Friday, 22 November 2013

The other side of trade

Following on from my discussion of trade surpluses and their effects (which lots of people didn't get, sadly), I thought I would talk about the other side of trade - which is capital investment.

This speech by the ECB's Asmussen contained the following intriguing line:

"If more German savings were invested at home, the current account surplus would fall by definition".

Now I can hear you all scratching your heads. What on earth has the behaviour of German savers got to do with the current account surplus?

Nothing, actually. This isn't about German savers. It is about German national savings - the proportion of national income that is not spent or invested at home. Essentially, what Asmussen is saying is that the German national savings rate is too high.

A nation's current account reflects its net foreign income, and its capital account reflects the change in ownership of foreign assets. A nation that is running a trade surplus has an equivalent capital deficit, by definition: money leaves the country and is invested in foreign assets, rather than domestically. So poor domestic investment is a consequence of an export-dominant economic model. That's what Asmussen is talking about. The trouble is, he thinks that by increasing capital investment at home, which would reduce the capital deficit, the trade surplus would automatically fall. But I'm afraid he has this the wrong way round.

In a closed economy,

Y = C + G + I

where Y = national income, C = domestic consumption, G = government spending and I = investment. Rearranging this, we see that

I = Y - G - C

In other words, investment is the bit of income left over after all private and public spending commitments have been met. Strictly speaking, saving (S) is the bit of income left over after all private and public spending commitments have been met, but we assume that all saving is productively invested in some way:

S = I = Y - G - C.

So far so good. But Germany is not a closed economy. In an open economy, the foreign sector matters. Net exports are the difference between exports and imports:

NX = X - M

Germany has a trade surplus, so NX is positive. A positive current account balance is a contribution to national income:

Y = NX + (C + G + I)

where C + G + I = domestic demand.

We can immediately see where the desire to run a trade surplus comes from - Y may be higher if there is a trade surplus. But if Y is constant (or delta Y < delta NX), then a trade surplus crowds out domestic demand. Domestic demand is public and private spending AND INVESTMENT. Therefore, if Y, G & C are all constant, a large trade surplus crowds out domestic investment. So what Asmussen is saying is that Germany should should seek to increase I (domestic investment), which would force NX to fall.

But I'm afraid this is simply impossible. The capital deficit is a consequence of the trade surplus, not a cause, and I is a residual:

I = Y - C - G - NX

Put bluntly, the money has to come from somewhere. In China, of course, there is perhaps excessive domestic investment - but they don't have much in the way of a welfare state and they have significant financial repression, so both C and G are low and most of the national income goes into I. I suspect that attempts to increase I in Germany would be achieved by the same means - cut government spending and depress household incomes - in which case increasing I would have no effect at all on NX. So that is not what is really needed. What Germany really needs to do is to reduce NX in order to increase I, not the other way round. This could be done by cutting exports, but it would be far better if it were done by increasing imports, And actually that would be best achieved by increasing both C and G, so looser fiscal policy, really: despite my caveat in an earlier post, higher wages and lower taxes would probably result in increased imports to some degree. That would diminish net capital flows out of the country and therefore increase the amount of capital available for domestic investment.

Asmussen should have made this clear. Increasing domestic investment would be consequent upon a falling trade surplus, not the other way round. By getting the causation the wrong way round, he has missed a golden opportunity to influence German economic policy for the better.


  1. This comment has been removed by a blog administrator.

  2. Sorry Dinero, can you repost please? I hit the wrong button and accidentally deleted your comment. Apparently there is no way of recovering it.

  3. OK no problem


    I was saying that

    The Euro area is a special case. , the trade surplus country does not need to lend euros to the deficit country to fund a trade deficit as the ECB will lend the deficit country the money in the operation of the payments system.

    1. Hi Dinero,

      No, that isn't true. The ECB provides liquidity for payments, but it doesn't provide trade finance. That comes from banks and is backstopped by governments. Prior to the Eurozone crisis the German banks were financing the trade deficits in periphery countries. When they abruptly pulled their funds, governments were forced to take over financing of those deficits. That's why sovereign debt went up so much.

    2. Dinero is more right than you, Frances. To say that governments (Troika) had to take over the trade financing of, say, Greece is not really correct. The borrower of Troika-loans is the Hellenic Republic through its Treasury and the Hellenic Republic is not engaged in trade financing (except when it buys, say, armaments from Germany). All other cross-border financial flows are settled by the Greek banking system through the ECBs target2 cash management system. When more money goes out than comes in, the hole was financed by private banks before the crisis and target2 balances were around zero then. When the private banks retreated beginning in 2009, the hole was automatically filled via target2 liabilities.

      Target2 claims are substantially higher than the sum of current account deficits because they also include capital flight. In fact, capital flight represents by far the larger portion of target2 claims against the South.

      One point which deserves attention but which is seldom mentioned is that, through target2, Germany actually reduced its macro country exposure to, say, Greece. Before, a loan to Greece by Deutsche was 100% Greek exposure of Germany whereas Germany carries only about 27% risk of the target2 claims. Thus, one could argue that Germany reduced its Greek exposure from 100% to 27% by shifting the claims from its banking sector to target2. Obviously, that’s not quite correct because Germany also assumed 27% of the target2 claims of other countries.

    3. Kleingut, you need to read the rest of this discussion. I've covered most of what you said already, including your observation about capital flight. But Dinero is not correct. He is ignoring the private sector side of trade finance. And you should also remember that Target2 claims are backstopped by national governments - hence all the discussion about what happens if a country leaves the Euro.

      You have to be very cautious when claiming that Germany is "financing" other countries via the Eurosystem. It could equally be argued that Greece is financing Germany via distribution of interest receipts on Greek bonds held by the Eurosystem, since Germany as the largest contributor to the Eurosystem receives the largest share of that interest.

    4. Frances, Germany is the largest net financier of the Eurozone, in fact, of the whole world since, I believe, Germany now has the world's largest current account surplus. By definition, current account surplusers MUST be net capital exporters. THAT (i. e. the financial risks associated with current account surpluses) is what needs to be explained to the Germans in order to make them do something about their surpluses. Whatever Germany's current account surpluses do to the rest of the world, Germans needs to understand that they are a tremendous financial risk to themselves. Thus, it is in their own interest to do something about it.

      Telling Germans to export less is silly. Telling them to import more would be sensible but won't carry the day. Telling them that their pensions might be in danger will get their attention immediately. Mind you, between mercantilists and Keynesian countries, the latter always have the greater leverage: yes, the mercantilists have the jobs and the cash, but also the loans. The Keynesians enjoy their consumption and they can blackmail the mercantilists.

      Germany now has gross foreign assets in excess of one year's GDP and they rise with every day of current account surpluses. Germans may think that they are safely invested but they can't be because money, directly or indirectly, finds its way from those who have it to those who need it. Since 2007, Germany has lost a 3-digit BEUR figure of foreign assets. However, Germans are not really aware of that because the losses have only hit tax payers in a very limited way. They were shielded in upper layers like banks and insurance companies. German tax payers need to understand that sooner or later the losses will hit them directly (i. e. when a Greek haircut needs to run through the budget) and that they are well advised to reduce their risk.

      Finally, I think it is incomplete to focus exclusively on the trade side of the current account and leave the services side unmentioned. The fastest way for Germans to reduce their surplus AND to help the South would be to multiply their vacations to the South. Obviously, the trade deficit is too high to be neutralized by more vacations...

      Given that Germany will have a current account surplus for quite some time, my advice would be that Germany export its capital more via investment than loans. Particularly investments in the South so that a country like Greece can build up productive capacities to make some of the products which Germans would buy. As of today, regardless of how much Germany would stimulate domestic demand, it would help Greece very little because Greece has only very little supplies to deliver.

    5. Kleingut,

      Great comment, and I don't disagree. I think you are unnecessarily critical of my post though:

      - Your comment that Germany is a capital exporter because of its trade surplus is the same as my observation that Germany has a capital account deficit because of its trade surplus, just expressed differently.

      - I have not suggested Germany reduces exports. I did suggest importing more would help, but in the first post in this series I warned that if German goods are regarded as better quality, it would be irrational of Germans to buy imported goods.

      - I also warned, in the second post in this series, that a persistent trade surplus (i.e. capital deficit) leaves the country very vulnerable to exogenous shocks. That would impact pensions, savings and jobs - all the things that this trade surplus policy (for it IS a policy) is designed to protect.

      - I have also been warning for literally years about the disastrous state of the German banking system and its over-exposure to distressed countries' debt. It's in better shape than it was, but at some point it will require substantial recapitalisation and I think many of the Landesbanken may have to be wound up.

      - Trade, in its entirety, includes services. I do not equate "trade" with "trade in goods". However, Germany is a manufacturing-dominant economy and its trade surplus originates entirely from its trade in goods (since its service sector is actually running a small trade deficit). Therefore I did not specifically mention services. I agree that the way forward for periphery countries is for them to develop services, rather than attempt to compete with Germany's manufacturing might. However, it still requires Germany to liberalise wage and tax policies, so that Germans can afford to take more holidays in Greece. Expecting Greece to reduce its population to penury so Germans can afford the extra holidays without increasing wages or cutting taxes is frankly unreasonable. And all the investment in the world won't help Greece out of its slump unless there is demand for its products.

    6. Frances, this article makes my point in an excellent way!

  4. This comment has been removed by the author.

  5. The ECB does not know that a Request for reserves to fund the draw down of a loan from a Greek bank has been generated because the Greek bank's customer is borrowing to buy a a German product.

    1. It is still a "request for reserves" for the bank, not trade financing. There is a very important difference.

  6. I am interested in this subject, I'm sure you could explain the difference if there is one.

    As Far as I can see commerce in the Euro area is the same as commerce in a nation with a national currency and a Central bank clearing system of reserves. Once the CB has an interest rate target the reserves are issued or else the interest rate goes up. There is no trade financing as far as the system is concerned, its all in the same system.

    22 November 2013 18:29

    1. 1) Reserves are commercial bank assets, not liabilities. All they do is facilitate payments. They are not "borrowings" of banks from central banks.

      2) Trade financing comes from banks, not from the central banks. It is a form of bank lending, which is funded from a variety of sources that may include the central bank. But that does not mean that central banks fund trade. They do not.

      Please don't confuse the operation of a real-time gross settlement system, which uses reserves as liquidity, with trade financing. They are very different things.

  7. Reserves are borrowed from central banks when they can not find cheaper sources of reserves from other banks

    There is no "trade financing" in the euro area

    in the absence of a definition of the "different thing" that you refer to the conclusion is that there is no conventional financial limits on trade deficits in the euro area.

    1. The conventional conceptual factors of a trade deficit don't apply within the Euro zone.

    2. Dinero,

      No, this is a gross misunderstanding of the Euro area's Target2 system.

      Reserves are NOT trade financing. How many times do I have to say it? They sit on the ASSET side of bank balance sheets. They are, if you like, what a bank has in its current account at the central banks. They have nothing to do with the financing of trade.

      Trade finance is simply a type of bank lending. And bank lending is constrained by capital requirements and risk appetite, not by the availability of reserves. Trade deficits are therefore constrained by the health and the risk appetite of commercial banks, not by the willingness of the central bank to provide reserves to settle payments ex post.

  8. Have you seen the target 2 balances they are all in credit to the surplus countries and deficit to the the deficit countries.

    That is regardless of Germany's loans habits because the ECB is obliged to provide the deficit countries with euros or else they would run out of currency and have to leave the euro.

    1. Yes, of course they are, Dinero. That is how the reserve system works. But most of the Target2 claims now are due to capital repatriation to Germany, not trade flows.

      The operation of the Target2 system has nothing whatsoever to do with the provision of currency to the countries of the Euro. Look, with all due respect you don't remotely understand this, and the comments on my blog are not the place to explain it. Can I suggest you read Karl Whelan's work on how Target2 works, and also this post by Beate Reszat:

      plus the links in her post. I'm not going to discuss this further here.

  9. Maybe he meant - if more investments in Germany grow the domestic GDP the relative export in percent may drop but not the absolute number.

  10. I like it to call it the traffic light effect,if a council brings in traffic calming measures then the small feeder road stay relatively quite,while main thoroughfares are slowed right down even to a stop,yet all the stopping & starting & uneconomic driving conditions means more money is used on fuel,parts which takes money out of the local economy ,just £1 per week in a small-town with a 100k of cars takes 100k out of the local economy,which reduces other enterprise & savings
    So it a council stopped cars entering it's borough & forced them onto buses,this would also effect the balance in the wrong way,with not enough money leaving the borough for it to service the petrol stations to make it viable to service the cars left,each needs to be as close to being balanced has possible for a sound energetic robust economy has possible

  11. There is no need to discuss Target 2 its not essential to the discussion about trade and money flows in the Eurozone.

    If I borrow 10 000 euro from a bank to buy a car from a car dealership in Germany the bank lends me the money the dealership does not need to lend me the money. As you have explained yourself loans create deposits.

    If I am in Greece or Germany I get the loan either way - The bank teller is not going to say to me no you can't have any euros because you are Greek and these euro here are for German people only.
    Its the single currency so it works as a single currency for all.

    As you said on the post entitled "About that ECB rate cut" there is one interest rate for all members.

    Regarding The intriguing comment from Asmussen If German banks did less direct lending to Greek banks there would be some effect on rates in Greece but I expect from what I have read on your Blog and elsewhere the ECB would want to counteract higher rates in Greece anyway.

    1. Dinero, if you read my blog then you will know that the ECB is not capable of counteracting higher rates in Greece. And I am afraid you are wrong about bank lending. Banks DO ask what the purpose of the loan is before they lend, and if the risk is too high they either charge much higher interest or they won't lend at all. So a loan to fund a car purchase in Greece really is more expensive than a loan to fund a car purchase in Germany, even if the car purchased is German-manufactured in both cases. Another way of looking at it is that the real price of euros in Greece is higher than the real price in Germany. Having a single currency does not mean that the currency has the same value everywhere.

    2. Sorry, I mean the real price of euros is higher in Germany, of course.

  12. I don't understand how you come to causality from a fully commutative accounting identity which is not directional.

    If we try to go back to the real world, I think he has a point. Imagine you stumble on a big pot of pounds sterling and decide to use it to start a music school (100 teachers, new buildings for the classrooms, new instruments) and at some point all is set in stone except where you put the school. You hesitate between Bologna and Birmingham. In either cases you plan to employ local teachers, local bricklayers, and source instruments from local suppliers. If you decide for Bologna, you'll give lots of people around Bologna pounds. What can they do with pounds? Buy UK stuff, or swap for euros with people who want to buy UK stuff. Basically your choosing Bologna for your school makes pounds easier to get outside the UK, and harder to get inside the UK, esp. compared to the alternative choice of building the school in Birmingham and giving pounds to music teachers and bricklayers there. This is very likely to show up in the trade balance, unless teachers and bricklayers save all they earn from you, in sterling cash or gilts, which seems improbable.

    (My example uses trade between currency areas because it makes it slightly simpler to understand, but the effect should be essentially the same within a currency area, or between people in a single village, the medium of exchange is not material to the argument and works under a barter regime too.)

    Maybe your idea on causality is rooted in a belief in all-powerful governments. The economy is the sum of the actions of people in it. Government are large actors so they can influence some, but then large modern government machines have a lot of inertia and in practice people in power can only tweak things at the margin. Belgium did fine for 2 years on auto-pilot. Active policy is actually hard to distinguish from auto-pilot in modern states because the leeway is so small.

    Trade deficits or surpluses are not what they are because the finance minister turns a big magic dial in the basement of the ministry that would allow them to set the value to whatever they like. There's no such dial. It's ironical that Asmussen is seemingly ending up with a more humanist view (people matter!) of the economy than you do.

    Imagine you wake up tomorrow morning in the shoes of the German chancellor, and that to be helpful the office as been merged overnight with that of finance minister. In the morning, you manage to convince your parliament that you should reduce the trade surplus. What practical measure do you take in the afternoon? Where is the magic dial? The best you can do is probably go on TV and say "folks, spend!"...

    1. Cig,

      Like many you have completely missed the point. This is a macroeconomic post: microeconomic examples simply do not work. I can only assume you have not read the preceding posts. However, Chris Dillow has explained the point rather more lyrically here:

      I do not deliberately assign any "causation", but I in the equation is a residual so can't be directly influenced as Asmussen suggests - as I said in the post, the money has to come from somewhere. However, Asmussen DID assert causation: he said that more domestic investment would result in a fall in the trade surplus "by definition". I was using national income accounting to demonstrate that it would not.

      I am certainly not a believer in "all-powerful governments". But as I noted in the previous posts, persistent trade surpluses are as much a sign of structural problems as persistent trade deficits - not surprisingly, because they are two sides of the same coin. In other words, persistent trade imbalances are caused by impediments to trade, usually due to protectionist behaviour by government. The other side of a trade surplus is inevitably a capital deficit: if governments wish to address their capital deficit they have to adopt liberal trade policies to eliminate the impediments that cause the trade imbalance. And that means giving up their fondness for trade surpluses - as I said in my first post in this series.

      I would say exactly the same about trade deficits, by the way - except that we have something of a "moralising" attitude to persistent trade surpluses which makes people much less willing to accept that they are caused by impediments to free trade. Generally governments don't deliberately set out to create trade deficits - but they do set out to create trade surpluses.

    2. In this sentence:

      "It simply is not possible to increase I without some other component falling *first*"

      what is the purpose of the word "first" other than to assign causality? (and deliberately so as you put in in bold type!)

      Without the word "first", the claim is 100% correct. You basically have a static model that provides no information on dynamic behaviour.

    3. My comment refers to the fact that I is a residual. Or if you prefer, it is the dependent variable.

  13. Here's a slightly classical account - not Keynesian or national accounting that scares some readers. I agree with your post but arrive at same conclusion from a different direction.

    When the sales value generated by the demand for a country's exports exceeds the sales value of its demand for imported goods, the excess value appears as surpluses in the country's trading account. The surplus is an excess demand (for exports relative to imports) and, as such, a pressure for prices of exports to rise relative to imports. There is no internal eurozone system of exchange rates to do this. Hence the argument for reforms to attain perfectly working competitive markets. However, if the price mechanism works then this excess demand transmits a demand that leads to higher prices of resources producing German exports ie it ought to be signaling a rise in wages.

    However, here lies Schäuble's "down to shop-floor level, companies and unions working together to make labor more flexible", it is a strange meaning of flexibility that restricts German wages by agreement from being upward flexible.

    When the price of labor is suppressed - kept below the level that would help to remove the surplus in the German trading account - then the demand for other inputs (producing any given level of output) and other production mixes are also suppressed { factor substitution) . This undermines demand for alternative inputs - that is real investment - distorts the production mix and screw up the economy's 'production function". Not allowing the trading surplus to disappear by suppressing wages thus damages the efficiency and growth potential of an advanced industrial economy.

    And so trading surpluses create investment problems and this is what the statistics show - Germany has a huge private investment gap.

    "The share of investment in gross domestic product is one of the lowest among industrialised countries. It has been declining rapidly, from an average of 23 per cent in the 1990s to less than 17 per cent today" Investment," see

    The foolish notion is that interfering in a country's labor markets to keep wages down will create a "competitiveness" that automatically restores growth. If it destroys the economy's investment mechanism, it can't - at least by peaceful means. It threatens growth and the economy's future competitiveness.

    A second foolish idea, is the macho idea that surpluses reflect a superior method of production or product: the German brand. There are better cars in the world than Mercedes; but they don't dominant sales. There are worst cars in the world than Mercedes, but they can dominate. It all a question of price. And if Indian (which they are doing) and China invest better than Germany, then even Mercedes won't be cheap enough.

    So one way to see German trading surplus is as losses to the very Germans that produced them. At one level it is the result of underpaid effort in producing excess output on the part of German workers who are also overpaying for the imports they consume. At another level, it is answered by asking what returns and benefits did such a sacrifice bring? Let me put it another way, where's the investment? The darker side of the trading imbalances and financing of deficits you have dealt with in your earlier posts.

    Trade surpluses eventually become a losses to the German people. They are not good. They don't represent how well you are doing, but how well you could have been doing.

    But the point is this. German trade surpluses is also a German investment problem . Unfortunately, as an imbalance does not exist by itself, it also breaks the European investment mechanism and by doing so moves Germany and Europe closer to being a cheap wage economy.

    From @albertjohn

  14. It may be worth pointing out that the reason wages in Germany are not higher is that post-reunification Germany struggled with unemployment for a long period of time, which had the effect that (1) the so-called "Hartz" reforms were introduced by a lift of centre government and (2) German trade unions exercised a considerable amount of self-restraint for a long time. It may also be worth noting that this trend is now carefully reversed.

    1. Actually, it is worth pointing out . As I was writing in that reply. I was thinking about German unification, cheaper East German labor, those reforms and the period in the 1990s when Germany was called the 'sick man of Europe'.

    2. Hi Cheshire Cat,

      I agree with a lot, but I do have a query with the notion that this is about a German "macho attitude". IMHO (as a teuton), the opposite is true. Because unemployment in the 90s and early 20th century is still vividly remembered, Germans are cautious about rapid wage increases which they think could threaten their job. As humans, they would be very happy to earn and spend more in principle.

    3. Hi

      I did not really mean to say Germans have a "macho attitude". I was targeting some German politicians and the idea that surpluses are strength. I also wanted a pun on the word macro - as anything macro seems to attract too much aggression in mainland Europe.

      I really wanted to make the point, particularly as I'm writing from Greece, that Germans have also lost out and are still losing out.

      i relate the unemployment in the 90s to German unification. There are similarities with the EU and monetary union except there no real lender of last resort central or central fiscal authority. There is probably not even a popular desire for one, and instead only a conflicting mess or blob of interests.

      Further I am not saying there should be rapid wage increases just purely for the sake of expenditures. i could say so, as a lot of happy Germans going on holiday to Greece would certainly make a lot happy tourist resorts happy. Instead I'm. arguing that keeping increases in wages below increases in their productivity is not a sacrifice that yields good returns when it contributes to a trading surplus.

      To compete with China and India, Germany needs better investments rather cheaper wages. it is the mix of physical and human capital that yields the productivity gains that keeps Germans living and employed on good wages.

      A final point, and this is what will make Germans cautionary about asking for higher wages. Wages are lower (and are being reduced in the name of competitiveness) in the periphery. There is mass youth unemployment and many migrant to Germany. This is unfortunate for both Germans and those in the periphery. It is also other side of the trading imbalance story.

    4. Great comment, Anonymous. I think German economic policy is driven to a considerable extent by memories of unpleasant experiences in the past.

    5. Thanks for the comments and best wishes.

  15. Hi again...

    Postscript: There is a reason why the (now slowly ending) period of surpressed wage growth in Germany started in 2005, and why German unemployment peaked in 2005 as well.

    Wages (real wages green):


    What CDU/CSU and the SPD are about to agree on is bound to lead to higher spending.

    Best wishes to all!

  16. Another little postscript: part of trade imbalances is most probably explained by measurement error. People worship data but nobody likes to check it. There are trades that are real-world balanced but show up as imbalanced in the trade numbers, because some of what is really foreign trade is miscounted as domestic consumption. I doubt the statisticians correct for it accurately. And as it happens the big exporters tend to be good at exporting things that are easy to count (objects) and net "import" things that are less easy to count (tourism induced trade, property, some services that are hard to allocate geographically, safe assets as a product, etc). This is unlikely to explain away all the current surpluses but I expect the problem to be material.

  17. Indeed, extra German investments could just mean that S goes up and NX stays the same.

  18. This kind of sloppy reasoning sometimes bothers me about Michael Pettis' work. He insists for example that China needs to orientate its economy more to domestic consumtion, and if I understand him correctly, this implies that China's current account surplus would decrease. But I do not see why this should be necessarily true. China could increase consumption and at the same time run a big currenct account surplus. NX could stay constant, I would decrease and C would decline. Saving declines but so does investment, with no consequences for the trade balance.

    1. That should be 'C would increase' of course.