Global capital and national monetary policies

Speech by Lorenzo Bini Smaghi, Member of the Executive Board of the ECB
at the European Economic and Financial Centre
London, 18 January 2007

Introduction[1]

Ladies and Gentlemen:

The issue that I would like to address today is how globalisation influences the conduct of monetary policy, especially in advanced economies.

This topic is at the centre of many academic and policy debates. One difficulty in dealing with the issue is that we do not yet have a general equilibrium model that would allow all the relevant interactions between the various forces unleashed by globalisation to be appropriately taken into account. As a consequence, the various effects produced by globalisation on economic conditions, and thus on the conduct of economic policies, tend to be examined in isolation, from a partial equilibrium perspective, and this might lead to inappropriate conclusions.

Just to give an example, there is a growing literature on the deflationary effects of globalisation analysing the effects of increased imports from emerging markets and the competition from low-wage countries, which might lead to a flattening of the Phillips curve in advanced economies. However, the same phenomenon of globalisation that has enabled some emerging economies to accelerate growth is also at the origin of higher oil and commodities prices, which have fuelled inflationary pressures in our economies. Furthermore, the increase in capital flows from high-saving emerging economies to developed capital markets, which is also a by-product of globalisation and is being studied by the literature, has lead to a lowering of long term interest rates and a flattening of the yield curve, with potentially inflationary effects, according to some.

What is the net effect of all these forces unleashed by globalisation?

The answer is not obvious, also because many other developments are taking place in parallel, related or unrelated to globalisation, such as the exchange rate policy in some emerging countries, which can also affect inflation and interest rates in advanced economies. Furthermore, it is very difficult to separate the supply-side from the demand-side effects, the trade-creation from the trade-reallocation effects, the substitution from the income effects at the global level. This complicates the conduct of monetary policy, even in large countries.

As an illustration of this difficulty, it is an open question whether the fall in real long term interest rates experienced in advanced economies over the last few years reflects a change in the world savings-investment equilibrium conditions, which should in theory be accommodated by monetary policy, or to an excess of global liquidity generated by expansionary monetary policies, which should be corrected.

I would like to start from this specific example to try assessing how various factors related to globalisation are affecting monetary policy, even in a large economy like the euro area. Unfortunately, I do not possess a general equilibrium model that could illustrate all the interactions and provide the right answers. I have to use the available - yet imperfect - analytical tools of partial equilibrium and try to combine the results of different partial analysis in order to improve our understanding of how globalisation is affecting monetary policy.

One should not be disappointed by the fact that the results of such an analysis are not clearcut. Economists and policy makers are used to that. What is more important, especially for policy makers, is to be aware of the risks of taking policy decisions in an uncertain environment, i.e. an environment of which they have only a partial understanding.

I would like to start my analysis by looking at the conditions under which, in a theoretical world of fully flexible markets, globalization influences the equilibrium level of the real interest rate. I will then examine how the current, somewhat “incomplete” process of globalization – and I will explain in more detail what I mean by “incomplete”, since indeed not all globalizations are borne equal[2] - may produce results that differ from the predictions of the benchmark model. I will then try to analyse how the current effects of globalisation, and in particular the unprecedented accumulation of global liquidity worldwide, affects the conduct of monetary policy and how the latter should react.

A general view of globalization

There are several ways to define globalization. Empirically, globalization is often associated to the rise in international trade in goods, labour and capital, which has greatly outpaced that of income over the last two decades.

A more qualitative feature associated to globalization is the rise of emerging economies, such as China and India, and their increasing importance in international trade. These trends are coupled with the international reallocation of production and specialisation. The increase in cross-border holdings of financial assets is often associated to “financial” globalization.

From a more conceptual standpoint, it might be convenient to think of a world made of two countries, say a rich North and a poorer South, using a terminology proposed a few years back by Krugman and Venables.[3] As a modelling device, globalization can be described in a stylised manner as a fall in trading costs, such as transportation costs.[4] The reduction in costs facilitates trade between countries and leads to a reallocation of production. It (arguably) speeds up the catch-up process of the South towards the living standards of the North, owing to an easier flow of capital and technology from the North to the South.

According to standard trade theory, the removal of trade barriers should have prima facie the following effects on the real economy:

  • A trade-creation effect, leading to a rise in the potential growth rate of the world economy, due to the faster catch-up of the South and to the more efficient allocation of production between the South and the North, which will arguably also benefit the North (say, the North can better concentrate on financial services production and leave manufacturing to the South);

  • A trade-reallocation effect (potentially disruptive, at least in the short term) leading to a reallocation of production which, from the standpoint of individual workers or firms implies a rise in economic uncertainty, especially in the highly specialised and/or immobile occupations or sectors in the North but possibly also in the South.

The developments in the last few years have certainly confirmed the predictions of the theory, with a substantial increase in world trade (reaching over 30 per cent or world production) and a substantial redistribution of trade and production patterns across countries, according to comparative advantage.

Globalization is also expected to produce effects on relative prices. These effects have been substantial in recent years, and have attracted a wide interest in the literature. The globalisation of low inflation, through lower import prices and increased cost competition is deemed to have facilitated the tasks of central banks in advanced economies and helped to maintain price stability. [5]

Financial Globalisation

Globalisation is expected to promote the convergence of asset returns across countries, as a result of increased capital mobility and price equalization. There is indeed some evidence suggesting that global factors matter more than in the past in determining long-term interest rates.[6] This evidence refers in particular to the convergence among developed countries, while there seems to be less evidence of real interest rates convergence between developed and developing countries. In other words, capital flows seem to have produced strong convergence of expected returns within the North, but not necessarily in the North-South direction, at least not as expected on the basis of economic theory.[7]

According to conventional models, financial integration between two economies with different levels of economic development (the North and the South) should lead to capital flows from the North (where the rate of return on capital and the expected growth are lower) to the South, with the result of raising the interest rate in the North and lowering it in the South. This is, so to speak, the “first order” effect.

There is, in theory, also a “second order” effect which plays in the opposite direction, insofar as individuals (households, firms, or even countries) perceive their future to be more uncertain, as a result of the redistributive effects produced by global competition, and seek further insurance by saving more or investing less. Both the North and the South could experience such second order effects. In principle, such effects should normally be more substantial in the North; however, in the South the absence of risk sharing mechanisms, due to market incompleteness and possibly also higher intrinsic risk aversion, could make second order effects loom even larger, as we will see in a moment.

The recent experience has contradicted the results of the standard model, as net capital has flown from the South to the North, a phenomenon known as the “Lucas paradox”.[8] This has also lowered interest rates in the North, instead of increasing them.

Some have tried to explain the counterintuitive results by arguing that the “second order” effects may have played a more important role than the “first order” ones.[9] This could be suggested by the high level of corporate savings in advanced economies, in spite of high global profitability and productivity growth, coupled with low interest rates, all of which should arguably have led to an investment boom. Although investment growth appears to be recovering in Europe and Japan, we have in fact not yet seen a full revival of investment at a global level after the sharp shortfall in the early 2000s.

It is certainly difficult to evaluate the empirical plausibility of this hypothesis, but there are reasons to doubt that economic uncertainty alone can be the dominant factor behind the low level of interest rates experienced over recent years, for at least two reasons.

First, from a more analytical perspective, it has proved difficult to build equilibrium models in which the second-order impact of uncertainty on risk premia is as large as suggested by the data, as is well known in the voluminous literature on the “equity premium puzzle”. I suppose that the same objection can be made if one attempts to explain the impact of uncertainty on precautionary savings. One would probably need to assume rather extreme model parameters (such as an enormous degree of risk aversion) to obtain that the “second order” effect dominate the “first order” one.[10]

Second, we have recently witnessed a compression of risk premia over a broad range of asset classes and a rise in risk appetite. This development seems hardly compatible with the hypothesis of an increased and generalised concern about uncertainty. Moreover, at a global level we have not witnessed an overall increase in the savings rate, which has actually fallen in recent years, despite the supposedly higher uncertainty brought about by globalization. Anecdotal evidence, however, appears to support the view that globalization is playing at least some role in discouraging corporate managers from investing, especially in long term fixed assets.

The incomplete nature of financial globalisation

The alternative view that I would like to offer is that it is not the “physiology” of globalization, but – so to speak – its “anomaly” that may be driving both the behaviour of world real interest rates and the constellation of current account deficits and surpluses. The key element in this approach is that trading integration has preceded financial integration, especially in emerging markets. As I have argued elsewhere,[11] it is the incomplete nature of globalization that may create imbalances, more than its very advancement.

As we know from the economics of information, credit markets are quite special markets. Basically, the risk of someone “taking the money and running” and defaulting on his obligations is always looming. Financial technology and infrastructure have taken hundreds of years to develop so as to deal with the problems posed by information asymmetries; a sound financial system is not created overnight. Moreover, financial development is tangled with other dimensions of development, in particular those of a legal and political nature.[12] It is not easy to create an institutional system which is both creditor- and debtor-friendly, thus reducing intermediation costs to the advantage of both parties and making transactions possible.

Borrowing constraints depend not only on the prevailing interest rate but also on the legal, technological and political conditions which cannot be easily transferred from the North to South. As a matter of fact, while it would not be surprising that over time a whole range of industries migrate from the North to the South, as a result of cheaper labour costs and other forms of comparative advantages in the latter, it is highly unlikely that the financial industry be subject to the same type of migration, at least not at the same speed. Thus, the North is likely to continue to have a comparative advantage in the provision of financial services for some time to come. Indeed even within the North, such an industry is increasingly clustered in a few financial centres. In a sense, it is perhaps wrong to term this process as an “anomalous” or incomplete one. It is perhaps anomalous compared with textbook models of perfectly functioning markets, but it may in fact be explained by the fact that financial technology is arguably the hardest to transfer from the North to the South, for structural reasons.[13]

Therefore, expected returns may differ across instruments and countries even with full capital mobility, because of the imperfections in one or more local financial systems. Building on this analysis, one can derive the apparently “perverse” result, which is however consistent with empirical evidence, that comparatively higher expected returns are associated with net capital outflows rather than inflows.

Let us consider the case in which economic agents in the South face tighter borrowing constraints than in the North, which cannot be fully relaxed, at least not sufficiently and fast enough, during the process of globalization. I use here the term “borrowing constraints” as a catchword to refer to a broad and complex set of issues associated with financial frictions and the liquidity of financial markets, whose overall effect is to increase the cost of borrowing and of lending locally. In the meanwhile, in the North we have actually seen further financial innovation and deepening of financial markets, which may have contributed to further reduce economic uncertainty and precautionary savings.[14]

Under conditions of asymmetric borrowing constraints, globalization can lead to a reduced propensity to save and a net inflow of capital in the North, with an associated rise in asset prices, and to the opposite development in the South.[15] Sizeable expected return differentials may persist between the North and the South, which are not fully arbitraged away by capital flows.[16]

The asymmetry in borrowing constraints can explain a current account constellation in which the South has a savings surplus and the North a deficit. Asymmetric borrowing constraints can also explain the link between globalization and low global real interest rates.

One (perhaps simplistic) way to express this asymmetry is to consider the financial assets and liabilities produced by North as "global" (assets and liabilities), while the assets and liabilities produced by South are instead “local”, i.e. they contain a much higher degree of asymmetric information. With globalization, net lenders in the South gain access to global assets of the North, but not all net lenders in the North gain equal access to the net borrowers in the South, because the latter’s liabilities are local. Only specialised investors, which are used to deal with asymmetric information, would be in a position to lend profitably to the borrowers in the South. In equilibrium, the opening of markets would lead to an excess demand for global assets, which determines a fall in the real interest rate in order to clear the market.[17]

Evidence on incomplete globalisation

Is the hypothesis of asymmetric borrowing constraints realistic?

It certainly seems to match the data, in particular concerning East Asian countries, which in recent years have played a decisive role in the process of globalization. In China, the absence of a modern financial – and in particular banking – system and of a proper protection of property rights is well known, and even recognised by the Chinese authorities. Since the development of a sound financial system is a complex endeavour and takes time, the current situation can be expected to continue for some time. A substantial portion of savings continues to be invested in assets of developed economies, most importantly US fixed income securities.

It should be considered that the outflow of capital from China is still limited by capital controls and restrictions on the ability of domestic residents to hold foreign assets. If these restrictions were removed, the pace of the outflows could accelerate, unless a substantial strengthening of the domestic financial system takes place in parallel, so as to strengthen confidence in domestic assets.

The largest part of capital inflows into China takes the form of Foreign Direct Investment, financed through advanced economies’ financial institutions. The latter are thus performing the bankers’ role, not only for their own countries but also for emerging economies. This has led some authors to characterise the current system as a “Bretton Woods II” arrangement in which US liquid assets act as collateral for risky FDI in the South. The US thus acts as a world banker.[18] Unlike in the “Bretton Woods I” system, however, the US is now a net importer, rather than a net exporter of capital, which marks a qualitative difference compared with the past.[19]

The distortionary effects produced by the lack of a solid financial system in some emerging markets are exacerbated by at least two factors. First, there seems to be in China a very strong “second order” effect of globalisation. The lack of safety net, the negative demographic trends and fast labour mobility (all of which, with the possible exception of demography, are at least indirectly related to globalisation) have raised precautionary savings. Since the local underdeveloped financial system is not able to channel these additional savings internally, they give rise to capital outflows.

The second, somewhat related, distortionary factor is the fixed exchange rate regime, by which the Chinese currency is pegged to the dollar at an undervalued rate, thus fuelling the accumulation of net foreign assets. In practice, the accumulation takes the form of foreign exchange reserves held by the monetary authorities, which are the counterpart of the residents’ high precautionary savings.

In other East Asian countries, like South Korea, Thailand or Hong Kong there also seems to be a “second order” effect of globalisation, as the legacy of the Asian crisis in the late 1990s has discouraged domestic investment and enhanced net savings, leading to the accumulation of foreign exchange reserves as a buffer against future crises. Although improvements have been made in the structure and supervision of the financial system and in the corporate governance and overall economic performance has strongly recovered after the crisis, there still seems to be a lack of trust by domestic residents and even by the authorities in their own financial systems.

Looking back in time, the combination of strong economic growth, a still underdeveloped financial system and some degree of financial repression is not at all new in history. In fact, the post-war growth processes in Western Europe and Japan also took place in conditions of relative financial repression. It is perhaps not a coincidence that real interest rates were particularly low around the world also in the 1950s and the 1960s, despite buoyant economic growth, particularly in Europe and Japan. To some extent, some aspects of history may be repeating themselves. This does not mean that the present situation is optimal nor that it will last forever.

Another way to test the hypothesis that financial asymmetries might be at the origin of the anomalous effects of globalisation is to compare the situation in East Asia with that in the New EU Member States, many of which are similarly fast-growing emerging economies with a good prospect of progressively catching up with the living standards of advanced economies. Similarly to Asian countries, the new EU member states have less physical capital, cheaper labour and fast rates of growth, and thus meet the conditions for attracting capital from advanced economies. On the other hand, the entry into the EU has stabilised the institutional framework, especially concerning the financial markets. Foreign financial institutions are the main driver of foreign investments. Most of the domestic banking system has reached international standards, being largely owned by foreign (mostly EU) capital, and provides efficient services to domestic residents, which do not need to invest abroad to safeguard their savings.

The situation in the New EU countries would suggest that borrowing constraints are much less relevant than in Asian countries. This matches the observation that the new member states record high current account deficits and net inflow of capital, primarily in the form of FDI but also of portfolio investment.

The comparison between the two areas does not entail any normative judgment about the sustainability of one model against the other. There are also many other factors that may affect their relative economic performance over time. Nevertheless, the comparison suggests that it is possible for an emerging market economy to open the capital account, in both directions, and still attract net foreign capital, as the theory would predict.

Latin America represents an intermediate case between East Asia and Eastern Europe. On the one hand, the presence of foreign (in particular Spanish) banks has contributed to alleviate financial frictions. On the other hand, concerns with the legal systems, persistently high inflation in some countries and some recent crises (in particular Argentina, Venezuela) pull in the opposite direction. While the region runs a limited current account surplus, this is largely due to the rise in commodity prices. In this respect it is interesting to compare Argentina (a country with partly weak institutions) with Chile (a country with strong institutions). Despite very favourable global financing conditions for emerging countries Chile continues to experience large capital inflows, to the point that for a long time it had to impose controls on their entry, while Argentina continues to pay a large risk premium on its debt.

Globalisation and other shocks

Globalisation is not the only factor that may have affected the size and direction of international capital flows in recent years. There are also other forces at play, which might blur the impact of globalisation and possibly lead to different conclusions for economic policy.

First and foremost, the increase in oil prices (according to some itself partly the result of globalisation) has negatively affected the long term growth potential of industrial economies and led to the accumulation of large current account surpluses in oil exporting countries, such as Russia and the Gulf countries, which have become major net lenders. Given their lower absorption capacity, compared to industrial countries, the world savings ratio has increased and capital has flown towards the latter, contributing to lower interest rates. While we do not have full and detailed information on the investment strategies of oil exporting countries,[20] it can be argued that their behaviour is less driven by domestic financial frictions than by the lack of sufficient investment opportunities in their countries and by inter-temporal considerations. Notably, the increase in their net lending position may be consistent with the fact that the rise in oil prices experienced in the last 5 years may not all be of a permanent nature and that oil reserves are finite.

There are other structural reasons that may explain the decrease in long run interest rates. One factor is the continuous innovation in instruments, trading techniques and strategies by financial market participants which improve market liquidity and the distribution of risks across economic agents. Another factor is the impact of demographic trends on the behaviour of certain institutional investors, notably pension funds, which increase the propensity for low risk long term bonds. These developments can affect both the level of the interest rate and the shape of the yield curve.

A different factor that may explain the low level of interest rates is the relatively expansionary monetary policy that has been implemented around the world, especially in advanced economies. The low level of short term interest rates may have induced a substantial reduction in risk and liquidity premia over the entire maturity spectrum, thus affecting long term rates, for example by acting on the incentives of institutional investors.

This development might not be entirely unrelated to globalisation. First, expansionary monetary policy might have been possible, without producing immediate repercussions on inflation, thanks to the deflationary effects of globalisation which I mentioned previously. Second, increased financial integration has facilitated cross border arbitrage transactions, inducing financial institutions to conduct carry trades, borrowing at low interest rates currencies and lending in the high interest rate ones, thus transmitting easier liquidity conditions globally. It is suggested, for instance, that the very expansionary monetary policy conducted in Japan, by keeping interest rates close to zero while economic activity has progressively strengthened, induces massive capital outflows and might affect monetary conditions in the other major economies.[21] Some observers and policy-makers have thus concluded that the downward shift of the yield curve in main industrialised countries may reflect a partly exogenous fall in term premia and may be expansionary, for given level of expected future interest rates.[22]

The impact on monetary policy

It obviously makes a lot of difference for the conduct of monetary policy if the reduction in real long term interest rates that we have been observing recently is the result of an equilibrium phenomenon, derived from real developments, or reflect excessive monetary creation by central banks. In other words, it makes a difference whether the high level of global liquidity, defined as an abundance of liquid assets in circulation worldwide, is an endogenous or “equilibrium” phenomenon, related to globalisation and its incomplete nature, or an exogenous (or “disequilibrium”) one, engineered by central banks. [23]

In the first hypothesis, monetary policy should not try to counter or to react to such a phenomenon, but rather accommodate it in order to avoid creating a disequilibrium. In particular, if the lower level of long term interest rates is the result of a general equilibrium development at the world level, mainly produced by a rebalancing of savings-investment behaviour and by other “real” and demand side effects, it should not add to inflationary pressures that would require any counteracting measure by central banks.

In the second hypothesis, instead, the low level of long term interest rates would signal an excessively accommodating monetary policy over the medium term that would sooner or later translate into higher inflation. This would require, ceteris paribus, a tightening of monetary condition that would lead to a relatively flatter yield curve, or even more inverted one, than might otherwise be the case.

The key questions are: what is the correct hypothesis for the conduct of monetary policy? What is the appropriate policy response to the current global developments? In what direction should the central bank react?

As mentioned in the introduction, the answer has to be based on the fact that we have an imperfect knowledge of the overall impact of globalisation on our economies. In the absence of a fully fledged model that allows us to estimate the effects of the different factors, related or unrelated to globalisation, we have to rely on indicators to assess the plausibility of the different hypotheses.

One could argue that the task of central banks should be easy, since the discriminating fact between the two explanations is ultimately inflation. If inflation remains lastingly low even in the presence of low short-term and long-term interest rates, then one could safely conclude that we are in the presence of an “equilibrium” development related to globalisation. If inflation picks up, then the second story, related to excessive monetary creation by central banks, turns out to be the right one. Unfortunately, central banks cannot afford the luxury to wait for inflation rising. Since monetary policy affects inflation only with a lag, they have to analyse a number of indicators which could shed some light on the relatively plausibility of the two explanations.

Therefore, I will look at a few indicators with the objective of estimating the relevance of the different hypotheses and derive an assessment of the appropriate response for monetary policy. This exercise has to be seen just as a stimulus for further research and discussion on this matter.

Let’s look first at the hypothesis that the low level of long term interest rates has been produced by the very expansionary monetary policy conducted in the major countries during the last few years.

Some indicators seem to confirm this hypothesis.

First, simple correlations between interest rates and underlying economic developments, such as those implied by Taylor rules, suggest that in the recent downturn monetary policy has been more expansionary than in the past. Second, the behaviour of monetary and credit aggregates, not only in the euro area but also at the global level, seem to confirm the ample liquidity prevailing in the current cycle. A third indicator that would confirm this hypothesis is asset price inflation, although this phenomenon may also be explained by other factors. The survey of banks’ lending behaviour in the euro area also suggests that financing conditions have been very favourable in recent years. Finally, monetary conditions have been quite accommodative not only because of domestic policy decisions but also as a consequence of carry trades, which have transferred easy monetary conditions from one area to another.

What remains to be explained is how expansionary monetary conditions have systematically affected real long term interest rates. In deep and liquid markets it should not be easy, even for a central bank, to distort the shape of the yield curve. This seems to be confirmed by the failure already in the late 1960s of “switching” operations conducted by the Fed aimed at lowering long term yields.[24] Carry trade operations, i.e. borrowing short and lending long, could be systematically profitable and affect long yields only in the presence of imperfect markets, in which long term rates do not reflect the expectation of future short term rates. On the contrary, liquidity premia seem to have fallen as markets have become more efficient in spreading risk across financial agents.

In the same vein, given the size of the markets, especially the US one, it is difficult to understand how cross currency carry trades can affect the long end of the market in the latter country, without provoking major exchange rate adjustments, in particular for the low rate currency. Some weakness of the yen has been observed, but hardly to an extent that would be consistent with a significant lowering of US yields. Again, for the cross currency carry trade to have a significant impact on long term rates, presumes that the open interest rate parity would systematically not hold, something which is still very controversial empirically.

An apparent paradox is that, with the tightening of monetary conditions, started in the US and then in the Euro area, the problem does not seem to have vanished. On the contrary, the whole question of the so-called conundrum has emerged while monetary accommodation was being withdrawn, in particularly in the US. One possible explanation for the paradox could be that monetary conditions have been tightened much less than what the increase in interest rates would suggest, given the underlying strength of the economy. Another explanation could be found in the cross-currency carry trades mentioned above, that have partly offset the effects of the monetary tightening.

Another issue to be understood is what has happened to the cumulated amount of liquidity injected in the financial system during the trough of the cycle, which doesn’t seem to have been reabsorbed with the rise in short term rates. What is the impact of a possible monetary overhang on real long term rates, and - more importantly - why is excess liquidity not translating into higher inflation, after due lags are taken into account?

Part of the response might be that the inflationary impact of growing liquidity has been compensated by the deflationary effects produced by cheaper imports, a factor also associated to globalisation. However, some empirical analyses suggests that these effects have been rather limited, and to some extent counterbalanced by the inflationary effect produced by higher oil and commodities prices.

Let’s look now at the alternative hypothesis that globalisation and real factor effects have led to a reduction in long term real interest rates.

This hypothesis seems to be confirmed by several indicators and analysis.

First, there appears to be some evidence of the negative correlation between oil prices and real interest rates over the past couple of years. This is consistent with the intuition that higher oil prices reduce the long term growth potential in industrial countries and the recycling of oil revenues further contribute to increasing the demand for assets in advanced economies.

Second, there is also evidence that the large flow of savings from emerging market economies has largely been held in liquid assets, adding to monetary holdings by financial institutions in industrial countries. The inability of the latter to invest equivalent amounts in FDIs or real assets in developing countries has led to an increase in the preference for liquidity in advanced economies’ financial systems. The flow of capital into the long end of the capital markets has increased financial deepening and has also affected the short end of the markets. To be sure, the lower return on long term assets has increased the preference for monetary assets.

There is also some evidence that other structural factors, such as demographic trends, might have had some impact on long term rates, directly and through the effect on the preferences of institutional investors.[25]

The increased liquidity of economic agents does not seem to have affected investment and consumption behaviour in a way that would suggest a different pattern compared to previous cycles. For instance, although in the euro area firm’s borrowing and liquidity position has increased substantially in recent years, also thanks to the low level of interest rates, there is no evidence that investment is behaving substantially differently than in past cyclical upturns. Similarly, although consumer borrowing has increased at a very fast pace, and assets prices have increased substantially, there is no evidence yet that aggregate consumption is more dynamic in the current cycle than in previous ones. This would suggest that the stimulating effect produced by lower interest rates on consumption and investment might be compensated by other (real) effects, directly or indirectly linked to global developments, which might also explain the increasing preference for liquidity by economic agents in our economies.

To sum up, there is some evidence in favour of both hypotheses, although the one suggesting that the lowering of interest rates is an equilibrium phenomenon seems rather robust. In fact, the two hypotheses may not be mutually exclusive. Even if the hypothesis that the lowering of long term interest rates is largely due to global equilibrium effects appears to be strongly supported by several indicators, the other hypothesis cannot be discarded. The risks for price stability stemming from unusually lax monetary conditions at the global level and low long term interest rates should not be underestimated.

What should then be the optimal policy reaction, aimed at minimising risks?

The evidence provided above would suggest that monetary policy should not overreact to the observed lowering of long term real interest rates in advanced economies. On the other hand, “benign neglect” would not be appropriate either. A very close monitoring of all the available indicators, in particular those arising from the financial and monetary sectors of the economy, is certainly warranted, to detect possible disequilibria arising in monetary conditions leading to potential inflationary effects. The availability of a sound monetary analysis may prove to be particularly useful in the current conjuncture.

In any case, what the analysis has shown is that further research is required on the effects of easier financing conditions on households’ and firms’ behaviour.

A corollary to this conclusion is that monetary policy might not be the only, nor even the most important, tool to deal with the risks emerging in the new environment. The strengthening of financial stability measures and the increased cooperation between supervisory authorities at the international level might be even more important.

Financial Stability Issues

Even assuming that the phenomenon we have been analysing is benign on inflation and growth, over the short to medium term, how long can it be expected to be sustained and what could happen in case recent trends are reversed?

There is no clear answer to this question.

The issue is certainly relevant for the stability of national financial markets. Indeed, financial globalisation, be it complete or incomplete, increases the possibility of contagion, between countries and institutions. In this respect the information available is scarce and certainly insufficient to conduct appropriate supervisory activity, even more so than for monetary policy.

In spite of the increased integration of financial markets, there is still scarcity of information about the behaviour of key players, at the global level. This is not only the case of major financial operators, like hedge funds, but also of several important monetary authorities in emerging markets which do not disclose their official reserve holdings. For example, the central banks and monetary authorities that disclose information on their reserve holdings represent only two thirds of world foreign exchange reserves. Many important institutions, in particular in Asia and the Middle East, do not provide information, not even to the other authorities, on their investment transactions. We repeatedly experience increased volatility in the markets when noise is spread about changes in portfolio composition.

Being a member of the international community should entail more transparency and exchange of information with the other authorities on the holding of foreign assets.

The lack of adequate information on cross border international flows and balance sheet positions of the key players in financial markets makes it difficult for national authorities to assess the fragility and vulnerability of their own economies and to take the appropriate measures in case of a large shock.

Finally, the analysis developed above suggests that an important factor in the adjustment of global imbalances resides in the improvement of financial market structures and conditions in emerging markets, in particular in Asia. In theory, such improvements should have gone in parallel with trade integration, so as to ensure a balanced development of globalisation. As globalisation has instead proceeded at different speeds, a quick catching up would be highly desirable in the financial systems, to avoid the further widening of imbalances.

Thank you for your attention.

References

Bernanke, B. (2005): “The global saving glut and the US current account deficit”, Sandridge Lecture, Richmond, Virginia, March 10.

Bergin, R. and R. Glick (2006): “Global price dispersion: are prices converging or diverging?” San Francisco Federal Reserve Bank Working Paper n. 2006-50.

Berument, H. and R. T. Froyen (2006): “Monetary policy and long-term US interest rates”, Journal of Macroeconomics, forthcoming.

Bini Smaghi, L. (2006): “Global imbalances – global policies”, remarks at the opening of the 253rd academic year of the Accademia dei Georgofili, Florence.

Bordo, M. D., Eichengreen, B. and D. A. Irwin (1999): “Is globalization today really different than globalization a hundred years ago?”, NBER Working Paper n. 7195.

Caballero, R., Farhis, E. and P.-O. Gourinchas (2006): “An equilibrium model of global imbalances and low interest rates”, NBER Working Paper n. 11996.

Campbell, J. R. and Z. Hercowitz (2006): “The role of collateralized household debt in macroeconomic stabilization”, Federal Reserve Bank of Chicago Working Paper.

CEPR (2006): Pension Funds: Dealing with the New Giants, Geneva Reports on the World Economy, 8.

Chinn, M. and H. Ito (2005): “Current account balances, financial development and institutions: assaying the world savings glut”, NBER working paper n. 11761.

Dooley, M., Folkerts-Landau, D. and P. Garber (2003): “An essay on the revived Bretton Woods system”, NBER working paper n. 9971.

Eichengreen, B. (2004): “Global imbalances and the lessons of Bretton Woods”, NBER Working Paper n. 10497.

Fogli, A. and F. Perri (2006): “The Great Moderation and the US External Imbalance”, NBER Working Paper n. 12708.

Greenspan, A. (2004): “Globalization and innovation”, remarks at the Conference on Bank Structure and Competition, sponsored by the Federal Reserve Bank of Chicago, Chicago, Illinois.

Krugman, P. and A. Venables. "Globalization and the Inequality of Nations", Quarterly Journal of Economics, Vol. 110, 1995, pp. 857-880

Lane, P. and G. M. Milesi-Ferretti (2006): “Europe and global imbalances”, paper presented at the 7th Jacques Polak Annual Research Conference, Novembre 9-10.

Levine, R. (1999): “Law, finance and economic growth”, Journal of Financial Intermediation, 8, 1-2, pp. 36-67.

Lucas, R. (1990): “Why doesn’t capital flow from rich to poor countries?”, American Economic Review, 80, pp. 92-96.

Mendoza, E. G., Quadrini, V. and J.-V. Rios-Rull (2006): “Financial integration, financial deepness and global imbalances”, paper presented at the 7th Jacques Polak Annual Research Conference, Novembre 9-10.

Miller, M. and L. Zhang (2006): Fear and market failure: global imbalances and self insurance, CEPR Discussion Paper n. 6000.

Mishkin, F. (2005): “Is financial globalization beneficial?”, NBER Working Paper n. 891.

Rajan, R. (2006): “Is there a global shortage of fixed assets?”, remarks at the G-30 meetings, New York, 1 December.

Reichlin, L. (2006): Panel remarks at the Thirteenth International Conference "Financial Markets and the Real Economy in a Low Interest Rate Environment", Bank of Japan, Tokyo.

Rudebusch, G. D., Sack, B. P. and E. T. Swanson (2006): “Macroeconomic implications of changes in the term premium”, Federal Reserve Bank of San Francisco, Working Paper n. 2006-46.

Ventura, J. and F. Broner (2007): “Rethinking the Effects of Financial Liberalization in Emerging Markets”, mimeo.



[1] I thank L. Stracca for his input in the preparation of the speech and L. Dedola, M. Rostagno and E Dorrucci for comments and background material. The views expressed reflect those of the author.

[2] See Bordo, Eichengreen and Irwin (1999) on a comparison between the current process of globalization and that experience at the turn of the 20th century. They conclude that the current process is more important quantitatively and more broadly based.

[3] See Krugman and Venables (1995).

[4] Note that the transport costs are used as a metaphor, since there is no much evidence of a substantial decline in transport costs over the recent decade or so, and costs may have actually risen in the wake of the increase in oil prices from 2000 onwards. See Bergin and Glick (2006) on the possible implications of rising oil prices on transport costs and international trade.

[5] See, among others, Greenspan (2004).

[6] See for example Reichlin (2006).

[7] In fact, the bulk of capital flows still takes place between advanced countries; see Mishkin (2005).

[8] See Lucas (1990).

[9] Concerns for higher uncertainty created by globalization could manifest themselves in higher precautionary savings especially in the South, i.e. a “savings glut” (see Bernanke 2004), or in investment restraint by corporations in the North (Rajan 2006). While different, the overall effect of both influences would be the same, i.e. to lower real interest rates around the globe.

[10] See, however, Miller and Zhang (2006) for a contrarian view. In their model, a precautionary motive in the South can produce large global imbalances by using non-standard preferences, in particular Loss Aversion.

[11] See Bini Smaghi (2006).

[12] See, among others, Levine (1999).

[13] See Mishkin (2005) for similar considerations.

[14] See Campbell and Hercowitz (2006) on the nexus between financial development in the United States and the Great Moderation. Fogli and Perri (2006) argue that, as a result of a relative fall in economic uncertainty in the US, this country may ceteris paribus attract capital inflows due to the reduced need for precautionary savings. Quantitatively, they find that the fall in business cycle volatility could account for about 20% of the total US external imbalance.

[15] The argument is developed formally in Mendoza, Quadrini and Rios-Rull (2006). See also Caballero, Fahri and Gourinchas (2006).

[16] See Lane and Milesi-Ferretti (2006) for an empirical evaluation for the US economy. The authors show that the US systematically gains more returns on its assets than it pays out on its liabilities, and they link this result to a “liquidity premium” to be paid on US assets.

[17] For a theoretical model of the interplay between globalization, financial liberalization and persisting borrowing constraints (related to limited enforcement of contracts and weak institutions in South), as well as the effect of this constellation on North-South capital flows, see Ventura and Broner (2007). Also, the reader who is old enough could remember a similar debate in the 1960s on the concept of “financial repression" and in particular on the importance of borrowing constraints in keeping interest rates artificially low.

[18] See Dooley, Folkerts-Landau and Garber (2003).

[19] See Eichengreen (2004) on the differences between the present situation and the Bretton Woods system.

[20] For example, in the Middle East oil producing countries official foreign assets are held by specialized agencies, not by central banks. These agencies are quite secretive institutions, which do not disclose their investment strategies. Moreover, investment is typically channeled in hubs such as London or offshore centres such as the Cayman Islands.

[21] It should be noted, however, that near-zero interest rates in Japan are more than a decade old, while low global long-term rates represent a more recent phenomenon.

[22] Rudebusch, Sack and Swanson (2007) report some tentative evidence that changes in term premia are expansionary in the US, but with a large degree of uncertainty both in the measurement of term premia and in their effect on the economy.

[23] Note that if the yield curve is shifted by globalization then global “excess” liquidity would not be in “excess” at all, and could simply be the equilibrium portfolio allocation following a fall in equilibrium (nominal and real) interest rates, in a standard money demand framework.

[24] Indeed there is some evidence that the impact of monetary policy on long-term rates may have fallen over time together with a stronger anchoring of inflation expectations; see for example Berument and Froyen (2006).

[25] See CEPR (2006).

Speaking engagements

Media contacts