russel Norman (@RusselNorman) September 17, 2013

Taxes are a bad, public services are a good. Saying the first doesn't mean denying the second.

More importantly, economists use the word 'burden' in a particular way. A few useful notes about Principles-level (maybe intermediate) economics for someone who thinks himself qualified to be finance minister:

• 'Burden' measures the total cost of a tax. The 'excess burden' is the amount by which the cost of a tax exceeds the amount collected. Treasury tends to reckon that excess burden is around 20%: it costs us about $1.20 to raise $1.00 in tax. The $1.00 raised is a transfer from the public to the government; the $0.20 is pure loss due to distortions in economic activity consequent to increases in our current mix of taxes.
• Tax incidence theory is important: it tells us who bears the burden of any particular tax. Suppose we wanted to add another 5% compulsory Kiwisaver contribution. The 'burden' of the tax would fall on both workers and on employers with the precise mix depending on how employers and employees change their labour demand and labour supply with changes in wages: it doesn't much matter whether we say that employers have to pay it or whether employees have to pay it. Regardless of statutory incidence, economic incidence - the burden - will remain the same. Meteria Turei understood this when she said that the accommodation supplement paid to tenants is largely a subsidy for landlords. Alas, public understanding of such things is imperfect, allowing for shenanigans where measures imposing burdens on one group are framed as costing somebody else instead.
• If a genie appeared able to provide public health services, for free, this would be a good thing, right? It's impossible, but it would be good. The services paid for by taxes are good, the taxes are bad. We need to be sure that the value delivered by services are greater than the burden imposed by the tax. At current measures of excess burden, a project must return at least $1.20 for every dollar in spending. 

Russel Norman suggests only "right wing" economists talk about tax burden. Here is a JSTOR search on "tax burden". There are 61 pages of search results with 100 results per page. Item number 177 on a date-sorted list is famous Right Wing Economist John Maynard Keynes discussing the Colwyn Report on Natinoal Debt and Taxation. Item 398 is rabid right-winger Nicholas Kaldor's call for wage subsidies to reduce unemployment (1936).

Burden is just the term used by economists to describe the cost of the tax and to help sort out the difference between statutory and economic incidence. Like "While X writes the cheque to IRD, the burden of the tax falls on Y and Z." That's it. It's the standard term used in the main texts to describe this thing. Richard Musgrave (centre, maybe centre-left) uses it. James Buchanan (right) uses it. Pick a random public finance text, you'll find "tax burden" or "excess burden" somewhere in it.

Update: egads, it gets worse. Lance Wiggs tries explaining that it's just a word we use. Russel Norman replies:

@lancewiggs Taxes do change behaviour. To call it a burden or distortionary does buy in to general equilibrium model w all its problems
— Russel Norman (@RusselNorman) September 18, 2013

Update 2: this is way too funny. A Twitter correspondent points me to two press releases by Russel Norman.

First:

"It's not fair to expect income-earning New Zealanders to carry a disproportionate share of the tax burden while some of New Zealand's wealthiest individuals pay none," said Green Party Co-Leader Russel Norman.

Second:

Unlike the National Government that has chosen to shift the tax burden on to the lowest paid New Zealanders, our tax changes would focus on those not currently paying their fair share.

Accommodation incidence

Poor people in New Zealand are eligible for a subsidy towards accommodation expenses: the Accommodation Supplement.

The Greens' Metiria Turei calls the supplement a "landlord subsidy" and points to it as part of the general problem of housing affordability: it pushes up the price of housing.

In the current state of the world, she's mostly right. Given a near-vertical supply curve for housing, because land use policy in New Zealand is a complete mess, anything that subsidizes demand mostly gets capitalized into land prices. So it is a subsidy to landlords, mostly via capital gains. 

If developers were allowed to build new housing in response to demand, either by increasing density or by building out, the incidence of the subsidy would be entirely different. 

In the graphs above, we map out supply and demand for rental accommodation among low-income cohorts. 

D represents their demand curve. It slopes down for the usual reasons: when housing is expensive, people demand less of it. Think less of the grosser forms of substitution, like homelessness, but rather of the intensity of rental use: families doubling up in accommodation units and many kids sharing bedrooms. When housing is expensive, people double up; when it's cheap, we have more space per renter. D+A gives demand when low-income renters have access to the accommodation supplement: the vertical distance between D and D+A is the level of the subsidy.

S is the supply curve: the price at which developers bring new low-income units onto the market. It slopes up as well. But, the slope differs between the graph at left and the one at right. On the left, supply is relatively inelastic. And that's the current state of the world in New Zealand. It is illegal to provide low income housing, or any kind of housing, cheaply. Councils restrict the supply of land such that its price is bid up. And, they make it illegal to put self-contained flats into existing homes: one of the quickest and least expensive ways of expanding the supply of more affordable units. And in that state of the world, the accommodation supplement does little to expand access to accommodation; rather, it mostly confers rents upon existing landlords. The quantity of housing shifts outwards from Q to Q', but most of the supplement is taken by landlords. It's then capitalised into land prices, helping to further push up the price of land that's made scarce by regulation.

In the happier state of the world, that pictured in the graph on the right, developers are able to bring new supply onto the market when demand for it exceeds the cost of providing it. The regulatory barriers are eased and the supply curve is consequently more elastic. In that state of the world, the accommodation supplement results in a greater supply of housing for lower-income tenants, with less of it turned into a transfer to landlords. The government is spending more in total on the accommodation supplement, but is also getting a lot more housing for its spending; it could achieve better accommodation outcomes under this regime even with moderate reductions in the supplement paid. That's because the spending mostly turns into new housing instead of into transfers to rentiers.

To keep the graphs simpler, I only rotated the supply curve. More realistically, the supply curve would have been pushed out, resulting in lower ex ante prices and higher ex ante quantities; there's less need for an accommodation supplement where regulatory inflation of land costs effectively bans developers from building low cost housing.

Stephen Franks illustrates the current political equilibrium:

For most of New Zealand's wonderful years of egalitarianism you could buy land for your house for around one year's average earnings, and build your house for about two and a half year's earnings. After taxes and living expenses you could expect to get rid of most of your mortgage over the next 10 to 15 years.

Then the baby boomers inherited political power. Already set for housing they don't need to be grateful to developers. They can despise subdividers. They'll rally to block densification, and 'sprawl' and highrises and infill units and anything that might offend their 1970s aesthetic sensitivities or glorious views. They feel the virtue in sending others to commute in trains from apartments on 'hubs' irrespective of the surveys that show fewer than 5% want to live like that.

The more scarce is housing the better off the boomers are. If you are already on the property escalator of course you will demand 'protection' of 'heritage' building and suburbs. It guarantees your overinvestment in housing. Artificial scarcity will not be exposed for the selfishness it is. Bankers of course agree. Otherwise they might find they've lent more than houses are worth, as they have in the rest of the world.

Zone more land to allow higher density use, allow more subdivision on the edges of town, and implement congestion charging so negative sprawl externalities are handled adequately. Current sets of land use restrictions build massive fragility into our systems so that private owners simply cannot respond to sudden changes in housing supply.

Tax incidence: alcohol edition

Pop quiz: if the alcohol excise tax is charged to consumers rather than to producers, are winemakers better off? The New Zealand Winemakers seem to think so:

While the industry accepted the government has to collect income tax, the excise is not reaching the market as a social policy goal, Smith said.

"We believe it should be shifted to be collected off the consumer rather than off the producer so it becomes a price disincentive as it was meant to be. The tax could still be taken, just at the other end of the cycle," he said.

Standard tax incidence theory says it doesn't matter whether a tax is collected from a producer or from a consumer; the burden of the tax depends on relative supply and demand elasticity rather than who cuts the cheque to the IRD.

There can be good administrative reasons for collecting the tax from one side of the market or the other. Exported product isn't subject to excise; who's best placed to keep track of what's exported and what's domestically consumed isn't obvious. If a winemaker operates a cellar door or otherwise sells to consumers online, he would have to submit an excise bill to the government even if it were retailers that otherwise had to do it. And while it's true that an alcohol producer will have some cash float issues around having to pay excise in advance of receiving payment for his product, it's also possible that administrative costs on the whole increase if all the excise submissions have to be handled by the hundreds of small retailers who might have a few bottles of wine along with other stuff in their shoppes. The opportunity cost of the short-term loan extended from alcohol producers to the government in terms of interest income forgone isn't zero.

I don't know whether having retailers, wholesalers, or producers submit the excise return minimises the aggregate administrative burden associated with alcohol excise; I can imagine it going either way. It's a question best left to the tax accounting specialists.

But this part does seem wrong:

Smith thinks the excise should be collected after GST so when an item is purchased, the excise tax is added on top.

"It sends a clear price signal to the consumer."

Winegrowers chief executive Philip Gregan agreed the excise should be firmly considered in a social policy context. "If you're going to achieve some social policy goals with it, then you're levying it entirely in the wrong place."

Most wineries aren't able to pass the increases onto the retailers and Gregan says their data shows some wineries haven't had a price increase in five years but have to keep absorbing the annual excise adjustment.

First, if excise is compensation to the government for some kind of expected external burden associated with consumption, then those government services included in the product price ought to have GST applied to them; excise should be subject to GST.

Second, levying excise at point of sale either makes no difference where it's included in the sticker price, or leads consumers to expect actual prices to be lower than they are by having sticker prices much lower than the price that will be levied at the till.

Third, if wineries haven't been able to pass along excise increases, they should also expect to bear the burden of a tax assessed at point of sale via similar reductions in demand.

But something that Kiwi winegrowers ought to be watching out for is the potential for changes to Australia's WET. As I understand things, small producers, including small NZ producers, get a rebate on the Wine Equivalent Tariff levied in Australia. I wouldn't be surprised if NZ wineries stopped getting that concession from the Australians.

HT: David Hargreaves. Thanks!