Baseline 2019

I'll be in tomorrow's budget lock-up and on TVNZ's ridiculously early Breakfast panel pre-budget roundup ahead of it. 

For rather some time I've been making the case that government needs to retrench Core Crown Expenditure to pre-Covid levels. 

The Great Wellbeing Budget of 2019 was not austere. It was a substantial increase in spending, as a fraction of GDP, as compared to the prior National government's last budget. 

So. 

Take every Vote from Budget 2019 as a percent of GDP.

Make that the baseline for a future budget. Maybe 2025 if they're somewhat ambitious, 2026 if less ambitious. 

Some programmes or structural changes since Budget 2019 might be considered worth keeping. Government would have to trim other spending so there'd be room for it. 

I'm less interested in overs and unders around automatic stabilisers. Benefit spending will be higher in 2025 than 2019 because unemployment is likely to be higher; note too though that jobseeker numbers remained very high relative to unemployment over the period after lockdowns. 

If Budget 2025 were "Hey, this is the same as Budget 2019 would have been, as a fraction of GDP, if unemployment had been at 2025 levels", all good. There would be no structural deficit. 

Let's compare a few numbers here. 

At Budget 2019, Treasury put up forecasts of Core Crown expenditure for the year to June 2023. The bulk of Covid spending should have been 2020 and 2021. The traffic light system ended in September 2022. 

Budget 2019 forecast a 4.3% 2023 unemployment rate. The 2023 half-year fiscal update had a 3.6% unemployment rate for year to June 2023. So higher unemployment isn't responsible for any differences in spend between 2019's forecast 2023 and the actual 2023. 

Let's compare Budget 2019's forecast for core crown expenses by functional classification for 2023 with actual expenditure by those classifications for the year ended June 2023, inflated by the extent to which nominal GDP for 2023 exceeded 2019's forecast GDP for 2023. Nominal GDP in 2023, divided by the forecast of 2023 GDP from Budget 2019, then multiplied by each functional classification line. 

Overall, Core Crown spend is $13.4b higher than had been expected in 2019, after inflating for nominal GDP increase over the period, and accounting for Budget 2019's expectation of about $10 billion in new operating spending by 2023 that couldn't be classified in 2019.

Or put it this way. Budget 2019 expected Core Crown spending in 2023 would be about $104 billion across those functional areas, plus $10 billion in new operating spending, for $114b all up, after adjusting for nominal GDP growth to catch population increase and inflation.

Actual 2023 was almost $127 billion. So $13.4 billion more than would reasonably have been expected.

Despite unemployment being a lot lower in 2023 than had been forecast in 2019 for 2023, social security and welfare spending were $3.8 billion higher than had been forecast for 2023 at Budget 2019.

Health spending was $7.7 billion higher. 

Finance costs are substantially up, and there won't be much government can do about that now other than avoiding making things worse.

Transport is $1.8b higher. Core government services $1.6b higher. Education $1.5b higher. 

Spending has blown out across the board. If you take the excess spend as a proportion of the had-been-forecast spend, the blowout is of course biggest in finance costs, with with housing and community development (60.3%), heritage, culture and recreation (57.8%), environmental protection (52.2%), primary services (48.9%), and transport and communications (47.1%) following - health only blew out by 36.8% as compared to forecast and social security and welfare by only 9.9%. But those two line items are so very very big. 

Getting back to Core Crown expenditure comparable to 2019's when measured against current GDP isn't a small job.

The Taxpayers Union commissioned a poll from Curia. Sample of 1000 respondents. 

They wanted to know whether people viewed Robertson's Wellbeing Budget, with Core Crown spending of 29% of GDP, as being too low, about right, or too high. 

Only Green Party supporters reported that it was too low, or at least in any substantial numbers. Among Green supporters expressing an opinion, about half of Green supporters thought it was too low, and half thought it was about right. 

Among the population at large, 9% thought it was too low, 34% about right, 29% too high, and 29% were unsure. 

Among those expressing an opinion, those thinking spending in 2019 was too high outnumbered those saying it was too low by more than 3 to 1, with most saying it was about right.

It isn't crazy to view the pre-Covid spending proportions as a decent baseline. And most people didn't think Budget 2019 was austerity - far more people said that it spent too much, as a fraction of GDP. 

Budget 2019 reckoned on Core Crown spend of 28.8% of GDP over the medium term.

December's Half-Year fiscal update had 2023 Core Crown spend at 32.2% - the mess I pointed to above. And forecast it would be worse for 2024 at 33.4% of GDP. 

People rightly recognise Budget 2019 as not being austere. 

Getting back to spending consistent with government's share of the economy in 2019 is still going to be a big job. 

I'm hoping tomorrow to see reasonable work in that direction at least. Especially if they also want to deliver tax reductions, given the size of the deficit. 

National will be damned as austerity merchants for any spending reductions, regardless of how many people viewed 2019's budget as too generous. 

If you're going to be hanged anyway, better to be hanged for a sheep than for a lamb. I don't expect any sheep here. But they could at least aim for hogget. 

Robertson's Seventh Budget?

Dan Bunskill reports on Finance Minister Nicola Willis's speech this week:

Willis said it wouldn’t be a “big-spending” Budget, knowing that Crown finances could get worse before they get better, but she wouldn’t “overreact” to worsening forecasts either.

It will not be “an austerity Budget, of the sort suggested by a few commentators seemingly enthusiastic to see the mistakes of history repeated”.

“Our Government knows how devastating it would be if we were to give up on overdue tax relief, to drastically cut-back on investment and public services, and to downsize our ambitions for growing New Zealand’s economy”.

"Drastically cut-back". 

Right. 

National hopes to get core spending down to 30% of GDP, sometime. 

Ardern's 2019 Wellbeing Budget was forecast to cost 28.8% of GDP over the medium term. Willis wants to avoid "drastically cutting-back" to a level higher than Ardern had set. 

Pretty easy to read those lines in Robertson's voice really, maybe other than the bit about overdue tax relief. 

As I'd put it in The Post earlier this week:

Tax revenue and government spending are both substantially higher than they were before 2020’s Covid lockdown – whether measured as dollars collected and spent, or as a fraction of overall economic activity.

Core tax revenue rose from just under 28% of GDP in 2019 to just over 29% of GDP forecast for 2024. But core government spending increased from 28% of GDP to 33.4% of GDP over the same period. The difference between the two is a problem.

Government spending increased by over six percentage points of GDP in 2020. Spending to deal with the worst parts of lockdowns and border closures mattered. But wage subsidies are now gone. Borders are open. And core government spending, as a proportion of GDP, is forecast to be only about a percentage point of GDP below its peak.

If you think the continued increase is because government has had to put more money toward healthcare, in the wake of Covid, and toward education, for dealing with the lingering effects of Covid through the school system, check the figures.

In 2019, education and health together were about $30 billion of a $100 billion government budget – 30% of the total. If education and health were pulling spending upward, they would now be a larger fraction of the larger budget. But Budget 2023 had education and health as about $44 billion of a $176 billion government budget – about 25%. Health and education are not what has driven the overall increase.

Compare growth in health and education spending with growth in other areas.

Spending at the Ministry for the Environment increased from $708 million to just under $3.5 billion over the same period. Transport’s budget also more than doubled – from just under $5 billion to just over $10 billion.

The Ministry for Social Development went from $26 billion in 2019 to $41 billion in 2020 – understandable when the wage subsidy was in place and lockdowns blocked jobseeking. But MSD’s 2023 budget was $43 billion, despite relatively low unemployment rates.

I'd there concluded:

Basic maths lays out the options. Spend less, take more in taxes, or a bit of both. Anything else reeks of denial. But surely a May budget set by a National-led coalition Government ought to balk at setting a core spending path entrenching government spending as a larger share of the economy than Finance Minister Robertson promised in 2019.

What I'm really hoping for in Budget 2024 signaled strong consolidation for Budget 2025. Set each budget line for 2025 as the same fraction of GDP as it was in Budget 2019 - as a baseline starting point. Ditch programmes that don't make sense to maintain stronger funding for ones they want to keep. 

I'll be at the lockup. The last lock-up was really rather fun. At least for me. Maybe less so for a few others. 

In a structural deficit, the only real tax cut is a spending cut

This week's column in the Stuff papers. A snippet:

Tabarrok warned that America had two political parties – “the Tax and Spenders and the No-Tax and Spenders” – and neither was fiscally conservative. In the two decades after Tabarrok’s warning, the federal government never achieved a balanced budget. America’s federal deficit ranged from 1.1% of GDP to over 14% of GDP and gross federal debt doubled, rising from 60% of GDP to 120% of GDP.

New Zealand’s Public Finance Act aims to avoid those kinds of outcomes.

The fiscal architecture is neutral about whether core government spending should be about 27% of GDP, as it was in 2018, or over 33% of GDP, which it is forecast to be in 2024.

The six-percentage point difference in core Crown expenditure, as a proportion of GDP, might not sound like much. But in a $405 billion economy, it amounts to almost $25b in increased core government spending per year across a population of just over 5 million people. Or over $18,000 for a family of four.

Some of that substantial increase in spending has been funded by a higher tax take. Core Crown tax revenue increased from just over 27% of GDP in 2018 to over 29% in 2024. The rest, after accounting for other bits of government spending and revenue, is funded by what is now a substantial structural deficit.

Taking on debt during the depths of lockdown to deal with the crisis was one thing. Now, that crisis has long since passed – but the structural deficit remains. The One Big Task for the coming budget is providing a credible path out.

It is not just a test for the current government. It is also a test of the Public Finance Act.

I'm all for tax cuts. Start by balancing the structural budget. That's the most important tax cut to make. Then see whether there are opportunities to go further. There should be. The government takes in a lot more in tax than it did in 2018, and is spending a lot more than was projected in the 2019 Wellbeing Budget. 

I'd filed the column a few days before having a chat with RNZ's Morning Report on proposed staff reductions in government agencies. I was a bit frustrated about that RNZ rarely informs listeners about whether cutbacks in an agency simply undo the last couple years of hiring, whether they revert staffing levels to levels comparable to pre-Covid, whether they go back to where things were when National left office, or whether they go further than that. If an agency has stayed fairly constant over time, and there haven't been substantial productivity improvements or other shifts, a 40% cut would mean they can't keep doing stuff they've been doing for a long time. But if one has tripled in size recently, and then cuts back by 40%, well, there are options around what things no longer get done.

Checking the last five years' data is really easy at the Public Service Commission's website. Click the link, scroll down to the table, click the tab to open the 'Five Year Trend' view. Done. 

I don't know why Radio New Zealand doesn't check. They were really outraged about proposed 40% staffing reductions at the Ministry for Pacific Peoples. That Ministry had 40 staff in Q2 2018 and 136 by Q2 2023. They reduced a bit at the end of 2023 to 121 staff. If a 40% reduction comes off that figure, they'd still be left with more staff than they had in Q2 2020.

Update: Kate MacNamara does some checking. 

2023, as forecast in 2019

At Budget 2019, Treasury put up projections through 2023. 

According to those projections, which incorporated Treasury's best guesses about everything that was going into the Budget, Core Crown tax revenue would be 28.8% of GDP in 2023. Core Crown expenses would also be 28.8% of GDP. 

Covid happened in 2020 and all bets were off for that period. But Covid-spend has now largely worked its way through. Spending should be getting back to what was forecast for 2023 in 2019, right? 

Nope. 

Core crown tax revenue at the Budget Economic and Fiscal Update 2023 was 29.3% of GDP and Core Crown Expenses were estimated at 32.5% of GDP. So spending in 2023 is more than three percentage points of GDP higher than the government had forecast in 2019. 

I covered a bit of this at the Financial Services Council conference last week in Auckland; Jarod Kerr and I were on a Chief Economists' panel with Jack Tame. 

RNZ this morning gave a fairly soft ride for the Public Service Association's rep on ACT's proposals around cutting government spending, with the PSA noting that population increase has mattered and that demands for services are higher.

It would perhaps have been helpful if RNZ had provided that minor bit of context. As a percentage of GDP, core crown expenses are more than three percentage points higher than Labour had planned on in 2019. Percentage of GDP will catch migration and inflation. 

The fights between ACT and National, and Labour/Green, on appropriate size of government are one thing; it's in part a value judgement. 

But government is currently spending more than three percentage points of GDP more than the Labour coalition had forecast in 2019. 

It's a very substantial increase in the government's share of the economy.

Even getting back to what Labour had promised the path would be gets portrayed as terrible cuts. Was the 2019 path really that draconian? 

Budget 2022

I have a hard time seeing how Budget 2022 is consistent with the new fiscal rules.

In shifting from a net debt target to a net debt ceiling, they added a requirement that budgets normally provide surpluses. It's the old Keynesean fiscal macro kind of idea: run small surpluses in normal times to build up padding for large deficits in bad times.

And it has the same political economy problem that this kind of thing has always had. Government prefers to ratchet up spending. The old net debt target helped to constrain against that. The new debt ceiling won't, until we hit it. 

So Treasury warned that the new setup required greater focus on fiscal discipline to make sure that dollars provide value for money. A "things must at least pass CBA" rule could take the place of a net debt target in preventing budgets from blowing out.

And yet.

We get hundreds of millions for cash for clunkers and piles of other climate initiatives and subsidies to corporates targeting emissions already covered by the ETS cap. 

We get an extension of the road user subsidy which makes no kind of sense. 

Instead of inflation-indexing the tax thresholds, the government gives a handout to lower-income households equivalent to the extra tax that was stolen from them through inflation pushing about $5000 in earnings from the 10.5% band into the 17.5% band - and packaged as government largesse rather than compensation for a small bit of the accumulated inflation theft. 

Government is running a substantial deficit while the economy is overheated. There's a strong positive output gap. Even on old Keynesean approaches, you're not supposed to do that. 

And think about some of the other ratchets that they've now set up.

They've adjusted one benefit setting that has been a bugbear for child poverty advocates for a long time. If you're on a sole parent benefit, child support payments from the non-custodial parent mean your benefit is clawed back. 

Why would they do that? Sounds mean right? 

The idea underlying it is that the government wants to target support, as much as possible, to those who don't have other means of support. So say that one uninsured parent dies in a car wreck and the other parent, who wasn't in the labour force before, winds up in a hard spot. The government wants to provide more support for that child than for a child where there's additional support coming in from a non-custodial parent. 

So they claw back some government support in the case where the non-custodial parent is able to provide support. That means government can afford to provide a higher baseline level of assistance, before clawbacks.

Now there can be arguments around administrative messes - if the government is not good at enforcing child support payments, or about ensuring that those payments get reported, then the system can encourage informal payments, or discourage all child support payments.

But in-principle, the setup isn't crazy. It at least tries to make sure that more money goes to kids in worse situations when there's a fixed pool of money available for support. And even if your political position is that that pool should be doubled or tripled, you would still do better with that kind of targeting if you wanted to make sure that the money went to where need was greatest.

They got rid of that.

So what's the predictable effect in a couple of years? A John Campbell special that will highlight the grave disparities facing sole parents who have no support as compared to those sole parents who receive child support payments? Is it fair that the kid whose parent died, or whose non-custodial parent has no income, has such worse circumstances than the kid whose non-custodial parent provides a lot of support? And then we run the ratchet again. Someone will recommend supplementary payments where there isn't that support, which will require monitoring and abatement when there is support, and then complaints about how that system runs and demands that the supplementary payments go to all parents, and we get to play the cycle again. 

The Dom wanted a short piece from me yesterday afternoon. I didn't attend the embargo this year, but pulled this together shortly after the embargo lifted. I hit on some of this. 

I wonder if there's anyone left in Treasury who would regret the new fiscal rules or whether all of that cohort left a decade ago. They should regret the new fiscal rules. The only real problem the rules work to solve is ensuring that debt targets don't unduly hinder infrastructure spending that does pass cost-benefit - and that problem would better have been solved through revenue bonds financed from fees or charges on the beneficiaries of the infrastructure, and separated from Council and Crown main balance sheets. 

Afternoon roundup

The browser tabs...

• The Internet Archive will help improve access to a lot of books currently held by the National Library. Naturally, this upsets some people. 
• Mike Joy wins a battle over what "natural levels" of nitrate in aquifers might mean.
• Places where the local schools are funded by higher levels of government wind up trying to privatize the public schools through zoning. Everything's tradeoffs though, eh? If schools are funded by the local communities, poor places get really bad schools. It's instead a good argument against even greater anti-NIMBY vigilance.
• Potential nudges to improve compliance with scanning-in for Covid tracing. I'd add one more: put the QR codes in lots of places. Make it easy. At the entrance is great, but if your phone isn't ready and you'd be holding up a queue, maybe you won't scan and you'll forget to on the way out. Put them everywhere. 
• The EMA makes a very bad argument about unemployment rates. The point of inflation targeting and increasing rates when you're hitting capacity constraints isn't to create some stupid Marxist reserve army of unemployed, which seems to be what they're arguing for. The point instead is that if a monetary push gets you past NAIRU, you just wind up back at higher unemployment rates anyway once contracts adjust AND you have higher inflation to go with it. Anyway, because either EMA put it badly or it was reported badly, a billion tweets followed claiming the Marxists were right all along. 
• Heather du Plessis-Allan cites me here on Treasury's diminished capabilities. Worth noting: all of that was prior to the current Secretary's appointment. She may well have improved things since; I need to get OIAs in to check. 
• Shipping is a mess.
• Remember back at the Budget Economic and Fiscal Update when Treasury put in a footnote saying they couldn't produce a structural deficit figure this time round because they couldn't tell how to place some of the Covid-spend? And it seemed odd because Covid-spend is supposed to be emergency = one-off = not-structural? Well, looks like the free school lunch programme is Covid spend. I simply do not understand how this works. Shouldn't the Auditor General be going after this kind of thing as misappropriation? 
• The government isn't bothering to check on whether people got their Covid tests before travelling. Some of them get caught ex-post, with the option to go into quarantine or return to Oz, after they've been circulating in New Zealand already. If they went back to Oz and turned up positive, what hope would we have on contact tracing them? However low your expectations of government competence are, you just keep being surprised. 

Do less, well.

Budget day's today; I'll be in the lock-up after 10. 

The best pre-budget take I've seen thus far is from Harman at Politik, or at least I hope he's right.

The Government has run out of bureaucratic capacity to undertake any more substantial reforms. 

Finance Minister Grant Robertson conceded this yesterday and suggested it would influence the Budget on Thursday. 

In other words, the Budget is unlikely to contain any new broad initiatives.

The public sector has had a lot of crazy stuff thrown at it over the past few years, and not just because of Covid. Things like massive tax shifts on investment properties that weren't well-canvassed with the bureaus before being launched. A complete restructuring of the health system in the middle of a pandemic. An immigration system that's falling apart. Series of absolutely absurd spending initiatives in minor climate change things that will do not a lick of good because they're covered by the ETS's binding cap, but will require bureaucratic time and effort to sort out. 

And more:

But what is likely to be missing on Thursday will be a big new initiative in any policy area. 

And that is because the “big” departments are already flat out on a number of major reforms — the Resource Management Act replacements; the Freshwater reforms; the response to the Climate Change  Commission; the Health reforms; the Immigration review; three waters reforms; the local government review and the future of work as well as big changes in education.  

“Along with the focus on our three core areas of housing affordability, climate change and child wellbeing, we have a complete overhaul of the health system, a complete overhaul of our planning laws and a complete overhaul of the way we manage water,” said Robertson.

The government is not only doing too much, it is doing too much of that too much too badly. 

Gemmell on the budget

Norm Gemmell's piece over at Newsroom is harsh but fair. 

Before the Budget, I called for a stimulating, but prudent, Budget. That meant:

- moving away from universal, to more targeted, wage subsidies

- setting up flexible spending programmes now that respond to needs as they arise

- reprioritising spending away from short- and longer-term ‘nice to haves’ to essential recovery support

- presenting a credible future debt track beyond the immediate recovery.

Pretty reasonable. 

What did the Budget deliver? Arguably none of the four.

Firstly, another two months of universal wage subsidies with slightly stricter conditions (a bigger fall in business revenue) doesn’t make them ‘targeted’. This is despite projections that the economy is already getting back to 80 percent or more of normal working, with a few sectors likely to continue suffering. And when we’re two months closer to that election, the $20+ billions of ‘unallocated’ Recovery Fund can be strategically dropped into the election battle as further subsidy extensions and other vote-targeting sweeteners.

There's good reason to have unallocated funding when you can't guarantee that there won't be renewed outbreaks. But where a lot of the current budget and post-budget spending announcements really sound like election campaign spending - like the increases in Early Learning Centre funding for places with 100% qualified teachers - we may fear that contingency funds will be put to similar purpose rather than held against worsened Covid conditions.  

Thirdly, what of ‘reprioritising’ spending? There is negligible evidence of the Minster’s pre-Budget promise that some spending would be ‘put on ice’. On the contrary, almost every spending ‘vote’ (or should that be ‘special interest group’?) gets a boost over 2020-24, including making sure future inflationary costs are covered. Sweeteners everywhere, from arts and culture to conservation to $75 million for the Ministry of Social Development’s office refit programme. No fiscal constraints here, no public sector pay restraint or postponing ‘nice to have’ project for a year of two.

Do we really need those new Inter-Islander ferries right now when we have massive, more urgent spending needs? And an extra $1.6 billion for ‘retraining’ and apprenticeships. How many Air New Zealand flight attendants or Queenstown tour guides does he think are up for retraining to move on to farms?

Matt Burgess made similar points in the Dom last Friday.

Why are we paying $3 billion (up from $0.4 billion ‘prescribed by formula’) to the New Zealand Super Fund in 2021-22 to pre-fund future pension spending, while we also borrow massively to fund today’s crisis? Answer: Labour in opposition criticised the National government for precisely this after the global financial crisis. It dare not be seen now to be doing the same, even though suspending payment makes more financial sense.

I wonder instead why the government isn't going even further: spending down the Super Fund while cutting entitlements from a decade or more from now to balance the thing out. 

From a healthy Crown net worth of 43 percent of GDP at the time of last year’s Budget, this FSR forecasts net worth at 34 percent in 2020. Fair enough – we are coping with extra crisis spending and lower revenues. But, following the Budget, Crown net worth is projected to drop to just 9 percent of GDP by 2024 and only get back to 12 percent by 2030. In other words, a decade from now, during which time another serious crisis could easily have hit us, the Government’s plans for net worth are so diminished that we would be woefully unprepared financially for another fiscal bail-out of the economy.

Let’s be clear. New Zealand economists are not calling for future ‘austerity’ to get the Government’s books back into balance within a few years. Instead, there is widespread support for suitable, not profligate, spending to assist faster growth of the economy to reduce the debt burden to previous levels over, say, a decade or more.

But the Crown’s projected financial vulnerability a decade on reflects Robertson’s failure to offer any credible plan to raise net worth through fiscal prudence down the line. This is despite official forecasts of quick economic improvement: real GDP growth is forecast to be massively positive at 8.6 percent in 2022, and 4.6 percent in 2023.

And Robertson wouldn't say whether he views 42% debt-to-GDP ratios in 2034 as the new normal for fiscal prudence. 

Crisis budgets and returns to prudence

The Spinoff asked me for comment ahead of last week's budget.

Here's some of what I'd told them, hit the link for the rest:

The government should have two priorities in this budget. First, and most importantly, the health system needs to be ready for the medium-to-long term changes that the pandemic has forced on the country. Second, the government needs to ensure the temporary measures necessary are in place for all of us to deal with the current mess.

Those two priorities lead to one unavoidable consequence. Permanent spending increases – increased spending on matters unrelated to the pandemic that are locked in and hard to reverse – means less room for dealing with the pandemic.

Why is that the case? The government will need to take on debt to deal with the current crisis. Tax revenues will be down and spending will have to be higher. But while the Public Finance Act allows this kind of response to emergencies, it also requires the government to think about the path back to prudent debt levels. This requirement makes a lot of sense: we just cannot know when the next crisis will hit; the Alpine Fault is overdue. We will need room to deal with the next crisis too.

Returning to prudent debt levels after this is all over will be much harder if the government locks in increases to baseline expenditures; that would limit how much debt the government could afford to take on now. An adequate response to the current crisis requires fiscal discipline about matters unrelated to the pandemic response.

After lockup ended, I wondered why the government didn't seem interested in the tiny investments it would take to be able to start reopening the border to safe travels: travel from safe places, or under strict quarantine and testing requirements.

I had a bit of a chat with Duncan Grieve about it at Spinoff. 

Then I filed this with Stuff:

The Budget headlines will focus on the extent of the Government’s additional support for households and businesses through extensions of the wage subsidy programme, worker retraining initiatives and job creation schemes in areas like environmental improvement and house-building.

The Government avoided locking in too many increases to entitlement spending. But it will still take until 2034 for its net debt levels to fall to levels twice what they were prior to this crisis.

What seemed more absent was a vision for economic recovery grounded in some of the advantages New Zealand now enjoys. The Budget presents a vision of a Government-led recovery with a comforting message: the Government is there to provide support, jobs or retraining — a sealed lifeboat bobbing on rough seas until the storm clears.

But small investments in setting rigorous quarantine facilities so others can safely join our lifeboat would pay off rapidly — and many times over — while enabling different visions.

One bit that did worry me during lock-up: the budget documents project net debt rising to well over 50% of GDP, but only dropping to 42% of GDP by 2034. I asked Minister Robertson whether he viewed 42% Debt-to-GDP ratios as being the new long-term fiscal prudence, or how much longer it would take to get back to normal levels. Rather than answer the question (which I thought was a fairly simple one), he said he viewed himself as acting prudently and in line with his obligations under the Public Finance Act. But I hadn't said that taking on debt was imprudent. I had wondered about whether the end point in the forecasts was just because they couldn't see getting back to prudent debt levels within the forecast window, or if they'd revised their view on what counted as prudent. 

It's important to know. Matt Burgess's column in the Dom was on-point.  

However, Government borrowing is considerably greater at $140b over the next five years. New operating spending, excluding welfare and health, peaks in 2022, long after the economy is expected to have turned the corner.

It is also hard to find evidence of the reprioritisation referred to by Grant Robertson. Nearly every part of the public service sees new spending.

Around the budget traps

Budget lock-up on Thursday was fun. I didn't get to ask Minister Robertson whether he still had confidence in his Treasury Secretary, but it probably would have been rude to ask. 

A few bits from me on it over at the Spinoff, our Insights Newsletter, and over in the Stuff newspapers - I think it ran in both the Dom and the Press on Friday. They reversed the headline, but s'all good.

The end of my Dom piece:

Treasury often undertakes this kind of programme evaluation work, or at least assists in overseeing it. But Treasury's core economic and analytical capabilities have been seriously eroded under its secretary, Gabriel Makhlouf.

Treasury was awarded $5m a year over the next four years to "deliver core functions and the wellbeing approach". Rebuilding Treasury to be able to deliver the evaluation capabilities necessary for achieving Robertson's vision may be a bigger job than that.

It would have perhaps been the wrong day to ask the minister of finance whether he has any confidence in Treasury's capability to deliver. I have doubts.

Programme evaluation has always been the poor cousin to programme announcements. When voters too easily infer a government's depth of caring by the quantity of its spending, rather than the outcomes that spending achieves, that outcome is not surprising.

I hope the wellbeing approach yet proves to be something more than that. It may otherwise be difficult to distinguish from business as usual.

Oh Treasury ... again

Tom Pullar-Strecker goes through the more likely explanations of the inadvertent partial budget release. 

There is speculation Budget documents may have been "hacked" by someone simply guessing the website addresses of documents prepared by Treasury but which were not yet supposed to be visible on its website.

Treasury has been approached for comment on whether that is what may have occurred, or whether the alleged hack might have been more sophisticated.

Puller-Strecker's interviewed experts think Treasury screw-up the most likely explanation, with pages being indexed that shouldn't have been indexed.

Whether this was due to somebody putting up placeholder content using near-live budget documents in a test environment that was less test than they'd thought, or a screw-up in the CMS where embargoed content was cached by Silverstripe in ways accessible to crawlers* - the current plausible explanations are Treasury screw-up.

The screw-up would fall under the Director of Operations' remit. But the Director of Operations is only a few weeks into the job, having replaced Fiona Ross, who left in April to lead the Ministry of Justice's Family Violence And Sexual Violence unit. You may remember Fiona from Danyl McLaughlin's reporting on DEVUCA worlds. 

It's still possible this was a leak from elsewhere. But if this winds up being confirmed as Treasury screw-up, it would be difficult to blame a DDO who's only been in the job a couple of weeks. 

And I agree with Hamish Rutherford's piece here: if Treasury really thought that market-sensitive information had been hacked, the budget would already have been released early to avoid giving any hacker a time advantage. Plus, were someone keen on really hacking Treasury, I'm not sure that the budget would be the most interesting stuff to go after. Tendering, commercial contracts, debt issuance - there's a lot of stuff that they'd have in the back-end that would be rather more interesting than a soon-to-be-released budget unlikely to have much market consequence. 

* I understand this to be a known issue in Silverstripe in some government implementations, but I could not possibly explain it.

[Update - link fixed along with spelling of Pullar-Strecker -- Doh!]

A few notes on the budget

Your take on last week's budget should depend on whether you have your economist glasses on, your politics glasses on, or your "economic policy is constrained by institutions and elections" glasses on.

The economics of it are pretty average.

The increase in the accommodation supplement is likely to flow through primarily to landlords. Radio NZ has reported on work at MSD showing relatively little effect of a prior increase in the accommodation supplement on rents, but the basic tax/subsidy incidence on this stuff depends on market conditions. If supply is more elastic than demand, meaning that a given increase in rent does more to stimulate new construction of rental properties than it does to decrease demand for rental accommodation, then the accommodation supplement is good for low-income tenants receiving it. If you run your study when regulatory barriers to building aren't as binding, you'll find that increases in charged rents are pretty minor. But if you extrapolate from that period to now, you just might be making a mistake.

I also worry that there's not been quite enough attention paid to the long term fiscal outlook, where the costs of an aging population drive us into substantial net debt from 2030; to effective marginal tax rates, which barely changed; or, to the small-scale fiscal discipline that would avoid cash giveaways to the film industry in favour of measures addressing either of the two prior points.

The changes to the tax thresholds only partially adjust for inflation since they were last changed in 2010. The increase in the threshold for the 17.5% rate overadjusts for inflation, but the 30% rate's threshold was underadjusted, and the 33% rate's threshold wasn't touched.

My piece at the Spinoff, last week, covered what would have been needed for inflation adjustment to those thresholds:

Suppose that the government adjusted the tax thresholds to account for wage inflation since 2010. The top income tax rate would then kick in around the $83,000 level, the 30% rate would come in at around $57,0000, the 17.5% rate would apply from about $17,000 and the bottom rate would apply below that. Treasury’s tax calculator says this would cut just under $1.9 billion from government revenues. For the same drop in government revenues, every tax rate could be cut by a percentage point and the 17.5% rate could drop by two points. Everyone would get to keep a greater fraction of the next dollar earned.

If we rank options, here's my preference ordering:

• Adjust for inflation automatically by knocking income tax rates down by half a percentage point whenever accumulated wage inflation warrants it. Treasury's calculator provides the full-year costs of a one percentage point change in each of the tax rates. Whenever inflation gives the government an extra $740m in revenue through fiscal drag, cut each of the rates by a half point - at least on the current costings. 
• Adjusting things this way means everyone sees a change in their marginal tax rate, with consequent dynamic benefits for growth. Adjusting the thresholds provides a big marginal tax change for a small group, and large inframarginal changes for everyone else.
• Adjust for inflation automatically through annual changes to the tax brackets.
• Let politicians pretend inflation adjustments are tax cuts in an election year. We're probably stuck with this one because politicians really really like being able to announce tax cuts in election years.

And that gets us to the pure politics lens, which has the budget as a triumph. That lens is boring and has been done to death already. 

If we run it instead through a "what's the best we could have expected given that it's an election year and given the political constraints" lens, it's not too bad. Just consider how much better things are here than in Australia - the subject of my Australian Financial Review piece on the budget.

Paying for roads

National has announced that petrol excise will increase for the coming three years. Some of my Twitter stream has been suggesting they're doing this to patch up the budget rather than to cover roading expenses. Can't it do both?

The most recent year-end financial statements, those for the year to June 2012, had the government receiving:

• Road user charges of $1,045 million
• Petrol fuels excise of $1,478 million ($847 million on domestic production; $631 million on excise-equivalent duties on imports)
• Motor vehicle fees of $175 million.

So about $2.7 billion in revenues for roads.

Vote.Transport in Budget 2012 had just under $3.4 billion for the National Land Transport Programme, of which a substantial portion was a loan from the Crown for cashflow management. A quick adding-up of the real expenditures on road related stuff looks like $2.6 billion in road spending - about what they collected in revenue. I expect that the loan that's on the books is for expenditures perhaps having been front-loaded during the year with excise dribbling in throughout the year, but perhaps somebody with more familiarity with the Crown Accounts can correct me if I have that one wrong. 

Simple cost inflation would require that petrol excise and road user charges increase if next year's roading expenditures are the same in real terms as they were this year. If the government has any plans on spending more on roads next year than it did this year, then excise has to go up by more than that. I note also that Note e of the Financial Statements, page 129, includes the following:

"Other earthquake costs do not include costs associated with the future repair of local roadways. This exclusion reflects that the first call for funding these future expenses will be from dedicated ring-fenced revenue in the form of road user charges, fuel excise duties, and registration fees paid to the New Zealand Land Transport Fund. Should the Government's share of the costs associated with the future repair of local roadways exceed the amount available from that ring-fenced revenue, the Government has a number of options to allocate future to this expense. The Crown's share of the costs for local roadways remains uncertain, as is the range of funding options available to the Government."

If the Crown's share of roading rebuild costs has increased, then petrol excise rising to cover it isn't crazy either.

A reasonable argument against the excise increases would be an alternative of issuing roading bonds to be paid off over a longer period of excise revenues, so the burden of current road construction is spread across future road users. And that would be reasonable for one-off shocks like the Christchurch earthquakes. But if every year we have to spend some amount to maintain the stock of existing roads, and some amount to build new roads to match population growth, and if the annual increases in the roading stock are roughly the same from year to year, then the annual burden should work out to being roughly the same under either regime.

I don't think there's sufficient evidence to conclude that National's ramping up petrol taxes as a deficit-fighting measure, except inasmuch as it allows the government to continue building roads without drawing on revenues coming from outside the National Land Transport Fund. But I suppose that next year's budget will reveal whether that's right. It would be rather disappointing if the government were using petrol charges to raise funds for other purposes, but I expect that will not turn out to be the case.

As for whether the Government will produce a surplus ... iPredict says there's a 15% chance for 2014/15.

Update: Liberty Scott posts something useful in the comments; I'm hoisting it up here. If the new roading expenditures are a one-off rather than part of an ongoing set of construction projects, then it really makes far more sense to use debt to finance them.  Here's Liberty Scott:

The fundamental problem is that the current capital expenditure on major roads is a one-off, as it really is a bunch of projects that have at best marginal economic benefit. There wont be equivalent major projects ever built again, so there is a major problem with the PAYGO funding system paying for capital the year it is built, even though that capital has a depreciated life (particularly if you consider earthworks and tunnels which largely never deteriorate) of 90 years +.

What should have been done is that the state should have borrowed for those projects and paid for them over many years, but MoT has been almost wilfully blind to this issue. What ought to happen in future is that there should be less long term spending on roads, because the network will largely be built out by the time all of the RoNS are finished.

A footnote

A switch from passive to active voice would have had me as a footnote to the tobacco excise increase RIS:

On the narrow  fiscal  grounds of covering the costs smokers impose on government, further increases in tobacco excise may  not be justified.  At over $1.3 billion per year, tobacco excise revenues may already exceed the direct health system costs of smoking¹⁰.  When the broader fiscal impacts of smoking are considered (eg shorter life expectancy reducing  smokers’ superannuation and aged care costs), smokers are probably already “paying their way” in narrowly fiscal terms.

¹⁰. A recent Ministry of Health study estimated health costs of smoking at up to $1.9 billion per year (15% of the Vote Health).  While this estimate used more detailed data analysis than previously available to estimate health costs, it is well above previous estimates (a  2007 estimate put the cost of smoking to the health system at $300-$350 million per year) and its methodology for comparing lifetime health care costs has been contested. [emphasis added]

I still contest that measure as it relies on an assumption that smokers otherwise would never impose end-of-life costs on health budget. I also have an OIA in for more of the background workings on the $1.9b estimate.

For a more efficient front line

New Zealand's headed for a pretty austere budget; the government's looking for front-line efficiencies:

The Government's "general strategy" remained trying to move money on to front line services, but English signalled pressure would go on for savings there too.

"Where the front line is inefficient, of course, it will have to be sorted out," he said. 

"We're not offering any guarantees ... Just by way of example, if you can give the police mobile technology, then you could get more front line hours with the same number of police - so it's more policing, safer public and we cut costs."

If we're really looking for front-line police efficiency, allow police to stop worrying about marijuana and instead focus on violent crime. Reallocate resources currently devoted to big grow op busts and put them into areas where social costs are real and substantial.

The Law Commission reported that cannabis enforcement cost the police about $116 million in 2005/06; it would be surprising if current figures were lower. Back of the envelope, but if that were shifted over into violent crime, we might expect some pretty serious savings. Treasury estimated a 1% decrease in violent crime saves $25 million in social costs; if the elasticity of crime with respect to police expenditures follows American estimates, which I'd take as a lower bound given that they're further out on a likely concave production function, we'd reduce violent crime by about 11% and social costs of violent crime by about $250 million.

Even better, legalize marijuana but regulate it similarly to alcohol and tobacco with restrictions on age of purchase and on permissible sales venues. Set an excise tax to keep the selling price to consumers about where it is now if demand-reduction is government policy. It's not simple to figure out what that tax should be because it's not obvious what production efficiencies could be achieved in a legalized production environment. But back-of-the-envelope numbers suggest excise revenues a bit over $150m (with a very wide confidence interval; I'd be very surprised if it came in at less than $100 million or at more than $300 million). If production costs in a legal environment were $30-$50 per ounce, and if we assume distribution costs roughly on par with production costs, then a legal market would deliver product to consumers at about $100 per ounce; if weed currently sells at about $300 per ounce, then excise and GST* could total about $200 per ounce. A 2001 estimate put the black market in marijuana at around $190 million in 2001; if marijuana follows the CPI, that's about 830,000 ounces. At $200/oz in tax, that's about $165m.

Recall that drug use in Portugal has gone down, not up, since they abandoned prohibition.** In that case, tax revenues are less than I'm here ballparking.

* Normally we'd ignore the GST revenues collected as they'd just be a shift in spending from other GST-subject goods. But moving to a legal market would bring in net new GST revenues. I discussed some related issues here.

** They have not gone to full legalization; read the Cato summary.

What if?

U.S. House Budget Committee Chair Paul Ryan asks a good question:

Let me ask you a question: what if your President, your Senator and your Congressman knew it was coming? What if they knew when it was going to happen, why it was going to happen and more importantly, what if they knew what they needed to do to stop it from happening and they had the time to stop it? But they chose to do nothing about it, because it wasn't good politics?


What would you think of that person? It would be immoral.

This coming debt crisis is the most predictable crisis we've ever had in this country. And look what's happening. [emphasis added]

Now, I've not been following American budget politics and am consequently agnostic about whatever budget plan he might be proposing; I'd be surprised if it made more sense to cut Medicaid, which helps poor people, than to cut cut Medicare, which helps old people many of whom are wealthy. There's more informed analysis of American budget politics elsewhere.

But I really liked the question. Recall New Zealand Prime Minister John Key's position on maintaining former Prime Minister Helen Clark's policy of zero-percent student loans:

Charging interest would bring in considerable extra revenue for the Government, but Key said he would be voted out if National did so.

"Bluntly, if you want me to be really crude about it there are 565,000 student loans out there. If we add interest back on the student loans, it doubles repayment time of the loan.

"If your loan is $50,000, and it's estimated it will take you eight years to pay it off, we effectively turn it into a loan that is about $90,000 with interest that takes you about 15 years to repay," Key said.

"That is about the only thing that will get [young people] out of bed before 7 o'clock at night to vote, but it's not politically sustainable to put interest back on student loans. It may not be great economics, but it's great politics. It is a bit of a tragedy because it sends the wrong message to young people, it tells them to go out and borrow debt." [emphasis added]

What do we think about this person? Here's what Matthew Hooton thought.

At least Key's not pretending it's good economics; politicians more typically convince themselves that whatever's politically popular is also economically sound.

But instead of leading a national conversation about the size of the deficit and ways of addressing the problem, including reintroducing interest on new loans and phasing interest back in on existing loans over time, he shuts the door on sound policy. Recall Justin Wolfers' argument against forgiving student loans; zero percent loans aren't that different - they forgive a good chunk of the proper present value of the debt.

Budget constraints do bind; Key last year suggested limiting access to loans rather than charging market interest rates. The coming budget is likely to have more restrictions on loan access: rationing the zero-interest credit. And such restrictions aren't necessarily bad economics either: if government-backed loans solve a market failure in credit markets that stems from inability to collateralize human capital, restricting people with other assets that could serve as collateral from accessing zero percent loans reduces costs without any particular policy cost. Or, if some degree programmes or student cohorts are exceedingly unlikely to see sufficiently enhanced earnings for the policy cost to be recouped through the tax system, restricting loan access there also doesn't necessarily do harm relative to a policy of loans issued at market interest rates. But it smacks a bit of government picking winners rather than letting students make their own choices after looking at the expected loan repayment burden at market interest rates.

And, there are other negative consequences for tertiary policy that flow directly from maintaining zero percent loans. The Government caps tuition increases, ostensibly for the students, but really because it can't afford to dish out loans at zero percent were tuition to rise. Similarly, the government has moved to limit domestic tertiary enrolment via capped enrolment at the universities - a policy that would be superfluous if tuition were higher and students charged market interest rates.

I sure hope the budget has something other than "Yeah, the deficit is really big. But everywhere we looked for savings would have somebody complaining, and Labour would then get elected and boy wouldn't that be worse. So we're not really going to do anything - not even commit to raising the age of superannuation eligibility two decades from now. It's horrible economics and boy the Treasury guys will get mad at me. But it's great politics."

iPredict says there's a 43% chance National wins government again in 2014. Pulling out a couple of high variance plays might not be crazy.

Why oh why can't we have a better press corps?

(with apologies to Brad DeLong).

Eric has covered off last week's budget well, so I had thought of writing a parody of the reaction of different interest groups, in the spirit of Fred Dagg's farmers' lament about the lack of a black-singlet subsidy or a gorse retention scheme in one of Muldoon's budgets. But this sort of response is already self-parody, so what is the point?

So instead I am going to put in a plea for better media coverage of future budgets. Here are some suggestions:

1. Don't publish any press releases from interest groups; instead, interview the spokespeople and ask each one two questions: First, "please comment on those parts of the budget that did not relate to your sector"; and second, if you had to put the same total resources into your sector as in this budget, how would you have allocated it differently?".
2. Don't publish any press releases from opposition parties; instead, interview the leaders or finance spokespeople and ask each one, "please state the areas in which you would have spent less money".
3. In a one-marshmellow-now, two-marshmellows-later exercise, promise to give twice as much coverage to anyone whose response is "I haven't had time to fully digest the information yet, let me get back to you tomorrow with a more considered response".
4. Simply refuse to quote any statement with the words "bold", "imaginative" or "Titanic" in it. (To be fair, on the last of these, I didn't see a deckchairs cliché this year, but it will be back.)

Other suggestions are welcome. And while we are at it, could someone please ask Phil Goff why, if a fall in unemployment following a small increase in the minimum wage during a boom period is post hoc ergo propter hoc evidence that a large increase during a recession would not reduce empolyment, why he isn't promising to raise the minimum wage to $20?