It may come as a bit of a surprise that there really isn’t any readily available database that shows the fiscal record of any particular administration. We will have to construct this for this topic, and there will be some complications involved in doing so.
For today, however, I just want to use some currently available information to present a bit of the ‘big picture’. I’m going to use Core Crown Expenses, which is the standard measure, and look at the period from 1998/99 to 2008/09. I think that timeframe is the best one for a simple analysis, but no year-to-year analysis is perfect. The budget for 1999/2000 was actually set by Bill Birch in the Shipley government, but the incoming Clark-Anderton government committed significant additional spending during 1999/2000 ahead of their own first Budget in June 2000. Similarly, the current government made significant changes to the 2008/09 appropriations but the budget for this was originally set by Michael Cullen.
Plus, of course, the Core Crown Expenses are a mix of deliberate initiatives, responses to cost and salary pressures, and automatic adjustments (benefit CPI increases, additional superannuitants etc). With these caveats in mind, then, let’s look at the figures.
The graph above shows the growth in Core Crown Expenses from the 1999 fiscal year (1998/99) to the 2009 fiscal year (2008/09). The yellow line is the CPI index for this period, so the area above that line represents a real increase in government spending. Anything below that can be argued to be simply reflecting general price inflation.
We can see that expenses increased from $34 billion to $62 billion during the period. That’s an increase of $28 billion, or almost a doubling of Core Crown Expenses. Probably about a quarter of that, or $7 billion, can be accounted for by general CPI inflation, so the real increase would be about $21 billion.
Where did all that additional money go to? We have data breaking the money down into Core Crown Expense Classes, a slightly odd grouping of spending areas that don’t seem to be used anywhere except the Crown accounts.
The graph above presents the composition of the $28 billion increase by Core Crown Expense Class. It shows that the two largest areas of additional spending were ‘Social Security and Welfare’ and ‘Health’, followed by ‘Education’. Between them, these three classes accounted for almost two-thirds of the total $28 billion increase.
Such broad categories, however, raise as many questions as they answer. For instance, what drove that extra ‘Social Security and Welfare’ spending? It might be thought that it went on additional benefit expenditure, but actually it was more likely to have been a combination of an ageing population increasing the cost of Superannuation, and Working for Families tax credits.
Also, since some Expense Classes are much bigger than others, those large slices of pie above may simply reflect the fact that those areas made up a large proportion of spending to begin with.
To test that, let’s use an index approach to show the growth pattern in each Expense Class over time irrespective of the overall size of spending in each.
This shows that, when viewed in relative rather than absolute terms, growth in ‘Health’ and ‘Education’ was actually only mid-range amongst the Expense Classes, and ‘Social Security and Welfare’ was one of the lowest-growth areas.
By far the fastest growing Expense Class was ‘Economic and Industrial Services’ followed by ‘Transport and Communications’. ‘Defence’ by contrast has grown quite slowly — however Core Crown Expenses do not include capital expenditure such as military equipment.
All this is just the first layer of the onion. Over the course of this topic I intend to peel further and look at the particular initiatives and expenditure lines that have driven costs in each of these Expense Classes, and also look at other areas like capital expenditure that are left out of this initial analysis. Let me know of any insights you have about how to proceed, or requests about particular areas of expenditure that you feel are important.
Update: I should also reference my data sources. Most of the data comes from the very useful Treasury file Fiscal Time Series – Historical Fiscal Indicators 1972-2008. The data for 2008/09 is from Core Crown Expense Tables in BEFU 2009.
Tags: Core Crown Expenses, Fifth Labour Government, public spending
Really interesting to see by looking at the line graph that all the action (in terms of growth in Core Crown Expenses) really took place post 2005 – the 3rd term. I would have thought there would have been more action before then.
I’d be interested in any dispassionate analysis you may have in the future about the underlying reasons for the growth in these areas post 98/99 – that is, by how much were some expense classes underfunded (by international/OECD measures, I guess). It would be interesting to see how much of the real growth in crown expenses was restoring underfunded public services.
Well done on laying this out nice and clearly.
Not that I’m wanting to create extra work for you David, but to partly answer Andrew’s thinking you’d want to at least create another version of the indexd graph from 1991 (ie when the Crown Accounts shifted to Accrual Accounting) That way you can see the relative spending growth through the 1990s and then this period under discussion.
Certainly it ought to show clearly the reason for the rise in operational defense spending from memory…
Yes, this is a very clear first layer. The relative standing of ‘economic and industrial services’ and ‘transport and communications’ in the line graph cased me to think about might be classed as investment expenditure and what be classed as consumption expenditure. While capex isn’t included the precursors to capex, such as planning, are. It would be useful to know to what degree any government is investing compared to paying for consumption. While a lot of consumption is, as you point out, out of a government’s control, some kind of assessment about thinking and spending for inter-generationally equity might be a marker for progressive view point.
Of course the military policy wonks (you know who you are) will bore you endlessly about the need for good command and signals as well as logistics before you put a front line soldier in the field . . .
Okay – now looking at that spending profile again it seems to me there are at least two things worth noting (and in a way you’ve alrady started to do so David.)
1) Big spending classes will be less effected by the decision by the govt to spend on a bit ticket item. That’s obvious. Whereas some of the smaller classes (eg the sky blue heritage, culture and housing one) will jump with one or two injections. Interesting to be reminded what they were, but only marginally interesting. I’m sure the real story lies in the slower but steadier increase in teh big spending classes (esp health).
2) How does the index graph deal with contingent liabilities. I seem to recall from Transport for example that we loved as a government committing to spending in outyears. We made a lot of that sort of commitment in earlier years in government but the spending itself was only just coming on stream as we left office. That should in part account for the change in magnitude of some of the lines in later years, without making it look that we ’spent it all at the end’ which isn’t really the case.
And I for one (thank Darel) am not going to wade into the military policy wonk field unless others provoke me (ie if Darel stays out, I’ll stay out)….for now anyway
@Andrew Thanks! I take your point about the ‘1990s underfunding’ question, but I suspect the answer is in most cases pretty subjective. A ‘big picture’ analysis similar to this one for 1990s, as James suggests, may be worthwhile, though. Still, my impression/recollection is that the overall numbers in big areas like Health and Education did grow ahead of inflation in the 1990s – we’d probably say ‘it wasn’t enough’ but more based on impression of outcome than it being immediately obvious from the fiscal numbers alone.
@Darel I like the concept of investment versus consumption but suspect there’d be some disagreement about it in practice. Is Working for Families consumption or are we investing in the next generation? Still, the balance of spending between various purposes is definitely something I intend to focus on in Stage 3 of this topic.
@James I generally agree about #1 but I do intend to do justice to both big and small spending ‘classes’. On #2, the overall Spending Contingency is included in Core Crown Expenses but this is only an issue for future years – contingencies do get appropriated for (or cancelled) once the fiscal year in question arrives.
Hi David, interesting numbers which I’ll have a further look at. On the other hand I have some concerns with representing spending in nominal terms as opposed to %GDP. Are you going to show those numbers at a later stage? (And if you have answered that question in the post, my fault for scanning it too briefly!)
No, I didn’t address that in the post, but I will now. I have to say, I tend to go along with Peter Harris’s comments at his recent Fabians lecture:
I don’t think we ought to be defensive about expenditure increasing under the Fifth Labour Government – that’s what they intended to do, and what they were elected with a mandate to do.
Having said this, I’m not averse to including some GDP proportions, as context, for the total spending figures. But the main thrust of this topic is going to be drilling down into particular spending areas and initiatives and at that level I think the %GDP stuff tends to lose meaning a bit.
Re: consumption vs investment
Is there any tools of analysis that others use? I would class WfF as consumption but can see the investment argument too. There are going to be value judgments so maybe if others have developed tools to decide which is which then several analyses can be done and people can choose which they think fits their beliefs better.
Hard.
I’m not aware of any such tools of analysis, but I’ll keep an eye out. If anyone else knows of any pre-existing approach that might be useful here, please leave a comment here or else send me an email.
Hi
I think it’s worth looking at total spending, but also having something like GDP to compare – isn’t the increase in population significant over the period?