PensionReforms' summary and comments
In a PensionReforms' first, we look at a Google Books' version of a publication that can be bought from the publisher but that can also be accessed online from the link given at the foot of this abstract.

The report, published in October 2008, looked at many cross-national comparisons for relative measures of income and wealth in 30 OECD member countries, based on different measures of poverty.  Its general thesis is that, across the OECD, inequalities grew in the decade between the mid 1990s and mid 2000s.  That continues a trend from the previous decade though the disparity today seems "... less dramatic than is often portrayed in the media."

The full report covers different groups in the 30 countries compared.  PensionReforms concentrates its comments on the way different countries treated those "of retirement age" in the mid 2000s.

International comparisons of incomes, wealth, poverty and other economic measures are difficult and must always be treated with great caution.  There are many ways of measuring poverty - relative income poverty, absolute poverty levels, or the headcount ratio that counts the number in poverty (the frequency of poverty).  Yet another measure is the amount by which the mean income falls below the poverty line, measured as a percentage of the poverty threshold (the poverty gap).

PensionReforms suggests that relative income poverty among the older populations seems an appropriate measure for developed countries.  It provides an indication of how that age group might experience living in those countries in relation to local living standards.  However, the way in which housing costs (for owners and renters) are treated can change things dramatically.  An after-housing costs analysis (not used in this report) would help to deal with that difficulty.

In discussing the position of those of "retirement age", the report used an income of 50% of the median equivalised household disposable income as the test of "poverty".  It is, of course, possible to use other percentage levels - 60% of median income is often used.

PensionReforms notes difficulties with the denominator (the "median income" number) used in such a comparison across countries.  The report defines "household disposable income" in a way that matters in the pensions area.  For example, 'social security' contributions paid by employers and employees are treated like payroll taxes, not income.  That seems in order because those contributions are just another way of collecting the money that governments need.  However, presumably, compulsory Tier 2 contributions (such as in Australia) are not treated like taxes even though they will eventually reduce the state's Tier 1 pension.  The same issue applies in Canada..  

It's also important to note that the report looks at household income so that, where a household includes at least one person "of retirement age", the whole household counts in the measure.

With those reservations, PensionReforms reports the position of citizens of retirement age in the mid-2000s and compared them with the equivalent numbers a decade earlier (from Table 5.3 in the report).  The table's highlights are:

 

1. Czech Republic, Netherlands and New Zealand at the top: Czech Republic, Netherlands and New Zealand are three countries that show an overall incidence of poverty in the "mid 2000s" amongst all people "of retirement age" of about 2%.  Only 13 of the 30 countries had poverty rates, on the defined basis, of less than 10% amongst all those of retirement age.  The average amongst all 30 countries was 13%.

 

2. Poverty levels extensive in some countries: There are seven countries where the poverty levels amongst all those of retirement age exceeded 20% and two cases where the proportion was exceptionally high - Korea at 45% and Ireland at 31%.  The next two 'worst' performers were Mexico at 28% and Australia at 27%.

 

3. General improvement: Over the decade between the "mid-1990s" and the "mid-2000s" there was an average fall in the percentage points of poverty incidence amongst those "of retirement age" of 0.7 across all 30 countries.  The stand-out improvers over the decade were Turkey (-8.1), Norway (-6.8), Greece (-6.6) and the Czech Republic (-6.5).

However, there were some backward steps over the decade as well - most notable in this group were Ireland (18.8 percentage points worse), Finland (+5.3) and Australia (+4.6).

4. Households "with a head of retirement age": The comparison also looked at all households where the "head" was "of retirement age".  The following shows the 'top' performers in the various groups with the lowest incidences of poverty:
-   All households: Netherlands (2% incidence);
-   Working households: New Zealand and Norway (1% each);
-   Not working households: Netherlands and New Zealand (2% each);
-   Singles: Netherlands and New Zealand (3% each);
-   Couples: New Zealand and Norway (1% each).

The 'worst' performers for the "Not working" category (arguably the most important group for a retirement income system) were Korea at 69%, followed by - in order - Mexico (39%), Ireland (36%), the United States (34%), Spain and Australia (both 32%) and Greece (31%).

Conclusions from cross-country comparisons are always difficult and the report notes an expected correlation between levels of "social spending" (for the retired, this meant only the "sum of outlays for old-age and survivors benefits") and "poverty".  The three countries where the social spending was lowest (Mexico, Ireland and Korea) also had the three highest levels of poverty.  On the other hand, New Zealand and Netherland both had social spending of less than 5% of GDP and the lowest levels of "poverty".  The report itself did not give sufficient detail to explain this apparent anomaly but it presumably relates to the rather limited definition of "social spending".

The average "social spending" across all countries was 7.5% of GDP.  That does, as the report points out, exclude health expenditure, a matter of some importance for those "of retirement age".

PensionReforms accepts that a different measure of poverty (such as 60% of median incomes) would produce different outcomes, particularly where there are significant retirement income interventions at Tier 2 and Tier 3.  The report focussed just on 50% of median income for those "of retirement age".

PensionReforms would also like to see more work done on the poverty measure.  50% of the median is a crude, single measure and it would be improved if it were based on an after-housing basis.  As the report notes, picking a particular figure can affect the numbers a lot for those of retirement age, given the cluster of cash incomes around the amount of the state pension.

Since the mid 2000s, the situation will have changed for all countries.  For example, the global economic meltdown has significantly reduced values of financial investments and that will affect countries where retirement incomes are more dependent on private provision.

PensionReforms suggests that the report's findings should give some countries pause for thought.  Poverty at any age in a rich country should be a cause for concern.  Poverty amongst the old should be of particular concern because they are probably the most vulnerable of all sections of society.  It seems from the results that complex, expensive interventions by governments aren't delivering the basic income required by the old to maintain a minimum standard of living.  The OECD's report card should raise some serious questions for most of the 30 countries involved.  Despite the data issues we have mentioned, an average poverty level amongst all "of retirement age" of 13% across all countries should be unacceptable.  (File size - on-line; 309 pp)  290