With today's Budget expected to herald a return to surpluses – skinny next year but plumper in the following years – it is time to start thinking about what to do with them.
Broadly there are three options: pay down debt, increase government spending or cut taxes.
The Government has already told us its policy on debt reduction. It is to bring it back down to 20 per cent of gross domestic product by 2020 (from 28 per cent now).
And we got an indication of its view on spending from Finance Minister Bill English last weekend. Since 2008 government spending had dropped from 35 per cent of GDP to 30 per cent, he said, "and we want that to be dropping to 26 per cent and 25 per cent in the next six and seven years".
No doubt that went down well with the party faithful he was addressing. But should the rest of us applaud?
It would represent a substantially more ambitious/onerous objective than the 30 per cent of GDP enshrined in the current Budget Policy Statement.
Let's remember that in 2008-09 when the ratio of government spending (core Crown expenses, to be precise) hit 35 per cent of GDP the economy was in the deepest recession since the 1970s.
The average for this indicator for the six years before that was just under 30 per cent. The minister is taking a (deserved) bow for getting it back to the levels prevailing under the supposedly spendthrift previous Government.
But a target of getting the spending-to-GDP ratio down to 25 per cent, while at the same time paying down debt and (eventually) resuming contributions to the Cullen Fund, implies a continuation at least, if not an intensification, of the kind of spending restraint evident in English's previous Budgets.
In the first nine months of this fiscal year compared with the same period of the previous year tax revenue was up 6.3 per cent, as you would expect with the economy now growing at an above-trend rate, plus a bit of inflation boosting the tax base as well.
The Government's operating spending, on the other hand was flat.
That is a bit misleading, however, as the previous period included some hefty one-off expenses related to earthquake damage to Canterbury infrastructure. If you take that line out of both periods, there was an increase in government spending of 1.7 per cent.
That is just 0.2 per cent when adjusted for inflation in the year ended March.
And the population grew by more than 1 per cent over that period. The statisticians have already told us the working age population grew by 50,000 in the latest March year and that does not include those under 15.
So in real per capita terms spending was cut.
The problem with a crude spending-to-GDP target is that it does not take account of demographics.
In particular it does not take account of the changing size and composition of the dependency ratio.
The dependency ratio is the young plus the old divided by the working age population.
First there is the ageing of the population. There are 4.2 people aged between 20 and 64 for every one 65 or older. By mid-century it is projected to be 2.5 people, with obvious implications for healthcare and superannuation costs.
Helpfully in this context New Zealand's fertility rate (the number of children per woman) is the second-highest in the OECD.
But that has implications for the demand for education. We spend less than the OECD average on education and while the average outcomes look respectable by international standards, a large minority of kids are not doing well at school. We also spend less than the OECD average on healthcare but are above average on one measure of effectiveness, life expectancy. That suggests the health sector should not be on the waiting list for fiscal liposuction either.
Between them health, education and superannuation account for more than half of government spending.
The population is also blown around by large flows of migrants in both directions.
Right now there is a strong net inflow, reflecting fewer Kiwis leaving and more returning.
The net inflow hit 32,000 in the year to March and economists, at ASB and Westpac for example, expect the annual gain to reach 40,000 late this year, adding nearly 1 per cent to the population.
But migration is very volatile. Two years ago we were suffering a net outflow. And it affects the demand and supply sides of the economy in a number of ways and with different lags.
A more fundamental reason for doubting whether public spending is as compressible, with no harm done, as English's 25 per cent of GDP target implies, relates to what one might call the country's social productivity.
There is some prima facie evidence at least that we are reasonably good at turning a not very impressive national income into decent social results that rank highly in international league tables for such things.
The 2014 Social Progress Index, published last month by a not-for-profit organisation in the United States, the Social Progress Imperative, ranks 132 countries based on more than 50 indicators of social and environmental performance.
New Zealand tops the rankings, whereas we rank 25th in GDP per capita.
While some of the indicators have nothing to do with the level and quality of government spending – the suicide rate, for example, or religious tolerance – most of them do, one way or another.
No one would have any trouble compiling a list of counter-instances to the general proposition that New Zealand is a decent place to live.
The report itself points to persistent disadvantage experienced by Maori and adds that "the place of children in New Zealand is also of concern with suggestions that the lives of up to 20 per cent of New Zealand children are neither as safe or as nurturing as they should be".
English's standard response to concerns like these is that it is the quality and effectiveness of government spending that matter and that we need to move on from the idea that you show you care about this group or that by just spending more money on them.
The question is how far that approach can be pushed before it crosses the line into false economy.
The argument that countries do better with lower ratios of public spending to GDP depends on how you define doing better.
It has to be broader than maximising growth in GDP.
The Social Progress Index is part of a worldwide trend towards a broader measure of welfare and progress, which has even reached the New Zealand Treasury.
The Government deserves some credit for quantifying in its better public services framework some of the results it is aiming for. That makes a lot more sense than aiming for some crude target for spending to GDP.