Friday, 20 August 2010

Were we reckless spenders?

Tom McDonnell: Tuesday’s €1.5 billion bond auction by the National Treasury Management Agency ensures that Ireland will successfully get through the year without defaulting.

One rare and related piece of good news is that Ireland has today dropped out of the top ten list of countries most likely to default (we were ninth as recently as Tuesday). The odds on us defaulting/restructuring are now just over 20 per cent. Greece, in contrast, is still considered to have a better than even chance of defaulting. Nonetheless the overall picture is still pretty grim and this brings us back to how we as a country tax and spend.

In a previous blog I looked at where Ireland prioritises its spending. The table of public spending shown below (click on it for a bigger view) compares Irish spending to EU 15 spending across the functional categories of government; for example, defence or health. Blue boxes show where Ireland spent a smaller proportion of its GDP on a particular functional category and orange boxes show where Ireland spent a higher proportion of its GDP on a particular functional category.

2007 is an interesting year because it provides a snapshot of spending just as it was just before the economic crash. In 2007, Ireland’s public spending/GDP ratio was just 79 percent of the EU15 average public spending/GDP ratio although this figure rises to 93 percent if we choose to use GNP as a better measure for Ireland (see table below). These numbers appear to indicate that Ireland’s public sector was smaller than European norms before the crash.

The Eurostat data shown below indicates that Ireland had the lowest level of public spending in the EU15.

However if we break the figures down by functional category we get a more complicated picture. Ireland (measured as public spending/GDP) spent relatively more than the EU 15 average on housing and community amenities; environmental protection; economic affairs (primarily physical infrastructure) and health. If we use GNP instead of GDP for Ireland then education spending and public order spending are also seen to have been above the EU 15 average.

The very low unemployment rate prior to the crisis had kept public spending on social protection, which is by far the largest area of public spending, very low prior to the crisis. In GDP terms social protection spending was just three fifths of the EU 15 average. This was the main driver keeping overall public spending below the EU 15 average. Our relatively mild debt burden (part of general public services) also helped in keeping our public spending at low levels.

It is self-evident in retrospect that such a low level of social protection spending could not have been maintained in perpetuity. This is because levels of social protection spending move counter to the economic cycle. We were at the peak of the cycle in 2007 (a precipice as it turned out) and therefore social protection spending was at a natural trough. The figures for 2010 will paint a very different picture.

Public spending only tells half the fiscal story. I’ll turn to the tax revenue side next week.


Michael Burke said...

Great post.

Antoin O Lachtnain said...

But did we spend enough? Did we spend too much? We have to consider our context. For instance, our population aged 65 and over is around 11 percent, which is half the average of the Union. It makes sense that we would be spending a lot less than average on health and social protection as a result. On the other hand it would make sense to be spending more on education since the population is young.

It would make sense to me to compare euro cost/capita rather than all these comparisons with GNP and GDP. That would give us a better idea of how effectively we are using actual money and whether or not we are indeed reckless spenders.

Tom McDonnell said...

@ Antoin:
Did we spend enough? Did we spend too much? You're right it depends on the context. Every country has different characteristics and that implies there is no 'one size fits all' best practice breakdown of spending.

I'll drill down into the different functional areas of spending in future blogs.

As to measuring how effectively we are using actual money? Choosing Euro cost per capita as your effectiveness indicator suffers from the same problem as using public spending/economic output as your indicator. The problem is that both are input variables. To measure effectiveness you also need output and/or outcome variables, for example, level of literacy or level of child poverty.

It is the output/input ratio that lets you measure effectiveness.

But there is no correct single answer to either of your initial questions. It depends on your goal or set of goals. Do you want to maximise the overall utility of society? Perhaps you want to minimise poverty? Or do you want to maximise economic output? and if so, now or in ten years time?

Ultimately the type of outcomes you want, and your willingness and ability to pay for that type of society, will determine the appropriate level and breakdown of public spending.

As to whether we were reckless, well recklessness will be a function of an economy's long term (whole economic cycle) ratio of public spending to public revenue raising.

Michael Taft said...

Tom - look forward to your 'drilling' in future posts. One point: all things being equal, there's no necessary reason to take GNP is a better measure of economic performance than GDP. GDP, after all, measures the totality of activity in the Irish economy. To exclude outward investments from indigenous companies, remittances, capital / profit flows is something that no other economy does when measuring levels of spending, taxation, debt, deficit, etc.

Most EU economies, including EU-15 economies, find their GNI (GNP plus EU funds) to be lower than their GDP, yet GDP remains the standard measurement throughout the EU. If profits are generated here it is irrelevant that they are repatriated (or placed in a deep hole or burned or donated to the developing countries) when we are measuring our economic activity.

Of course, there is that caveat - if profits are generated here. If they are not then they should't be in our GDP in the first place because they are counted in some other GDP. Therefore, it is our GDP that is faulty. Excluding net factor payments doesn't compensate; it merely glosses over.

So, until such time as the Government faces up to this issue and provides a reliable calculation of genuine economic activity, GDP should remain the standard benchmark - just as it is for the Maastricht guidelines.

Tom said...

On the GNP/GDP question, there's another issue that seems to me to be constantly overlooked, which is this: whatever measure we use it has to be the same for every economy.

If there's an argument for using GNP as a better indicator for ireland, it still has to be compared to the GNP of other countries, not their GDP.

The figures vary or tend to vary to some extent for all countries. That Irelands variance is high doesn't mean that the variance in another country is irrelevent.

The difference between what is considered high spending and low spending is often only a few percentage points, so any distortion created by a false comparison is important.

Anonymous said...

apologies for the incorrect forum.

Can I suggest that TASC set up a group facebook page.

Antoin O Lachtnain said...

What is wrong with euros/capita exactly? If you want a competitiveness benchmark, that seems like the one.

Maybe you cannot measure outcomes, but you can measure outputs and compare them to other countries. For instance, if we compare our health spending/capita to (say) France, can we see in broad general terms whether we have a comparable standard of day to day care? A more sophisticated comparison could then take into account issues like demographics.

Tom mcDonnell said...

In principle I wouldn't at all be opposed to use something like... public spending expressed as Euro per capita in constant prices and adjusted for purchasing power parity... as a primary input variable at the macro level.

However as you drill into the cofog areas you quickly find that it becomes necessary to unpackage public spending into its various component parts, and only some of these are relevant to your outcome variable.

In one sense the cofog work is designed to identify and quantify these input variables
e.g., if we can identify every line item of public expenditure that is associated with day to day care, we can track the level of spending in this area as it changes year on year - and if we can identify and quantify relevant output and outcome variables - we can then also measure effectiveness over time and from country to country.

It should also become possible to identify what 'type' of public spending is effective in achieving desired outcomes.

Antoin O Lachtnain said...

Why would you use purchasing-power parity? This is a small, open export-oriented economy. That puts us under pressure to have good value public services not in relative terms but in absolute terms. That is what we should be looking at.

The other effect of using purchasing-power parity is that it would let the private sector off the metaphorical hook for its part in driving up public sector costs.

It is unrealistic to expect to be able to analyse down expenditure line-by-line in all European countries. That would take years. We are probably stuck with whatever Eurostat have produced.