Counting jobs

I really really hope that somebody in the bureaus is running proper probity over whatever Shane Jones is up to in the Provincial Growth Fund. Because every time we hear news on it, it's just a bit off. 

Here's Jason Walls in today's Herald:

Numbers published on the Ministry of Business, Innovation and Employment (MBIE)'s website show more than 10,000 jobs are expected to be created as a result of Provincial Growth Fund (PGF) announcements made last year.

Of that, 7000 – almost 70 per cent – are estimated to come from just one project: The East Bay of Plenty Regional Development Project Implementation.

This project was allocated just under $240,000 for "funding a position to manage and report on 65 key economic development projects".

That 7000 jobs estimate is based on a feasibility study provided to the Provincial Development Unit, which oversees the PGF, by the Ōpōtiki District Council.

It isn't nuts to run forecasts of expected employment from projects that haven't yet gotten started, but this kind of accounting doesn't seem right.

And this seems just a bit off as well, from the head of the Provincial Development Unit:

The day after Goldsmith said just 54 PGF jobs had been created, the Provincial Development Unit issued a press release which said "job creation does not happen overnight" and that more than 10,000 jobs may be created as a result of all the PGF investment approved to date.

Speaking to the Herald, the head of the Provincial Development Unit Robert Pigou said the statement was only released to provide "clarity" around how the PGF was structured.

He said he was not able to say how many jobs had been created because of the PGF last year, as he "did not have that detailed information yet" but said it would be higher than 54.

"We're working with the [PGF] applicants because many of them are just starting to get into that employment phase [of the projects]."

The 10,000 jobs expectation is based on the estimates of the PGF project applicants, a Provincial Development Unit spokesman said.

"When applicants outline their project, they include an estimate of the number of jobs to be created both directly through the project and indirectly, as a result of the investment."

For larger scale projects, such as the $40 million to expand the TranzAlpine service, financial modelling by an independent economic firm has been used.

I would expect Shane Jones to advertise numbers based on the job creation estimates provided by folks seeking grants from him; he is a retail politician.

I would expect a neutral and impartial public service to be a bit more sceptical about claims made in those applications and not use them to tally up big numbers to put on the Ministry's website. And that makes me a bit nervous about what the heck is going on over there more generally.

It’s always been a mistake to try and sell this stuff on short-term job creation numbers. The point of the fund, as I’d understood it, was to get infrastructure and other bits in place for longer term development in the regions. There are better and worse ways of trying to do that – a pile of loans to business that no commercial outfit would take a punt on isn’t a great idea, but getting better infrastructure in place so that regions that don’t have a strong enough economic base can have a fair go – that’s not crazy.

So I think it was silly to advertise this stuff as aiming at lots of jobs quickly.

That said, it is a bit odd to count as current benefits the estimated increase in jobs that might transpire if a bunch of other projects that haven’t yet been funded all panned out as well as expected by the folks pitching the projects. I haven’t seen the underlying studies, but if the Herald’s description is accurate – it’s an odd way to do things. It’s fine to base expected numbers on projections from studies, but we might have expected the basis to be a bit stronger than that.

I also really hope that there are sufficient checks in place in the bureaus vetting these projects and approving them, making sure that none of them would cause problems for any of our international agreements around subsidies, and making sure that there’s sufficient probity around the projects.

The bout round the bach? The fight in the Ferns?

I chatted this afternoon with Jim Mora's Panel at Radio NZ about the potential government funding of a coming boxing match. There's apparently been Cabinet-level discussion of whether the government should throw money at the boxing match to make sure it happens in New Zealand rather than someplace else.

I covered a few points:

1. A boxing match is a commercial endeavour. If investors thought it would be more profitable here than it would be hosted elsewhere, they’d be putting in private money in anticipation of that return. 
2. Government funding to bring it here then only makes sense if:
   1. It would not have happened here unless it were funded (likely); but more critically,
   2. There is a real benefit to New Zealand in hosting the match here that would not be enjoyed were it hosted elsewhere; and, further,
   3. That this benefit, relative to the government’s outlay, is bigger than the government can get from spending the money elsewhere, or from leaving it in taxpayers’ pockets in the first place.
3. Governments love subsidising big sporting events. They talk a lot about the extra spending that tourists coming to events bring with them, but the kinds of studies backing these things up are usually pretty flawed. 
   1. First, they’ll count spending by visitors rather than profits on spending by visitors. Where there are costs involved in providing services to visitors, those need to be brought into the analysis.
   2. Second, they’ll assume that the visits would not have happened but for the event, and that the visits that do happen do not displace other visits. Both of these are often wrong. People who had always planned on visiting New Zealand and who like the event might shift the timing of a visit to coincide with the event. And other people who could only visit New Zealand during the time of the event might be put off: hotels get booked out for big events, for example, and rental caravans can be hard to find during things like the Lions’ tour. Just looking at the spike of visits during a big international event isn’t enough. You also have to account for displacement.
   3. Finally, none of this much enters into the picture for a boxing match which would draw fewer tourists in than a big sporting tour anyway.
4. Minister Joyce noted that he’d hope that the government might be paid back if the match turned a profit. If I were operating under that kind of contract, I’d be pretty sure to pay myself and staff bonuses big enough to make sure there weren’t profits, or to buy my supplies from a related company at inflated prices to make sure there weren’t profits to pay back. But maybe the companies with whom the government strikes these kinds of deals are more publicly spirited.

Government funding of boxing matches... I'm just reminded too much of Mobutu Sese Seko fronting the costs to host the Foreman-Ali fight: the Rumble in the Jungle.

RNZ's panellists weren't having a bar of it either. Neill Miller has great economic intuitions - he also pointed to most of the stuff that I covered. And I doubt I'd disagree with Gordon Harcourt's take that the main point isn't the economic benefits of having tv cameras reporting from an indoor arena in Auckland instead of wherever else, but rather the potential drawing of voting support from South Auckland fans of Joseph Parker.

I followed on immediately after Marewa Glower's excellent discussion of vaping as a way of avoiding the weight gains that otherwise may come with quitting smoking.

Update: Catch also Sam Richardson on point here. I also heard him on Radio NZ on the way in to work this morning.

And Northland might just be in Chretien's Quebec

Quebec...Northland...

“For 25 years, the people of Lac-St.-Jean and Chicoutimi were promised roads by the former Jonquière MP [Lucien Bouchard, former head of the Bloc Quebecois]. Now that they have a Liberal MP, they have hope.”

That was Jean Chretien, 2002.

Today:

Osborne, who has risen from electorate treasurer to potential MP in just weeks, even warned a flagship National roading project to extend State Highway 1 as a motorway further into Northland, could be derailed if Peters beat him.
"If the people of Northland want the Puhoi to Wellsford motorway extension, which is vital to open up Northland to our biggest market . . . if we want that, there's only one choice, and that's to vote for me," Osborne said.
If he lost the by-election there was a "real risk" the road might not go ahead.

Surely the PM will clear this up and confirm that his candidate really didn't mean what it sounds like he meant.

Road hog (or at least pork)

Were I teaching my undergrad Public Choice class again in two weeks instead of heading off to Wellington to join up with the NZ Initiative, this would have been Assignment 3. It couldn't have been the first assignment, because that comes after they've gotten basic constitutional political economy and social choice, but before they've gotten the basics of their own electoral system and the mechanics of pork barrel politics. It couldn't have been in Assignment 2, as the kids then wouldn't have yet gotten the economics of bureaucracy. It could have been Assignment 3, or on the take-home exam.

In theory as developed and discussed in lectures and in your readings, we found that First-Past-the-Post systems tend towards geographic-based pork-barrel policies while PR and MMP systems tend instead to demographic-based pork.  

During this year's election campaign, National announced a substantial set of roading projects to be funded outside of the normal NZ Transport Agency funding process. Normally, roading projects are chosen by a process largely outside of politics: collected petrol excise and road user charges in the Land Transport Fund are used for road maintenance and for new roading projects, where the projects are ranked in importance by bureaucrats rather than politicians. These ones were chosen by the National Party and funded from revenues outside of the Land Transport Fund.

a) Explain the basics of pork-barrel policies and why they might differ between electoral systems.

b) Why might a government agree to leave roading decisions to an arms-length body?

c) Was the roading announcement above surprising? Why or why not? Explain National's decision.

One secret in my assignments and exams: I often didn't really know what the answer was. All that grading rubrik stuff we were supposed to have done this year, well, that wouldn't have gone well for part c of this question as I don't have a great answer here. Students who could apply the theory well to the case at hand would do well, and those who came up with explanation that didn't demonstrate any understanding of the theory failed.

In Part a, I'd expect some discussion of the basics from Persson & Tabellini and from Stratmann & Baur.

In Part b, I'd expect some discussion of the benefits of delegation. In a FPP system, such delegation is more needed than in MPP, because every MP will be tempted to skew things towards their own districts. We might expect a delegation to a partisan committee that weighs up overall benefits to the governing party, but not to a neutral committee. Under MMP, you could get bipartisan sustainable agreement to leave it to a nonpartisan committee because the benefits of distributing geographic pork are more limited, as far as the Party's concerned, but individual district MPs will still try to push for it anyway.

In Part c, well, I haven't a great answer. Maybe National's doing it as a sop to the hinterland in general, reckoning that Auckland won't care enough to punish them for it. I wonder what my students would have done with it.

What's really irritating here is the precedent. Once we've broken the general user-pays nature of the National Land Transport Fund, all bets are off. When the system's working, the fees motorists pay are what pay for highway construction and maintenance, with some local Council contribution (half) toward local roads. If you want rail lines and other stuff, that comes out of general revenues: it can't come out of petrol excise, 'cause that's for roading projects. When you break that deal and run partisan allocation of funding for transport projects rather than leaving it to the Transport folks who are meant to be running things through a cost-effectiveness filter, well, who knows what messes might yet come.

Do read Transport Blog's excellent discussion:

Perhaps the most worrying thing about this announcement isn’t so much the projects themselves but that the government is getting more and more involved in picking projects rather than leaving it up to the NZTA to decide on spending based on merit. It started with the RoNS and last year we got the Auckland package.

When highways are ribbon-cutting triumphs for MPs rather than mundane bureaucratic decisions, I expect we get worse outcomes.

I also thank @zippygonzales for related discussion.

Update: Just remembered that I was in a Twitter argument last week about how we needed to keep highway funding separate from public transport funding because it's harder to go on dumb C>B roading sprees when you have to fund it from petrol excise and road user charges. Hmm.

Canada's Economic Action Plan and the Pork Barrel

The National Post reports that Conservative districts received 57% of allocated projects but hold only 46% of the seats.

That by itself isn't evidence of distortion.

If, for example, infrastructure money is best spent on fixing the atrocious Trans-Canada highway, and if that highway passes mostly through rural areas, and the Conservatives hold most rural districts, then we'd expect Conservative districts to receive more money. What we really need is a measure showing the predicted level of spending in each district given its underlying characteristics (other than partisanship), then look to see how partisanship distorts those funding decisions.

What would be some first cut measures? Unemployment for starters. We might expect that higher unemployment districts would be first in line for spending. Second, municipal government fiscal capacity. If these are mostly projects where the province and the municipality has to pick up part shares of the cost, then if the province or the municipality is less likely to be able to pick up their part shares, and if those districts tend to be NDP or Liberal, then again we'd see the current correlation between party and spending. You could easily then say that there's partisan bias in the project design: that the criteria were established to benefit Conservative districts; but, that's somewhat different from subverting the funding rules to benefit Conservative districts.

Finally, I'm curious about the patterns of spending within Tory districts. Did it mostly go to safe seats or marginal seats? Cabinet or backbenches? If it's like Liberal spending under HRDC's job creation grants program, I'd expect distortion to favour Cabinet members in marginal districts especially.

The numbers don't look good, but we need a bit more work.

Ban the cheques

Colby Cosh says ban the cheques. Putting party logos on federal government cheques for community projects ... very distasteful.

The fact is that our system of government allows MPs on the side that has the confidence of the House to engage in feudalism. This is the only appropriate term for a structure whereby communities and organizations are encouraged to display fealty to a person, and subsequently receive reciprocal benefits directly from his hand. It has been a feature of every Canadian government, whether Liberal or Conservative, and I suppose there might even be some things to say in its favour. It is certainly what the vassals expect; it is, at root, the feudal instinct that makes voters ask what their MP has done for them lately, or why the government isn't "doing something" to help a particular region or industrial sector.

I'd previously commented on the story here.

Plus ca change

From today's National Post:

Pork-barreling in Canadian politics is hardly new. The task of the politician is to climb the tree to shake down the acorns to the pigs below, conceded Sir John A. Macdonald in a moment of, presumably drunken, sincerity.

Now we have Gordon Landon, regional councillor for the Town of Markham and the nominated Conservative candidate in the Ontario riding of Markham-Unionville.

In what will likely prove to be a career-limiting move, he admitted on live television that the reason his riding has not received federal funding for a medical testing centre is that the Member of Parliament is a Liberal.

From Jean Chretien's mouth, 2002 (link is to Free Dominion, but the quote was also in a Paul Wells story from the National Post, February 2002...will they ever put up their back archive?)

“For 25 years, the people of Lac-St.-Jean and Chicoutimi were promised roads by the former Jonquière MP [Lucien Bouchard, former head of the Bloc Quebecois]. Now that they have a Liberal MP, they have hope.”

The National Post story suggests heavy weighting of federal Canadian stimulus spending towards Conservative-held districts. I found no such distortion in overall spending under the much-maligned HRDC Transitional Jobs Fund and Canada Jobs Fund under the Liberals: if anything, the average Liberal district received much lower funding correcting for the unemployment variables on which decisions were meant to have been made. However, Liberal cabinet ministers received much much more.

It would be nice to have the dataset that the Libs are talking about. In the HRDC case, it was much easier: the scandal resulted in the release of the data, and grant allocations were supposed to be based on unemployment criteria; consequently, I could check actual allocations against the criteria for being given grants. Not sure that could be done in this case. I'd also worry about using a sample of all projects where the sample was chosen by the Liberal opposition rather than by a random draw.