The analysis that follows accompanies Manuel’s political overview, over at GEM, of the recent events in Germany as well as Edward’s economic survey of the current state of play in the German economy. Essentially we are going to have a look at, arguably, one of the more salient features of the German economy in the recent period, namely that of her dependence on exports to grow. What we are going to ask here is then furthermore whether this presence of export dependency is related to the fact that Germany is one of the oldest economies in the world measured on median age (currently running at approximately 44 years). This is a bold claim and if it is unlikely that we will be able to provide decisive evidence for our claims that Germany; 1) is dependent on exports to grow and 2) that this can be traced back the economy’s ageing population, we hope that at least we will provide some perspective. As an editorial note, the arguments presented follows closely Vistesen (2010) which is essentially a working paper under preparation at this point in time.
In the first instance, it is worthwhile to point out that while export dependency in the form it will be presented here enjoys very little, if any, backing in the academic literature it remains one of the more popular ways to narrate the German situation in the context of its recent economic performance, not least in a financial crisis context (see e.g. Martin Wolf Martin Wolf (2008) – Global Imbalances Threatens the Survival of Free Trade). The main question which arises is thus how the global economy will, or indeed can, emerge in a situation where hitherto external deficit nations (the US, Spain, etc) now have to live less off of foreign borrowing while those surplus nations supporting these deficits cannot arrive at stimulating domestic demand. This question which ties together a lot of the contemporary discourses on the global economy is important to keep in my as we move along into a more static world of academic theory.
What is Export Dependency?
We define export dependency as a high and increasing connection between the variation of total output and the variation of the external balance, the latter which will be proxied by the trade surplus in the analysis that follows. Consider consequently the following very simple empirical relationship for an open economy;
Where Y(t) is national income modeled as a simple function of its components; consumption expenditures, government spending, investments, and the current account. All partial derivatives are naturally positive and if we focus strictly on the partial derivative for national income with respect to the current account, we get;
In this simple framework, export dependency may arise as function of the increasing importance of this derivative in explaining the variation of total output which simply means that the extent to which we observe the trade surplus as an increasingly strong driver of economic growth we are moving closer to a state of de-facto export dependency.
Now, this still does not answer the question of exactly what export dependency is the present context since while it is certainly one thing to be export oriented or perhaps even export reliant, the term dependency implies a much strong and potentially malignant situation as it translates into a situation where domestic economic activity becomes overtly reliant on matters elsewhere and external to the economy.
In this sense, export dependency arises in the distinction between an economy where external demand may simply be an extra boost to growth beyond a vibrant domestic economy and an economy where domestic activity is not sufficiently able to spur growth to an acceptable degree and where the positive contribution from extra demand become a prerequisite to maintain growth. It is exactly in this context that the demographic perspective becomes interesting since one obvious and intuitive consequence of the prolonged process of ageing as a result of lingering below replacement fertility as seen in Germany is that domestic demand proxied by consumption and investment demand will decline to reflect the decline in the labor force relative to the total population. This would then be a natural consequence of a standard life cycle analysis set in a closed economy where savings have to equal investment in every period.
Of course, the rub which arises as a result of this quite natural and logical process is that the need to keep and maintain economic growth does not dissipate with ageing. If anything, in a society such as the German this need will remain, or even increase, as an ever growing headache in the context of how to maintain (or viably reform) key institutional structures such as pension and health care systems whose continuing existence, whether one likes this or not, is contingent on headline economic growth. In this sense, the open economy opens up a window of opportunity to fight this situation and essentially to make up for an increasingly lackluster domestic economic environment by exporting excess capital and capital goods abroad to earn a return either in the form interest income of a pure addition to growth in the form export revenues.
Demographics and Open Economy Macroeconomics
If the introductory section above should convince you what export dependency is and why it may be related to ageing the fundamental question here is whether and how demographics may ultimately act as a driver of open economy macroeconomics and international capital flows. The empirical and theoretical contributions here are extensive, so we won’t deal with them in detail but merely point out that the idea that demographics may affect capital flows is not an original postulate and has been investigated in seminal contributions such as Summers et al. (1990), Higgins (1998), Feroli (2003), Bryant (2006), Supan et al. (2007) and Ferrero (2007) to mention just a few.
Despite considerable uncertainty surrounding the exact effects, these studies jointly find that demographics indeed do form strong drivers of open economy dynamics and that, by a applying a standard life cycle framework, come reasonably close at providing some interesting results even if for example the hypothesis of rapid dissaving is still highly contested. But what is a standard life cycle framework then in the context of the effect of demographics on the current account? Well, life cycle theory as postulated by Franco Modigliani and Richard Brumberg (see Deaton (2005)) simply states that a consumer’s saving pattern will be hump-shaped to reflect the fact that she will need to spend her working years saving for retirement where she, by definition, will not be receiving labor income. This also indicates that the current account should follow a similar pattern if modeled entirely as a function of the age structure of society
This figure is essentially drawn free hand on the basis of the empirical estimations of Higgins (1998) and comes with an important interpretation. According to Higgins (1998), and in a general life cycle perspective, we need to distinguish between two centers of gravity as economy moves through the demographic transition: One is when the demand for investment is largest as a function of age and thus when, one could argue, the capacity to absorb external investment is largest (e.g. the demographic dividend). Such periods are traditionally thought to be associated with a current account deficit since whereas investment demand may be large, domestic savings are not likely to be adequate to cover this thus implying a current account deficit to the extent that foreign savings stand ready and able to move. The second period occurs later as the total labour force begins to decline relative to the total population. This has the effect of depressing investment demand, but since this is also the period in which savings are maximized (really not counterintuitive when you think about it), it should, all things equal, be associated with a current account surplus.
The green line is a pure hypothecial construct and has no empirical counterpart (yet). It essentially represents a constraint in the form of the economy’s need to rely on external demand to provide economic growth. In this sense, the constraint is non-binding up until the economy moves into old age (say a median age of >42 years) after which the expected result from ageing based on the life cycle theory stands in stark contrast with the fundamentals of an ageing economy where external demand becomes one of the only real sources of continuous growth in headline GDP. Remember here, as a rather important point aside, that economies such as Germany need this growth if they want to maintain the continuity of their welfare societies.
Within the typology of Malmberg, Germany would clearly be in the ageing phase, but since the threshold here is very unclear as a general rule, we could say that it is placed somewhere in the maturity phase inching over to the ageing phase and thus the period in which we should hypothetically observe dissaving as the savings supply will decline faster than investment demand. The first thing to observe here is then that the fact that Germany is running a surplus is not surprising given this model framework since, but the crucial and all encompassing question becomes whether this becomes a de-facto state of dependency with respect to economic growth since the decline in domestic demand will continue to depress the potential growth rate of the economy. In the jargon of the typology above, thesis of export dependency attacks the assumption of dissaving and postulates instead that keeping domestic savings above domestic investment demand and thus running an external surplus is one of the only ways in which an economy such as Germany may hope to achieve the desired growth rates as she, inevitably, moves into the unknown with an increasingly skewed old age structure.
An Empirical Perspective
In order to get a hold of the argument in the specific context of Germany it is worthwhile to start out with the following charts which depicts the ratio of consumption to GDP, the ratio of the trade surplus to GDP, the ratio of government spending to GDP as well as the ratio of total savings (including the trade surplus) to GDP. The data is taken from OECD in current prices (seasonally adjusted) and covers the period Q1-1960 to Q3-2008. All manipulations are based on own calculations.
The first thing to notice is that up until the very end of the 1990s the trade surplus did not represent a substantial part of total income in Germany. If we take the large perspective we can say that up until the latter part of the 1980s Germany was running consistent trade deficit, a deficit which narrowed substantially throughout the 1990s and moved into a substantial surplus at the entry to the 21st century. Moving on to consumption it is possible to observe a slight negative trend in the share of consumption to GDP and especially in recent 10 years, the negative trend has been strong and thus we could say that decline in consumption has been substituted for an increase in the trade surplus, at least; it would appear so from just eyeballing the graph. It is also interesting to observe that the point in time where consumption to GDP peaked from somewhere around 1975 to 1985 coincides quite nicely with the peak of the trade deficit in a historical context.
Moving on to the share of government spending to GDP the trend is, contrary to the corresponding figure for consumption and investment, is upwards. However, there is a certain sense of optical deceit here since if you disregard the sharp increase in government spending as a share of GDP from the 1960 to the middle of 1970s, the trend has been very much like the one for consumption with the interesting detail that the share of government spending to GDP has not declined to the same extent as consumption.
If we finally look at the graph which plots the share of investment of GDP in combination with the trade surplus to GDP, it is obvious that while domestic investment has steadily declined as a share of GDP total savings have been more stable thanks to the addition from the trade surplus. In fact, we look at the total savings to GDP in 2008 it resides at the same level seen in 1970. What is crucial here of course is the composition of this saving base in the sense that the external surplus takes up a substantial part (around a quarter).
This last point is interesting with respect to the framework above since while the life cycle framework would definitely predict that Germany would be running a current surplus at the given juncture, it also predicts that, at some point, domestic savings will decline so as to make the external balance a negative contribution to output in GDP. The question we must ask ourselves is whether this is an optimal response to the increase in ageing since this process is also driven by a secular decline in the ability of consumption and domestic investment (not to mention government spending) to spur economic growth.
Specifically, it would be interesting here to check whether Germany may have reached a point in terms of its age structure (proxied by median age) where the contribution from external demand to output growth has been important. To that end, let us have a look at the following graph;
Notwithstanding the fact that this graph can hardly be seen as decisive evidence either way, it is worthwhile just observing the trend of the two series. The first thing we should note is that throughout the median age bracket 33-38 years the external balance has been deficit with a moderate trend towards balance as we move closer to a median age of 38 years. This is an interesting observation in so far as goes the life cycle framework sketched above where we can say that all things equal, this period was when Germany was moving towards its maturity phase in which it now finds itself slowly but surely moving into the horizon where we may only speculate to effects of further ageing. If we relax the typology a bit we can add that if a given economy may have a propensity to run a surplus as a function of its age structure it may only grow to be dependent on the continuing accumulation of this surplus as it enters a later stage of the age transition and it is this point that we are interested in here. Looking at the graph, we can superficially infer that the trade surplus to GDP increased sharply from the point where Germany moved above and beyond a median age of 39 (roughly the beginning of the 2000s).
This point in time is interesting since the sharp increase in the trade surplus’ share of GDP has occurred at the same time as a sharp decline in investment and consumption and thus a substitution in growth derived from internal domestic demand and towards one derived from external demand. The key to which this signifies export dependency is the extent to which this substitution in some sense is involuntary in the sense that it is a natural and necessary (and optimal?) way to compensate for the fact that the ability of domestic demand to generate growth declines as an economy enters the latter part of the age transition.
One issues which is worth stressing here towards the end is that the prediction of an entry into some form or the other of rapid dissaving as a function of age seems very difficult to reconcile with the economic realities of an ageing society such as Germany’s. This is not to say that it won’t ultimately occur, but it is important to emphasize that the extent to which we will see this, it is going to be a very serious issue for the structure of the Germany economy since one would assume that with an external deficit at the same time as an inability to create domestic growth as well as a potentially very large public debt overhang the German economy and indeed society will be in a very difficult position. It then seems a more likely outcome that Germany (and other ageing economies) will fight this and one of the only ways to do this (besides reversing the underlying demographics trends) will be through keeping domestic investment accumulation persistently above the domestic economy’s ability to absorb this and thus rely on the ability to offload this excess investment off to foreign takers.
List of References
Cutler, David M., James M. Poterba, Louise M. Sheiner and Lawrence H. Summers (1990) – An Aging Society: Opportunity or Challenge? Brookings Papers on Economic Activity, Vol. 1990, No. 1: 1-56
Deaton, Angus (2005) – Franco Modigliani and the Life Cycle Theory of Consumption, Research Program in development studies and Center for Health and Wellbeing, Princeton University; the paper was presented ad the Convego Internazionale Franco Modigliani Feb. 17th-18th March 2005
Ferrero, Andrea (2007) – The Long Run Determinants of the US External Imbalance, FRB of New York Staff Report No. 295
Borsch-Supan, Axel H; Alexander, Ludwig; and Krüger Dirk (2007) – Demographic Change, Relative Factor Prices, International Capital Flows and their Differential Effects on the Welfare of Generations, NBER Working Paper No W13185
Bryant, Ralph C (2006) – Asymmetric Demography and Macroeconomic Interactions Across National Borders, Brookings Institute, the paper was presented at a conference hosted by the Reserve Bank of Australia in 2006
Higgins, Matthew (1998) – Demography, National Savings, and International Capital Flows, International Economic Review, Volume 39 (1998) Issue (Month): 2 (May) pp 343-69
Vistesen, Claus (2010) – Ageing and Export Dependency, Working Paper 2009 (forthcoming)
Japan would be another obvious candidate here and essentially Japan and Germany are very similar on this account.