Some evidence about health expenditure in the Republic of Ireland

Posted on February 02, 2018 by Tom Healy

Tom Healy, Director NERI
Tom Healy, Director NERI

Never let the data get in the way of a good story. In their landmark book, How Ireland cares: The case for health reform , published in 2006, Dale Tussing and Maeve-Ann Wren recount some wild claims by a Minister for Health regarding the extent to which Ireland was spending far in excess of other countries.

Surprisingly, the claims along with the underlying assumptions have stuck. Sometimes, it is suggested that public policy is oblivious to data for much of the time. In this case, it would appear not. When the Great Recession hit in 2008-2011 the champions of austerity (remember ‘there is no alternative’) were not slow to claim that Ireland spent far too much on health and that we had, in particular, far too many nurses per 1,000 of population. At least some of the decisions made during that period in relation to pay, recruitment, capital investment and closure of hospital wards were to have serious and long-term implications for many people in Ireland.  Apart from the fact that authoritative OECD data did not and do not bear out the often advanced claims about nurse numbers, there are well-known limitations in regards to the count of practising nurses in Ireland suitably ignored by many commentators and about which I will return in a future blog. For now, let us concentrate on the big picture in regards to spending on health.

We are blessed in having – since last year – a new System of Health Accounts (SHA). This has changed the game in relation to international comparisons. Unfortunately, in the case of Ireland, the new data system only kicks in from 2013 while the latest available year is 2015 ( Ireland's System of Health Accounts, Annual Results 2015 ). However, Trojan work has been done by the relevant public agencies to implement (successfully) SHA for the purposes of reporting to Eurostat and OECD.

SHA is not perfect and there are lingering issues especially in regards to the boundary lines between ‘health’ and ‘social care’. In the case of Ireland, ‘social care’ elements embedded in long-term nursing care settings (for example) are counted in under ‘health’. This seems reasonable although it is not clear that other countries are as careful in this regard.  SHA covers all aspects of health under three major classifications:

  • Who pays (Government, private insurance of out-of-pocket households)?
  • Who provides (hospitals, GP surgeries, long-term care, etc.)
  • What type of care if provided (curative, inpatient, outpatient, etc.)

Following the line that Ireland ‘overspends’ on health, the European Commission during the years of fiscal austerity took the puzzling step of focussing on Gross National Income (GNI) rather than GDP as a benchmark for expenditure comparisons when it came to health. This is odd because, for just about everything else, GDP is the definitive measure (adherence to budgetary rules, fixing of contributions to the EU budget for Ireland, etc.)

A comparison of spending in 2013 shows Ireland (the Republic of) to be close to European averages when it GDP is used as the point of reference but significantly above the average when we replace GDP with GNI (Chart 1).


Point proven?

Roll on to 2015 (Chart 2). Something very odd has happened in the meantime. Health spending as a proportion of GDP or GNI was in freefall under both measures. The explanation rests in The Great Statistical Leap Forward in 2015. This is when activities by a very small handful of large multinational companies with regards to intellectual property and overseas contract manufacturing had a sudden and unprecedented impact on GDP and on GNI. Apart from the unreliability of GDP and GNI for any meaningful international comparisons, there was the matter of age-structure in the Irish population. Since Ireland has a relatively younger population, it ought – other things equal – spend less than other countries. This is a fair and important point although the exact extent of age-related bias in spending is difficult to measure and possibly exaggerated (a matter to which the NERI will return in the near future following a paper by my colleague, Paul Goldrick-Kelly: The Fiscal Implications of Demographic Change in the Health ).



We can only work with the data we have. One way of measuring health spending is to divide total current health spending by total population across a group of reasonably comparable countries. What do I mean by ‘reasonably comparable countries’? I suggest the following ten countries by virtue of (a) their membership of the European Union at this time, (b) their overall level of income and income per capita, and (c) their cultural and geographical proximity to Ireland: Austria, Belgium, Denmark, Finland, France, Germany, Ireland, Netherlands, Sweden and the United Kingdom I call these the Northern European Comparator Group (NECG) identical to the ‘peer group’ used by my colleagues Paul Goldrick-Kelly and Tom McDonnell in Taxation and Revenue Sufficiency in the Republic of Ireland save for the inclusion in the case of the latter of Luxembourg. In case anyone thinks I am trying to cook the averages by suppressing the 18 other EU Member States I show an average for the whole EU (EU27 because for some reason Malta is not included in the data for the EU28 countries). Chart 3 shows a comparison for 2015 on spending per capita adjusted for price differences so that, as much as possible, we are comparing similar values of spending per capita across the range of countries compared.



But, what if we look a little under the bonnet at the components of expenditure? Perhaps, there are areas where Ireland is out of line in such a way as to suggest structural inefficiencies? Let’s look at chart 4. Here, I have selected some of the major components of health spending using Eurostat data and standardised SHA measures. 


There is nothing to suggest that Ireland is out of line on any of the major components. The fact that spending, here, on day curative and rehabilitative care is relatively low while inpatient spending is relatively high is difficult to explain and will require further investigation. Total spending on administration (including and excluding the private insurers’ profit margins) is not high in Ireland. So much – at least at this very aggregate and macro view of health spending – for claims that the system is sclerotic and overstaffed by administrators. There are many issues to do with the structure of Irish Health that require urgent reform (as ably spelt out in the epic Sláintecare report published by an Oireachtas Committee last year). It is surely the case that the use of public monies in hospitals could be better allocated and in a way that levels the playing pitch for ‘public’ and ‘private’ patients (why can’t we be all just ‘public’ patients in a well resourced and efficient health care system?). Hence, the fact that Ireland falls below the statistical average for the NECG group of countries does not establish that we spend wisely or efficiently in all cases. In fact, we need to up the game and spend more in order to spend less than might otherwise be the case in the long-run.

Societies based on high levels of productivity, excellent public services and more egalitarian distributions of wealth and income tend to be healthier societies. The sick society is one where the privileged few (or many as the case may be) have access to the best treatment on the basis of ability to pay over need. That is no way to run a country especially given our high levels of income per capita compared to most EU Member States.

None of the above discussion is to be taken in any case to suggest that Ireland’s health system is efficient compared to other countries or that there are not areas where savings could be made. In a forthcoming NERI publication we will suggest that some major long-term cost-saving policies should include a transition to a single tier health universal health service, a drive to educate and promote healthier lifestyles and choices and investment in primary health care to keep as many people as possible for as long as possible out of acute hospitals (which unavoidably eat health budgets). Moreover, the relationship between spending on health and health outcomes (e.g. mortality and morbidity) is a complex one related to many factors and not just spending. What I am suggesting is that those who claim that we over-spend should pay closer attention to the data – limited and partial as it is.

So, when the next recession comes as it will (at a year and a decade we do not know and at a scale and impact we cannot foresee), beware of the austerity hawks who know a little about a lot but ignore wider economic, ethical and institutional realities.

Digital Revolutionaries