Published by AHAA (Apologies for the tabulation presentations.)
Introduction
By way of background, doing some
unrelated work I came across some international data on the healthcare sector
which seemed to contradict my, and the conventional wisdom’s, view of the
healthcare sector. Broadly, it is that the sector has been underfunded. That is
not what the international data seems to say. By international standards New
Zealand healthcare funding appears satisfactory – no worse generally than
average. Why then is there so much grumbling which seems to suggest the
healthcare system is in crisis?
Data is always treacherous
because of data definitions and structural change over time; international
comparisons are even moreso. I proceed with care – and honesty (hence the more
conversational tone). The first part of the paper summarises the data. My
commentary on its interpretation is in a second part.
OECD: New Zealand’s International
Ranking: Health Spending of GDP
Each year the OECD produces a
survey of healthcare indicators. The latest is Health at a Glance 2023.
It contains many comparisons for the 38 OECD countries (and 10 others which,
being much poorer, I ignore). Figure 7: [Public and Private] ‘Health
expenditure as a share of GDP’. Table 1, derived from it, ranks the top 13
countries.
I was astonished that New Zealand
was so high – seventh. I do not put a lot of weight on the exact ranking. It
bounces around from year to year. Moreover, the differences between the 5 and
11 ranking are small after allowing for measurement difficulties. Such cautions
do not alter the conclusion that New Zealand is well above (say 16 percent) the
OECD average at a level not incomparable to that of countries whose healthcare
systems are often admired.
1. Health Expenditure as
Percentage of GDP: 2023
<>
|
1
|
United States
|
16.7
|
|
2
|
Switzerland
|
12.0
|
|
3
|
Germany
|
11.8
|
|
4
|
France
|
11.6
|
|
5
|
Canada
|
11.2
|
|
6
|
Japan
|
11.1
|
|
7
|
New Zealand
|
11.0
|
|
8
|
Austria
|
11.0
|
|
9
|
Belgium
|
10.9
|
|
10
|
Sweden
|
10.9
|
|
11
|
United Kingdom
|
10.9
|
|
12
|
Netherlands
|
10.1
|
|
13
|
Finland
|
10.1
|
|
|
OECD38
|
9.5
|
International Ranking: Comparable
Health Spending per Capita
Table 1 is important for fiscal
managemeny. [1] Three adjustments are made in Table 2 which refines the
understanding for health policy purposes:
–
total actual spending (not relative to GDP);
–
valuing the spending in the common (PPP) prices, removing country-cost
differences;
– converting the spending to age-adjusted per
capita. (A country with an older population has its health spending scaled down
relative to a country with younger population because its health needs are
relatively higher.)
Table 2 shows the outcome for the
top 14 countries measured with OECD = 100. [2]
2. Age-Adjusted Health Expenditure per capita 2022
<>
|
1
|
United States
|
153
|
|
2
|
Germany
|
131
|
|
3
|
France
|
127
|
|
4
|
Czech Republic
|
124
|
|
5
|
Denmark
|
120
|
|
6
|
New Zealand
|
119
|
|
7
|
Switzerland
|
114
|
|
8
|
Luxembourg
|
114
|
|
9
|
Belgium
|
114
|
|
10
|
Norway
|
110
|
|
11
|
Netherlands
|
109
|
|
12
|
Austria
|
106
|
|
13
|
Japan
|
102
|
|
14
|
Canada
|
102
|
|
|
OECD37
|
100
|
New Zealand now ranks 6, one
higher than in Table 1 because its healthcare prices are relatively low and its
population structure is relatively favourable compared to the rest of the OECD.
(Observe how the US score tumbles because it prices are so high; it still
retains its top ranking.) Allowing for measurement difficulties, New Zealand
seems to rank somewhere in the 3 to 9 range.
Moser’s law says if a statistic
is interesting it’s probably wrong. (If it is not, it is really interesting.)
It behoves the researcher to check the quality of the data. The figure which
has most concerned me here was that New Zealand health sector production prices
seemed low (so that the country gets a lot of healthcare bang for its
international buck). However, Japan, Czechia and Slovakia are similarly out of
line.
Even so, I am cautious. New
Zealand health care prices seem to be about 60 percent of the level reported to
Australia. Part of those differences may indicate superior efficiency, but
given over 80 percent of the costs of a healthcare service are for wages and
salaries, it seems possible that New Zealand remuneration rates are markedly
below Australia’s. We know that to some extent that is true, but as much as the
OECD comparison suggests? I return to the cost difference issue in the
commentary.
OECD: Evaluating by Outputs
Rather than Inputs
The statistical conventions
measure the costs of healthcare, not the outcomes. For instance, they values an episode of surgery at the cost of
doing it, rather than the benefit from doing it As far as the conventions are
concerned it does not matter whether the operation is necessary, low need or
high need; whether it is a failure, a moderate success or resounding success;
whether it makes a significant difference such as saving a life, or a small
improvement such as removing an ingrown toenail; whether it is done with
consideration and respect or with brutality. (The disconnect is
well-illustrated by the US top ranking in terms of resources but, notoriously,
its markedly lower rank on its population’s health outcomes.) Statisticians are
well aware of these limitations but there have not been the resources to tackle
them.
The OECD report provides a
dashboard of health indicators too complex to detail. Here are the assessments
about New Zealand:
Better than OECD Average
Self-rated
health
Smoking
Air
Pollution
Effective
secondary care (Acute Myocardial Infarction – heart attack)
Effective
secondary care (stroke)
Close to the OECD Average
Life
Expectancy
Avoidable
Mortality
Chronic
conditions
Alcohol
Population
coverage for core health service
Population
satisfied with availability of quality health care
Financial
protection (coverage by compulsory pre-payment)
Mammography
Screening
Health
spending
Practising
physicians
Practising
nurses
Hospital
Beds
Worse than OECD Average
Obesity
The above list is those where the
OECD provides a judgement. The report provides many more measures. A
well-resourced researcher could interrogate the data (where they are available
for New Zealand) and add to the list. There is also usually enough data to make
ten-year comparisons.
For this paper’s purposes it is
sufficient to conclude that New Zealand health outcomes are usually close to
the OECD average, but there are some areas where the country is seems to be
doing better than average (and perhaps a few where it may be doing worse).
Commonwealth Fund: Health
Performance Indicators
The US-based Commonwealth Fund
confines its international comparisons to ten countries with six separate
aggregate indicators plus an overall performance measure in Mirror,
Mirror 2024: A Portrait of the Failing U.S. Health System: Comparing
Performance in 10 Nations. Its focus is on how poorly the US does
compared to the other nine. Here are its rankings.
<>
|
|
Overall Ranking
|
Access to Care
|
Care Process
|
Administrative Efficiency
|
Equity
|
Health Outcomes
|
|
AUS
|
1
|
9
|
5
|
2
|
1
|
1
|
|
NETH
|
2
|
1
|
3
|
6
|
3
|
7
|
|
UK
|
3
|
2
|
8
|
1
|
5
|
8
|
|
NZ
|
4
|
5
|
1
|
3
|
8
|
3
|
|
FRA
|
5
|
6
|
7
|
4
|
6
|
5
|
|
SWE1
|
6
|
4
|
10
|
7
|
—
|
6
|
|
CAN
|
7
|
7
|
4
|
5
|
7
|
4
|
|
SWIZ
|
8
|
8
|
6
|
10
|
4
|
2
|
|
GER
|
9
|
3
|
9
|
8
|
2
|
9
|
|
US
|
10
|
10
|
2
|
9
|
9
|
10
|
New Zealand does well on all the
dimensions – 4/10 on overall rankings, 3/10 on health outcomes – except on the
equity one. [4]
As a general rule, the
differences between the country measures are not very great, except in the case
of the US. The study tabulates the ordinal measure of ranking but provides only
graphs – not tabulations – of the cardinal measures which underlie them. The
countries generally bunch together – excepting the US – so that measurement
errors, which are inevitable in international comparisons, could change the
rankings. They would not change the basic conclusion that New Zealand is
bunched with the other nine.
Except on the ‘equity’ dimension,
New Zealand does really badly. It is bottom except for the US.
Commonwealth Fund: Health
Performance Indicator of Equity
Supppose the measured difference
between the top of the seven countries above New Zealand (Switzerland) and the
bottom (France) is X. [4] Then New Zealand is almost another 0.4X down (and
halfway between France and the US. Moser’s law suggests this outcome is
interesting but we need to check the data.
The report describes its equity
assessment procedure as:
Our Equity domain reflects how people with below-average and
above-average incomes differ in their access to health care and their care
experience. Australia and Germany rank highest for equity, meaning they are the
countries with the smallest differences in health care access and care
experiences between below-average and above-average income residents. New
Zealand and the US rank last on equity, having the highest income-related
differences in reported cost-related access issues and instances of unfair treatment
or feelings that health concerns were not taken seriously by health care
professionals because of their racial or ethnic background.[5]
The one comment the report makes
about New Zealand in terms of its equity ranking was:
Australia and New Zealand’s poor performance for rural versus non-rural
respondents contributed to their lower rankings. [6]
A reference to the post-code
lottery?
The extent to which the
international equity measure covers New Zealand concerns such as waiting times,
being forced into private treatment by lack of public supply and unmet health
care needs is not clear. They may apply in the comparator healthcare services
too. (One where New Zealand does badly by the comparator standards is that most
have national surveys of unmet health needs.) [7]
Some Local Data.
Before commenting on these
findings, a couple of New Zealand specific data bases are added.
According to the Statistics New
Zealand National Accounts the contribution of the health sector to GDP has
risen from 3.3 percent in the 1971/2 year to 7.2 percent in the 2021/2 year.
(An endnote explains why the number differs from the one the OECD uses.[8])
That is an increase of 0.7 percentage points of GDP a decade. The linear trend
has short term fluctuations around it, but it is consistent with the finding
that healthcare funding has a remorseless upward trend.
A second local source of data is
a release of slides by the Ministry of Health: Health System Changes
Comparative Analysis between 2012/13 and 2023/24: Key metrics. [9] It has
graphs of the following:
–
Vote Health Appropriation time series;
–
Caseweighted Discharges time series (2 slides);
–
Discharges and Bed Days time series;
–
Presentations to Emergency Departments (ED) time series;
–
Clinical and Non Clinical FTE time series (2 slides).
They report that over the 11
years when population had increased 17%:
–
Caseweighted Discharges had increased by 25%;
–
Discharges had increased by 30%;
–
Bed Days had increased by 21%;
–
ED Presentations had increased 29%;
–
Clinical FTEs had increased 39%;
–
Non-Clinical FTEs had increased 59% (or around 37% if a transfer of staff from
the Ministry of Health and the Health Promotion Authority to Health NZ are
deducted).
On these measures, treatments
increased faster then the population over the period. There are other outcomes
not covered by the slides – including the whole of the primary care sector.
It is true that nominal spending
has risen substantially faster over the period, averaging 6.3% p.a. (excluding
COVID-19 funding). The increases were smaller under National (3.4% p.a.) than
Labour (9.6% p.a.) – reported outcomes were rising faster then too. There are
no indications of how much this increase was due to rising prices (including
remuneration catchups).
Part II: Commentary
In order to avoid rambling, I
focus on but four issues.
First, the big conclusion for me
is that by international standards New Zealand healthcare compares well. It is
not perfect, but neither are the comparators (which certainly do not include
the US). Yes we grumble – there are failings – but the aim should be to reduce
the failings (knowing that the comparators also have grumblers and are trying
to to reduce their failings too). But the approach should be of continuous
improvement, not redisorganising – the system is not failing in a major way.
Second, the largest failing New
Zealand healthcare has in international terms seems to been in equity –
especially the urban-rural divide. However this is a tentative conclusion and
needs to be verified – the OECD data base has some material which could be
investigated. There may be a post-code lottery, but we need to understand why
it exists instead of jumping to the conclusion that, say, a centralised system
based in Wellington would resolve the problems facing Kaitaia or the Catlins.
Not surprisingly the centralisation seems to have had no effect on post-code
lottery.
Third, the apparent cheapness of
our healthcare system compared to our comparators may be a concern. Medical
professionals are in an international labour market and while it is is not
perfectly mobile, New Zealand may be losing skilled personnel to Australia. (If
we want international quality personnel, we may have to pay top international
dollar.) Whatever, we do need to know more about international cost
differences.
Fourth, much to my surprise, New
Zealand does not seem to be under-funding its health system. The OECD data in
particular suggests we are funding enough resources – compared to our
comparators – (although insufficient compared to our ambitions). However accounting
conventions are obscuring this conclusion. Following an examination of a batch
of papers relased under the OIA, I concluded that the Minister of Health ‘had
got to stop talking about a “health crisis” and focus on the funding crisis’. The
findings in this paper might suggest the conclusion could have been ‘to stop
talking about a “health crisis” and focus on the accounting crisis.’
Having said that, the previous paragraph might suggest we should be funding
more to retain staffing from migrating. [10]
Of course, there are serious
challenges facing the healthcare sector. Even if there were none we should be
trying to do better. The surprising conclusion from the data is that the
healthcare system seems to be doing comparatively well. [11]
Endnotes
[1] The ranking does not
distinguish between public and private funding.
[2] The OECD does not give this
table. I have constructed it by combining two which it does provide.
[3] The missing country arises
because it was not possible to collect the data for Sweden.
[4] The report offers two
rankings. Reported here is the more refined one. New Zealand does even worse on
the less refined one. That the rankings jump around between the two, warns they
may not be robust. However the US is bottom on both and New Zealand second to
bottom on both.
[5] Some of the input data is
subjective. Subjective data may be culturally biased. For instance, a few
decades ago little attention would have been given to gender differences.
[6] Australian Institute of
Health and Welfare, “Rural and Remote Health,” last updated Apr. 2024; and New
Zealand Government, Rural Health Strategy 2023.
[7] The report sometimes
specifically applauds New Zealand’s healthcare performance on some dimensions.
[8] The OECD figure includes
contributions to health spending from other industry sectors as well as the
health sector. Examples include outsourced food and cleaning services and
building and construction.
[9] On 25 July, 2024 Lester Levy,
just appointed Commissioner of Health New Zealand, stated
‘There’s pretty good evidence over the last 10 years that despite
significant increases in the revenue of this and its predecessor organisations,
the clinical outputs, the clinical outcomes have been relatively flat.’
<><>
https://www.rnz.co.nz/news/national/523162/more-health-nz-job-cuts-to-come-commissioner-lester-levy-says
I immediately put in an OIA
request for the evidence. On 5 November, 2025 (much later than provided for by
the OIA), I received a reply which consisted of five slides (plus two near
duplicates). Their content is reported in the body of the text. Wanting to cite
the source in this paper, I asked for a website reference. The response was:
“I can advise that the
presentation doesn’t have a web-source. However, the information contained in
the analysis is drawn from publicly available information such as the HWIP
and population statistics on StatsNZ website.”
Apparently the ‘pretty good
evidence’ is not worth presenting rigorously.
[10] The international
comparisons say nothing about the efficiency of administration. There is a
belief that New Zealand has ended up with too many layers of generic managers
(who, not incidentally, are far from ideal for managing a continuous improvement
regime). There are current pressures to reduce the numbers – but not the
culture which generated them. However, the substantial downsizing of the public
health unit of the Ministry of Health – ostensibly to release resources for
treatment – may be the opposite of the concern in the following endnote [11].
[11] I footnote the concern that
within the budget envelope there may be big gains from shifting spending from
expensive forms of end-of-life care (such as drugs with only marginal impacts
on extending quality-life years) to the earlier task of prevention and
detection. This is likely to be a common problem in all health systems. A
particular problem is that underfunding primary care adds to the pressure on
secondary care. Earlier we observed that ED presentations were rising at almost
double the rate of the population. It is so typical of the narrowness of
thinking that the targeting is concerned with waiting times at ED, rather than
whether the numbers can be reduced.
<>