US Public Spending In Context--Part 2

by: Paul Rosenberg

Sat Apr 11, 2009 at 20:45

This is Part Two of a response to Chris's diary, "Long-term, Center-left Victory on Public Spending Highly Likely".  In it, Chris argued that there was very good reason to expect a permanent expansion of public spending:

As a country, we are on the brink of a substantial, long-term increase in social investment that will move our economy much closer to the mixed, and substantially larger public welfare models, of Canada and Western Europe.

It's my position that the arguments Chris marshaled are significant, and deserving of attention, but that they don't really show such a victory, but only the potential for it.

In Part 1, I laid out four reasons for taking this position, and argued that turning potential into reality would depend, in part, on getting a clearer picture of how spending had varied in the past, how it was projected to change, and how the US compared with other countries.  I dealt with the US record in Part 1.  Now it's time to compare us to other countries.

Paul Rosenberg :: US Public Spending In Context--Part 2
If we want to compare US spending to other welfare states, the first and third charts below are helpful as a starting place. They are from "The Struggle between Equity and Efficiency: Evidence from the Luxembourg Income Studies" (PDF), Luxembourg Income Study, Working Paper No. 498, by Norbert Berthold and Alexander B. Brunner, which I referred to in both my previous diaries on the subject today.  

The charts use a variant of the "The Worlds of Welfare Capitalism" model, which splits the conservative welfare state group into "Continental" and "Southern" groups (also known as "conservative" or "coporatist" and "Mediterranean" or "familial" respectively).  The countries included are:

Nordic: Sweden, Finland, Denmark, Norway
Anglo-Saxon: Great Britain, Ireland
Continental: Germany, France, Austria, Benelux-Countries, Switzerland
South: Italy, Spain

Please note that "social expenditure" does not include eduction, while it was included in the welfare state core in my earlier diary about spending in the US.  The US comparison chart below has been adjusted to reflect this difference.

Note that all four "worlds" move roughly in parallel, with expenditures peaking in the early 1990s, and declining until 2000, after which we see a mild rebound. A similar pattern is visible with the US (top line in the chart below), but at a much lower level, peaking at around 7 points lower than the Anglo-Saxon peak, and with a much stronger rebound, thanks to the spike we're now passing through due to the stimulus and the bank bailout.  Afterwards, however, it drops back down again, several points below the Anglo-Saxon average.

Turning to social expenditures, the early-90s peaking phenomena is much less sharp and/or distinct.  It's very mild in the US, and more prolonged, lasting into the second half of the decade--similar to the pattern for the continental (conservative) welfare states--while the southern states rebounded much more quickly, already recovering in 1996, when the US and the continental states were still peaking.  The Nordic and Anglo-Saxon states were the only ones whose profile was similar between overall spending and social expenditures.  The US shows only a modest 1-percent post-2008 rise, less pronounced than the mid-Bush to mid-Clinton rise 1989 to 1996m which measured 2.5 percent.  It brings us to just 16% of GDP, a couple of points below the Anglo-Saxon average, and well below any of the other "worlds".

The above comparisons are helpful because they extend over time, but since they draw from different data sources, there is no guarantee they are truly equivalent comparisons.  Indeed, most static comparisons I've seen show much bigger gaps, with the US as a much more distinct outlier.  The trends above suggest part of why this may be so--the US has gradually increased socil spending since the early 80s, while others have declined, or at least been more erratic.  So it's certainly plausible that we're less of an outlier than in the past.  Still, the closeness seems overstated.  We pass on now to other data, that while static shows us to be more out of step.

First, from Working Paper No. 419, Poor People in Rich Nations: The United States in Comparative Perspective (PDF) by Timothy Smeeding (October 2005), we have two charts showing the US as an outlier in poverty rates, with two different corralations.  The first shows the relationships between non-elderly poverty and wages.  The second shows the relationship between cash social expenditures and non-elderly poverty.  In short, we have more poverty because we have more low-wage jobs and because we provide less cash to compensate for our low-wage job market.

Next we look at some more detailed comparative data about the impact of social spending on poverty rates for different groups, from LIS Working Paper No. 426, "Government Programs and Social Outcomes: The United States in Comparative Perspective" (PDF), also by Timothy Smeeding in April, 2005.

We begin with a chart showing 1980s and '90s cash & near-cash social spending on the non-elderly poor for 6 sets of 17 nations.  The note from this chart explains:

* Total Nonelderly Social Expenditures (as percentage of GDP), including all cash plus near cash spending (e.g., food stamps) and public housing but excluding health care and education spending. OECD (2002b). Anglos include Australia, UK, Canada; Scandinavia includes Finland, Norway, Sweden; Northern Europe includes Belgium, Denmark, Netherlands; Central/Sourthern Europe includes Austria, France, Germany, Italy, Luxembourg, Spain.

This is a bit of a hybrid categorization, as the "Northern Europe" group contains one social democratic state (Denmark) one conservative state (Belgium) and one in-betweener (Netherlands), but the other three groups (Anglo, Scandanavian, South/Central Europe) are all pure types, (liberal, social democratic, and conservative, respectively).  Here's the chart:

The figures here--focused on the non-elderly poor, not the middle class or elderly poor--show the US as much more of an outlier, closer to Mexico throughout the 1990s than it is to any of the other advanced industrial nations, even the others classified as "liberal" welfare states.  The liberal and conservative groups are very closely correlated, with roughly twice the level of spending than the US.  The mixed Northern Europe and pure social democratic groups are not closely correlated, but fall in roughly the same range, a bit under twice that of the liberal and conservative groups.

Now let's look at some data tables that provide a closer view of what this means.  First this snapshot macro-economic overview:

Our spending (2.8% of GDP) is less than half of Canada (6.0%) or the UK (6.4%), the next nearest countries, and both examples of other liberal welfare states.

Next, this look at spending on all the poor:

"Social insurance" refers to universalist benefits, such as Social Security, while "social assistance" refers to targeted means-tested programs.  The note to this table says:

4 Refunds from the Earned Income Tax Credit (US) and the Family Tax Credit (UK) are treated as social assistance, as are near-cash food and housing benefits such as food stamps and housing allowances.

We start off with a poverty rate from the market that's a bit on the low side--only Finland and the Netherlands are lower. But we lag far behind in poverty reduction from both social insurance and social assistance. Our percent reduction from social insurance (18.6%) is just 40% of the average (46.9%). We do somewhat better overall, which only goes to show that we depend disproportionately on programs that target the poor--and there's an old saying that programs for poor people are poor programs.  Overall, our 28.3% poverty reduction rate is 45.8% of the average, which is 61.8%.  Still, no other country comes anywhere close to our poor performance. The next worst is Canada, with a 52% poverty reduction rate.

Next, non-elderly poor families with children:

As can be seen, among single-parent families, social insurance does very little for the non-elderly poor in America.  It barely has any impact, reducing poverty less than 1%, whereas it reduces poverty almost 60% in Belgium.  Our overall poverty reduction rate for these families is just 14.8%, barely more than half of the next-worse country, Canada at 27.0%, which itself is just over half of the average, 52.6%.  This is what demonization looks like.

Or does it?  Here's the real shocker--as bad as we do with single-parent poor families, we do even worse with two-parent families. In fact, social insurance plus taxs actually increases poverty!  The overall poverty reduction rate of 5.8% is just 1/8th of the average 47.8% reduction rate.  No other country even comes close to being so bad. The Netherlands, with a reduction rate of 20.2% is 3 1/2 times more effective than we are.

Finally, the elderly poor:

At last, we're not really an outlier!  Well, at least when it comes to social insurance.  Social Security and Medicare combined give us a poverty reduction rate of 56.9%, better than Britain's 50.3%  No wonder is Versailles conventional wisdom that these two programs need to be cut!

But, when you go to the bottom line--overall reduction, we're back in last place, 57.7% reduction compared to Britain's 68%--we do just 85% as well as the next worse country.  Once again, we are dead last, and it isn't even close.  And that's the best we do.


What all the above shows is that the closer you look, the farther we are from drawing close to other countries.  We may be drawing near to the back of the pack when it comes to overall government spending, but the more you look for spending that would make us a nation where no one is left behind, the more of an outlier we are--and look to remain.

Tags: , , (All Tags)
Print Friendly View Send As Email

a lot to digest here (4.00 / 1)
Thanks so much.  The U.S. is certainly an outlier.

New Jersey politics at Blue Jersey.

GINI (0.00 / 0)
It would be nice to see this correlated with GINI figures.

I did something like this with wages over time and across countries. The correlation (by eye) is pretty high. The countries with have the highest levels of income equality also have the best wage levels.

One could say that the wealthy keep the working man down - sort of sounds like Marx...

(Look especially at the last chart.)

Policies not Politics

Labor Relations (4.00 / 1)
One of the key parts of a functional social democracy is its labor relations regime.  And how the labor relations regime works has a big impact on union density and sectoral union density.  This in turn has a big impact on the rest of this all, from income inequality to wage rates, etc.  

So it might be interesting to see an inclusion of the labor relations style of the European social democracies in context.

Esping-Andersen Does Discuss This (0.00 / 0)
in "The Three Worlds of Welfare Capitalism", but more from an historical point of view than a systematically analytical one.  It was particularly important, for example, that urban-centered unions in Scandanavia were able to set the agenda, so that they articulated a framework that proved appealing both to rural workers and, over time, to urban professionals as well.

"You know what they say -- those of us who fail history... doomed to repeat it in summer school." -- Buffy The Vampire Slayer, Season 6, Episode 3

[ Parent ]
Exactly (0.00 / 0)
These dynamics are key.  France's labor relations system comes to mind as well; same with (West) Germany's.  

I haven't read the Esping-Anderson piece yet, but I am going to after working up some ANES data tonight.  

Social democracies are not agnostic (much less antagonistic) toward unions or employee-representation; they encourage it.  Very different world we live in here in the U.S.  

[ Parent ]

Open Left Campaigns



Advanced Search

Powered by: SoapBlox