Finding findings about International Economics & Finance

To content | To menu | To search

Sun March 4 2012

Influence of social expenditures on the economic competitiveness

An article from Pr. Krugman's blog leads me to an idea on the links between the welfare state and the economy's competitiveness. My assumption is, the more there is social redistribution the more there is an effect on exports. I'm thinking about two opposite main effects;


  • Domestic markets partially fuelled by public transfers can absorb a part of the development costs of new products and allowing firms to offer relatively lower prices to exports.
  • Firms turn mainly toward their domestic market because of higher average incomes (perhaps a price distortion effect) and mobilize fewer efforts to export (a kind of "export gap").


So to explore this hunch I've used the same OECD Factbook database than Pr. Krugman and tried to see what data can tell to me.




The two variables are expressed as a share of GDP in 30 OECD's countries in 2007, last available year (and before the economic financial and debt crises may have modified redistribution pattern and exports' demand).

Here are some points I've noticed:


  1. There are 12 points above the linear regression curve and four of them are very close to it.


  1. There is no country with less than 16% of Public Social Expenditures (PSE) which generate exports bigger than 50% of GDP.


  1. Most of countries of this panel generate exports smaller than 40% of their GDP and spend between 15% to 25% of GDP in PSE (OECD's countries average is 19.5% ([2]).


  2. Among the five bigger amounts of PSE spent, only one country (France) has exports lesser than 50% GDP. There are just 8 countries with more than 50% of country's GDP in exports; half of them have more than 25% of PSE.



I conclude to the presence of threshold effects in the PSE/Exports dynamic.

It seems that a country needs at least 15% of PSE to have good exports' performances(more than 40% of GDP, OECD weighted average exports share of GDP is 26.7% in 2007) [1]. An upper limit of PSE is not so evident to define as the dispersed top 5 exporter's countries show it to us.

Of course there are more complex effects, for example, firms benefitting (directly or via individual customers) of an important share of PSE represent a creditworthy market (backed by public expenditures) for other firms that can sell them goods at a slightly more expensive price.


To better understand the PSE/X interactions, I've ranked OECD's countries according to the gap between GDP's s share of exports and the one of social expenditures (X-PSE). And then I colored in red countries with a current account deficit.


You can note the pretty clean distribution of the countries' panel in 2 groups. 3 exceptions remain which are Hungary, Canada and Japan. Each of them may have specificities which explain the particular result of the economy. For example Japan is an export oriented economy (confirmed by the trade surplus) but with very big issues on wages and prices stagnation, housing's prices still dissociated from the economic trend (in terms of households income which leads to more public redistribution incomes) and a pretty generous retirement system.


I can however conclude that countries with a small gap between exports and social expenses are less likely to generate a trade surplus.


Anyway, threshold's borders are not sharp. Trade performances seem to be mixed around a gap of 20 point between exports and PSE (Poland +20.97 and Finland +21.1). I was aware that the gap would not explain all trade performances. The encountered difficulties with various results in the threshold area are making obvious the fact that complex interactions influence the net trade of an economy. For example, USA and Ireland spend the same GDP's share of public social expenditures [3] and each of these countries is the extreme point of the graph, even in the cloud points' chart. Obviously this is just a first step which need further analysis but from my point of view, there are too many and too clean correlations to let the coincidence explain these results.


I will explore other potential impact of PSE in OECD's countries in further posts.

Please feel free to comment.



[1]Except for South Korea and Chile. OECD's countries not weighted average is 41.3% of GDP.

[2]OECD's weighted average is 19.2% of GDP.

[3] USA 16.2%, Ireland 16.3%

Thu February 16 2012

Is cutting Greek minimum wage economically relevant?

We would like to react at the agreement given by Greek government & assembly to cut private minimum wage by 22%. It seems an obligation to receive next financial aid from Troika (IMF, ECB and European Union). According with the Greek labor minister, "it's very big" (see previous link to Bloomberg) but it is in return for €130 billion of public aid and €100 billion of debt cut from banks creditors. Now we know the deal, which troikahas not fully approve yet, despite the strong commitment of the Greek side (there was an official vote in the assembly, some deputies lost their seat because they were against this proposal).


Given the terms of the deal, what is the economic purpose of a minimum wage cut? Official documents are missing but it appears as a consensus in officials' statements that to cut minimum wage will improve competitiveness and growth. We are going to check which aspects of Greek economy have justified that kind of "very big" change.




Let's check the current and trade accounts of Greece. We have added some others European countries to make comparison, Portugal would have been relevant too.



Obviously it's hard for the Greek economy to balance exports and imports1. In this way restoring (or developing, because the sold was never positive) price competitiveness of exports by lowering wages may look like a nice –and quick - response. But if we watch besides near the components of the current account, we found a more shaded story.


Both services and goods exports are growing, since end of 90's. In particular, since the entry in the euro (2001), goods exports increased of 48% until 2010 (and even +71,7% to the high point of 2008), services respectively +34,7% and +54,3% (2001/2008). These are nice yearly average growths! Here we have a real "Eurozone" effect; exports are increasing in consequence of the entry in a common monetary zone.

We also note that services represent 62,5% of total Greek total exports, which is not a frequent profile in Europe, except for tiny countries like Luxembourg. Travel services are a proxy of tourism performance and we see that the entry is not so good for the industry, one can easily think the contrary, and perhaps there is a price effect when swapping from drachm to euro.

Although it is always useful to stimulate exports competitiveness, it seems that Greece already achieved a nice performance since its entry in the euro. Yearly average growth rate of exports (goods and services confounded) is of +6,2% until 2008 (before the financial crisis that reduced world trade). One can conclude that imports increased quicker than the exports during the last 10 years and not that exports have made a bad result. It changes nothing about the current account (the deficit is growing faster than GDP) but that puts in perspective the urgent need to lower the minimum wage of 22% to encourage exports.


As we can see it, the imports increase very heavily in GDP terms, but the trend is not linear. Indeed we note that a peak is reached just before the entry in euro and followed by a decrease for 3 years (in fact it is explained by a stable level of imports and a strong economic growth

). It may be interesting to look more closely what's happened in these years to see what can be transposed in the present situation.

To conclude with the competitiveness, a good performance indicator is the market share of exports.



Greece have a steady share of 0,4% of world trade. This stability is a good performance by itself. More open economies have seen reducing their world market share on the same period. Il appears that improving Greek competitiveness will not be so simple that cutting exports prices. This economy achieved honorable performances in this domain during the last 10 years. Greece is selling abroad -as shows the success to maintain its world market share- but trade deficits remains. The question of raising exports volume would be less in the price than in Greece's capacity to produce a larger range of goods and services for exports2.

Therefore the problem rests less on a question of value than on a question of volume. In this way acting against export prices through wages is a wrong policy choice.

But European Union know that well, at least since may 2010, when the report "The economic adjustment programme for Greece" clearly affirm and explain why a cut on minimum wage will have no effect on export competitiveness (p21).



On reason for the wage cut is that Greek labor cost have increased faster than German one, which explain both competitiveness and growth issues of this economy. That's true in nominal terms, but if you look in real terms then you'll see a very different trend (nominal cost deflated by CPI).



What do we learn from these 2 charts?

First, there is a good reason why Greek's labor cost is increasing fast; average inflation in this country Is bigger than those in Germany. So to maintain purchasing power (not succeeded in real terms), labor cost has to follow prices. We have here a traditional problem of non-optimal monetary zone; asymmetric inflation. Of courses if euro monetary authorities didn't find a solution to this problem in the ten previous years, we can't expect more than "cutting wage is the unique solution to the fast relative growth of labor cost". It seems to us that it deserves a bigger monetary policy brain storming than that, at least because it could happen again in this country or another in the monetary zone.

Second, real labor cost decreases(also in Germany) then it should be difficult to maintain domestic consumption with this trend combined to the minimum wage cut. Yes it's a Demand based approach but ceteris paribus inside the economy, if labor cost and prices of goods and services have close trends there is no constraint on Offer, problem is on Offer addressed to the rest of the world. For the moment it seems hard to imagine that a 22% cut on wages leads to a similar cut on domestic prices, at least because it is very likely that a big part of domestic consumption is linked to imports. So cutting minimum wage shouldn't make a positive choc on domestic offer, nor help with the unemployment issue3.


Finally the first thing one should ever do before talking about heavy (or "very big") economic measures is to take time to understand what is the economic model of the target economy. If we do so with Greece, then we can see a very obvious fact, this is not and export based economy.


In Greece, households consumption represents by far the biggest GDP's share of the above panel (76,2% in 2010, according to Eurostat), and even of all European Union. Following countries are Cyprus (71,6%), Malta (69,7%), Latvia (62,3%) and UK (61,2%), the closest country (in all senses) is Turkey (74,2%). So Greek's households consumption represent two third of country's GDP and government and Troika are cutting the minimum wage? How economic growth may be possible at short term?

To solve the debt problem by economic growth and public solvability, what Greece need the most for now is a clever macroeconomic policy which plans and defines countries' development for several years and takes into account all aspects and specificities of a not-really-ready-for-globalization modern economy. Solutions may be a little more complex (and we hope efficient) than a downsizing at a country scale.


You can also find another nice analysis about the unfounded obsession of cutting Greek minimum wage on Rebecca Wilder's blog.





1 Being about the current account, there are some incomes flows, creating an increasing but still small gap these last years with the commercial balance.

2 Exports expressed in percentage of the GDP are indeed among the lowest of the euro zone, around 20% (source eurostat).

3 Again you can refer to the same occasional paper of European Commission stated above for more detailed reasons of wage cut's inefficiency in Greece.

Wed February 1 2012

20 years of wages in globalized economies

At first, globalization was a labor cost's story. Developed countries were looking for cheap labor and developing countries for rich customers/investors (in the foreign direct investment way). We just take a look back on the last 20 years to see what happened in terms of wages and purchasing power. Three data series have been used; yearly average growth of wages (W), manufacturing wages(mW) and CPI, in two periods 1990/2010 and 2000/2010 [1]. We used a sub-period because of numerous changes that occured at the millennium turn, such as China's entry in WTO (December 2002), the crash of internet bubble and the recovery of Asian developing countries, Mexico and Russian federation, which had a tough decade in the 90's!

A bit of theory

Well it's more hypothesis than theory, we can expect at least the following hierarchies:

W>mW>CPI (1)

mW>W>CPI (2)

We can easily make the assumption that first relationship is manly observed in developed countries while the second one occurs in developing countries. It could be that way because the most exposed activity to "globalization's wind of change" is manufacturing. Growth rates spread between average and manufacturing wages is either positive or negative, depending on direction of the manufacturing labor Demand flow (from rich to poor countries).


Chart 1 Average yearly growth rates in %. (Sources: IFS, Laborsta (ILO), BEA for US wages and National Bureau of China)


What did we found?

We will focus only on growth rates comparison (and we're leaving wages and CPI levels a side for now).


There are 3 groups of countries:

CPI> W or mW:

  • CPI in Mexico grows faster than the average of wages. The tequila crisis was a big shock, especially for services industries.
  • Less predictable, US manufacturing wage growth rate is lower than CPI growth over the 2 decades. Slightly but still, it's a surprise. Manufacturing wage bargaining was thought much efficient in this country, given the strong labor unions. At least the average wage growth is bigger than those of the CPI.
  • Italy was in the same case than USA, with a bigger negative spread between CPI and manufacturing wages growth rates.


  • Only 3 countries in this group, Germany, Hong Kong and China (USA and Italy have a close pattern but underperform CPI growth). Economies of these three countries are still export-oriented. The long run trend of low manufacturing wages growth (compared to the national average rate) give us a hint on how to succeed in this kind of economic specialization (and explain why all countries can't be them).


  • Majority of countries showed in the chart have experienced a bigger growth rate of manufacturing wages than the average wage (and than CPI). Unfortunately we didn't found manufacturing wages data for Eurozone, Greece (prior 2000) and Russian Federation.

Except for Mexico, average wage growth compared to CPI growth didn't show any evidence of purchasing power losses

Chart 2 Average yearly growth rates in % (Sources: IFS, Laborsta (ILO), BEA for US wages and National Bureau of China)



Few changes can be found in the most recent period.

  • US mW >CPI (perhaps most of the industrial transition phase is over and only competitive manufacturing activities remain and offer better wages) and still mW<W, so it should lead to an export prices' competitiveness advantage (as it is the case with China and Germany).
  • Japan is deeply involved in a deflation process. This lead to negative growth of average wages but manufacturing wages seem to resist.
  • Hong Kong made a big change in wages' trends. Inflation control has reduced yearly growth of manufacturing wages a little bit under those of CPI (+0.04% mW, +0.08% CPI, per year).
  • Italy reverses its wages growth hierarchy and now mW>W.

The global trend of this decade appears to be wages and CPI moderation, yet no big changes are to be found in wages patterns and purchasing power of employees.



We didn't found losses of purchasing power , except in Mexico. But there are clearly two groups of economies;

  • Countries where mW>W. Most of the time these countries have a deficit in net current account (France, Greece, Italy [2], Spain, UK and Mexico).
  • Countries where W>mW. These economies run chronic net trade surplus (goods), Germany, China, and Hong Kong (even with the change during the last 10 years).
  • USA are a whole and unique case; a feeble growth of manufacturing wages (compare to average wages and even to CPI for the entire period of 20 years) but huge current account deficits.

We just touched the surface of the lake and we provided more descriptions than explanations. But the purpose of this kind of article is to throw a stone in this lake and see what kind of circles it generates. In another word this is an ideas box, perhaps someone (perhaps us?) will use the indicators to make a deeper study. At least we hope it will be followed by some enriching exchanges of views on these subjects.

We have wages data for several others countries, we will show the charts in another post, and comment the growth levels.

[1] Manufacturing wages are only available up to 2008, so we don't capture any "subprime and to be continued crisis" effect.
[2] Italy had a net current account surplus until 2000 and has only deficits since then.

- page 1 of 2