Globalisation and monetary policy
Speech by José Manuel González-Páramo, Member of the Executive Board of the ECBSeminar, Suomen Pankki – Finlands BankHelsinki, 15 March 2007
Ladies and gentlemen,
I am very pleased and honoured to have been invited to speak here today at the Suomen Pankki. Thank you very much, Governor Liikanen, for the invitation and for giving me the chance to talk about the challenges that policy-makers face in the context of globalisation. And thank you all for participating in this event. Globalisation is indeed a subject of great interest for our policy and research analyses within the Eurosystem. The reason for such interest is all the more clear when we consider that our knowledge of the different channels through which globalisation is affecting key elements of the monetary policy framework, such as the inflation formation process and the monetary transmission mechanism, is still rather limited.
Allow me to begin by briefly recalling some important features of globalisation. Globalisation is the term used to describe the growing interdependence of economies via trade, production and financial market linkages. It has accelerated in the last 20 years mainly for two reasons. Firstly, because of the boom in the use of information and communication technology, which has reduced the costs of transporting goods, services and information across the globe. This process has been accompanied by a strong rise in foreign direct investment and this, in turn, has led to new ways of organising and doing business: production processes have become increasingly internationalised as companies have established affiliates abroad to gain access to foreign markets and reduce input costs [Chart 1]. The second reason for globalisation was the opening-up of emerging countries to international trade and production. The greater involvement of emerging Asia in world trade as well as central and eastern Europe following the collapse of the Soviet Union has intensified competition and caused major changes in the pattern of global trade [Chart 2].
The recent acceleration of globalisation has occurred in parallel with a decline in the level and volatility of global inflation [Chart 3 and 4]. Although the more general acceptance of price stability-oriented monetary policy frameworks, fewer negative shocks (so-called “good luck”), technological progress, fiscal discipline and structural reforms have all played a part in reducing inflation rates and macroeconomic volatility over the past two decades, globalisation is often included as a factor. Globalisation and its implications for domestic price developments raise several issues for monetary policy. Has monetary policy become less effective in a globalised world? Is the objective of maintaining price stability still relevant in the wake of globalisation? Do central banks have tools that are efficient enough to maintain price stability in a globalised world? How should monetary policy be conducted under these new circumstances? These are the questions I intend to address today. But to judge the relevance of these questions, we first have to consider the impact globalisation might have had on the variable that is most important from a monetary policy perspective: domestic inflation. For this reason, let me focus first on what theory and empirics tell us about the effects of globalisation on inflation.
What do theory and empirics tell us about the impact of globalisation on domestic inflation?
Theoretically, globalisation may have an impact on inflation through several channels. These channels include the direct and indirect effects on of: (1) increased trade openness and stronger international competition from low cost suppliers in emerging economies, which has resulted in both lower prices of imported manufactured goods and may have limited the pricing power of domestic corporations; (2) higher international competition in labour markets with the ensuing wage moderation, also out of fears of relocation abroad by domestic firms; (3) upward pressure on productivity growth, arising from a number of factors (such as better diffusion of technology, increased incentives to innovate in the face of fiercer international competition, “creative” destruction prompted by foreign competitors, etc.); (4) stronger global demand for energy-related commodities and industrial raw materials. In addition, it has been argued that globalisation may have affected the inflation formation mechanism by increasing the sensitivity of domestic inflation to foreign demand conditions.
Let me start with the trade and import price channel. The opening-up of trade with low-cost countries has put downward pressure on the import prices of developed economies. Lower import prices may depress domestic prices both in a direct and an indirect way [Chart 5]. The direct effect is via the (increasing) import content of production (intermediate and final goods) and consumption (final goods). The decline in input prices may then – at least partly – be passed on to final prices of domestically produced goods. Low-cost imports also have indirect effects on domestic prices by inducing adjustments in the domestic economy. If consumers start to substitute domestic products with cheaper imported goods, domestic firms need to become more competitive in order to avoid losing market share, either by cutting costs or squeezing profits. One way of reducing costs is to shift input demand towards cheaper inputs, notably labour, from low-cost countries. Another way is to cut labour costs. The lower costs of cross-border activities provide an additional tool for adjustment for some of the firms, namely offshoring. Due to the transitory costs of the adjustment, demand for labour is expected to decline in the short run, while the impact of increased globalisation on profits and mark-ups is less clear-cut. Indeed, firms could be obliged to squeeze their profits due to the increase in competition . At the same time, however, a possible downward impact of increased openness on both input and labour costs could be associated with increased profits if this reduction in costs is not fully passed on to output prices. This is, however, less likely the more firms are exposed to competition.
Turning now to actual import price developments in the euro area, the data indicate that there were recent episodes when import prices stagnated (1995-99) or even declined (2001-2003), suggesting that import prices may have contributed to low domestic inflation in the euro area – at least for some time [Chart 6]. Separating aggregate import prices into commodity and non-commodity components shows that the subdued rate of growth of non-commodity prices was the driving force behind the low overall rate of import prices during these episodes. At the end of 2006 manufacturing import prices in the euro area were still below the 2001 level and, in recent years, have helped non-energy industrial goods inflation to stay below 1% [Chart 7]. On the other hand, the prices of imported commodities, notably of oil, accelerated in 1999-2000 and even more sharply after 2003, pushing up the headline HICP inflation figure, particularly via the energy component, and offsetting the negative impact of lower manufacturing import prices on domestic inflation. Let me however also emphasise at this point that the changes we see at the domestic level are relative price changes, while overall domestic prices are less affected. I will come back to this later when looking at the implications of globalisation for monetary policy.
Relative to the other channels, empirical work on the import price channel is abundant, maybe because of the relatively lower level of complexity of this channel. The empirical literature addresses three main questions: how much of the import price developments are due to globalisation?; what are the impacts of imports from low-cost countries on domestic inflation?; and how do higher commodity prices affect domestic inflation?
Starting with the relationship between import prices and globalisation, there is no doubt that, besides shifting trade relations, import prices are also determined by exchange rate fluctuations, global economic activity and technology. The IMF claims that globalisation was not the main factor determining import prices in recent years for three reasons:  (1) the downward trend in relative import prices started in the early 1980s, well before the speed-up in globalisation; (2) the recent fluctuations in these prices do not appear unusual in either magnitude or persistence; and (3) they appear to broadly reflect fluctuations in global economic activity. On the other hand, based on a simple accounting framework, the ECB has estimated that the higher level of imports from low-cost countries had a dampening impact on overall euro area manufacturing import prices of approximately 2 percentage points per annum on average between 1996 and 2005. 
As regards the impact of import prices on domestic prices, the direct impact of trade with low-cost countries has been estimated by the OECD on the basis of an accounting identity, where both the impact of increasing import shares and lower price levels of low-cost countries are taken into account.  The OECD has estimated that the increasing imports from China moderated domestic price inflation directly by 0.2 percentage point in the euro area between 2001 and 2005. However, there are serious uncertainties about the size of this estimate, given the volatility of this measure. 
The above OECD estimate is mechanical and does not account for any adjustments in the domestic economy due to increasing competition, and thus may underestimate the actual impact of import prices on inflation. Indeed, according to a more comprehensive approach by the IMF, which is based on an estimation of an augmented Phillips curve, import prices contributed to disinflation in the late 1990s by 1 percentage point on annual average in the US and by 0.5 percentage point in seven other advanced economies.  Import prices also helped temporarily reduce inflation during the global slowdown in 2001-2002. 
Let me now turn to the second channel through which globalisation influences domestic prices: the impact on labour markets. As I mentioned briefly in relation to the import price channel, increased competition may make employers less likely to concede wage claims. Besides this impact however, there are at least three other channels through which globalisation may influence labour market developments and have an impact on wages. First, according to classical trade theory, the opening-up of trade with low-cost countries leads to specialisation on the basis of comparative advantage. This would imply that in developed countries production and thus labour demand shifts towards high-skilled workers. Given the rigidities affecting labour supply, this shift should result in falling employment levels and increasing wage inequalities, at least in the short run. Second, globalisation also increases the labour supply. Indeed, the available labour force in the world has potentially doubled from 1.5 to 3 billion with the opening-up of China, India and the ex-Soviet bloc.  Although labour mobility is still limited even within the euro area and the empirical evidence on the impact of migration on wages is scarce, the increased flexibility of capital flows implies that firms now have easier access to a global labour supply than before. This third channel – offshoring, or the threat of offshoring – has supposedly contributed to increasing the sensitivity of labour demand to real wages and a weakening of trade unions’ wage bargaining power in developed countries in recent years.
A brief look at the data indicates a structural break in the wage-setting process in recent years. The wage share in many major developed countries has been on a trend decline since the 1980s, indicating that the increase in real wages has remained well below productivity growth [Chart 8]. Euro area annual nominal wage growth has hovered around 2% despite recent oil price shocks hitting the economy and unemployment reaching historical lows. Again, the moderate wage developments can be attributed to several factors other than globalisation, such as technological progress, structural reforms of labour markets, a credible price stability-oriented monetary policy and well-anchored inflation expectations. But public debate has often focused more on the impacts of globalisation: how plant relocation and immigrants have threatened domestic jobs and pushed down the wages of domestic workers.
In light of the strong media coverage, the empirical evidence on the impact of globalisation on the level of aggregate wages and employment is somewhat disappointing. An IMF analysis revealed a significantly negative, but small 0.2 percentage point impact of openness on manufacturing wages for a sample of industrialised countries. Estimates for the euro area tend to indicate stronger negative impacts on employment than on wages, supposedly due to existing nominal wage rigidities. [Chart 9]
The third channel via which globalisation affects domestic prices is the productivity channel. What trade theory tells us is that increased world trade allows countries to better exploit their comparative advantages and economies of scale. This more efficient allocation of resources enhances productivity. Also, increased competition will push the least efficient firms out of the market, thus driving up average productivity. Firms facing increased competition will come under pressure to innovate, because new production technologies increase productivity and reduce production costs. Finally, with lower barriers to flows of information and cross-border production schemes, globalisation encourages the transfer of new technologies between countries, firms and sectors.
Despite these clear theoretical considerations however, so far there has been little direct and empirically robust evidence of the extent to which globalisation may have boosted aggregate productivity either in the euro area or in the US in recent years. As I see it, this can be explained by several factors, such as the difficulty of separating globalisation effects from technological development, or short-run adjustment costs from long-run positive effects.
Let me now turn to the channel through which globalisation may have an impact on the inflation mechanism: the declining sensitivity of domestic prices to domestic capacity contraints. Increased openness may imply that domestic demand is less constrained by domestic output. In other words, stronger import competition limits the possibility of domestic firms raising their prices during domestic booms, which results in a lower effect of domestic demand on domestic price developments. Moreover, the increase in openness would also cause the prices of domestically produced goods to be increasingly determined by foreign developments via higher import penetration and the stronger export orientation of firms.
The idea of globalisation weakening the sensitivity of prices to domestic conditions has not gone unchallenged in academic thinking, mainly for two reasons. First, there are factors other than globalisation that appear to contribute to the lower sensitivity of prices to domestic output gaps. The global disinflation process in the last two decades has resulted in a lower inflation environment, in which price adjustments are less frequent. Also, the stronger credibility of monetary policy has helped to anchor the expectations of economic agents. Consequently, in formulating their inflation expectations agents are less prone to extrapolate past trends in inflation and thus economic agents are able to “see through” the cyclical volatility when setting their prices or making their wage claims. On the other hand, it is often stated that the sensitivity of prices to domestic conditions should have increased, rather than decreased, in recent years. Indeed, stronger competition leads to lower profit margins and less room for manoeuvre for firms. As a result, their response to changes in the cost structure or demand conditions should be faster. The structural reforms of the recent decades, which aimed at higher flexibility in both product and labour markets, should also have raised the sensitivity of prices to domestic demand conditions.
Despite stronger competition and greater flexibility of markets, empirical evidence seems to suggest that the sensitivity of inflation to domestic demand conditions has declined in many developed countries in the last two decades. In technical jargon, this if often referred to as a “flattening” of the national Phillips curves. According to the estimates of the BIS , the coefficient describing the cumulative impact over one year of the domestic output gap on prices in the euro area declined from 0.4 over a sample covering the 1980s and early 1990s, to 0.1 over the sample period 1993-2005.  Moreover, the results show that using a proxy for global economic slack adds considerable explanatory power to traditional inflation equations. The significance of the global factor increases in their sample over time, thus partly taking over the role of domestic demand. However, when recalling these results, we should not forget the uncertainty about the measurement of the global output gap. Indeed, when other external factors such as import prices and oil prices are included in the equation for the euro area, the global factor becomes insignificant, indicating a possible correlation of the global gap measure with other external factors .
Commitment to maintaining price stability in a global economy
We are now familiar with the main channels through which globalisation is affecting the economic environment in the euro area. Let me move one step further and draw some tentative implications of globalisation for the conduct of monetary policy. In this part of the speech, I would like to address three issues of vital importance to central banks. First, is the ability of central banks to maintain price stability impaired by the unleashed forces of globalisation? And second, how should central banks respond to the challenges presented by globalisation.
No matter whether we are living in a global world or not, inflation, in the long run, is without any doubt a monetary phenomenon . As a consequence, globalisation does not affect the central role and overriding responsibility of central banks to preserve price stability. Central banks are the ultimate guardians of purchasing power. Moreover, a solid anchoring of inflation expectations at a level that is consistent with the central bank’s definition of price stability becomes even more important in a rapidly changing global economy, where inflation dynamics can be affected in various ways and adverse inflationary shocks are more easily transmitted from one country to another . Let me illustrate these points with a simple but revealing example.
One of the effects of globalisation, as I mentioned early on, has been the sharp rise in energy prices over the last few years, as an increasing number of countries have been competing for more or less the same oil resources. In this context, let me present two different scenarios, one in which the central bank credibly commits to maintaining domestic prices stable, and another where such a commitment is absent. In the first scenario, the corporate sector, which uses energy as an input for its production processes, observes a significant and sharp rise in the price of oil. Since there is a credible commitment to price stability over the medium term and inflation expectations are well anchored at low levels, companies easily understand that central banks will not accommodate the inflationary pressures arising from the shock. Therefore, they have less incentive to fully pass on the increase in input prices to consumers because, if they did, the relative price of their production would increase and their competitiveness could be significantly eroded, especially in an increasingly competitive global environment. As a result, the oil-consuming corporate sector is forced to cut its mark-ups. The least competitive businesses may be driven into liquidation as mark-ups shrink, and unemployment may increase in the short run while economic resources are transferred to other, less oil-intensive, sectors. But, in this simple example, second-round inflationary effects are very limited, the economy has coped with a potentially very damaging oil-price spike by reallocating its production and employment away from oil-intensive sectors, it has “creatively destroyed” the least efficient businesses and it has minimised macroeconomic fluctuations in output and inflation.
Let me now turn to the second scenario, where a credible commitment to maintaining price stability is lacking. In this case, firms see the price of energy rising but, as inflation expectations are not well anchored and economic agents have imperfect information about real-time economic developments, companies may not be able to tell whether they are witnessing an increase in the relative price of energy or a comparable increase in the aggregate price level . If they mistakenly conclude that they are facing a shock to the aggregate level of prices, companies find it optimal to pass on the upsurge in input prices to consumers, as they do not perceive that this would make them relatively less competitive. As a result, consumer price inflation jumps, confirming firms’ self-fulfilling expectations. Furthermore, when unions and workers notice that consumer prices are rising, and their real wages are falling, they try to negotiate wage increases that set the real wage above its initial level because they want to protect their purchasing power in case inflation keeps rising at that pace. At the same time, nominal interest rates increase across the maturity spectrum, as they incorporate higher inflation expectations and inflation-risk premia. Consequently, firms face much higher production costs, not only from the increase in energy prices, but also from higher real wages and financing costs. Therefore, many firms, and not only those in the oil-intensive industry, may be driven out of business, followed by a decline in production and employment throughout the economy.
This simple example is only meant to illustrate the benefits of a credible commitment to price stability. However, the experience of the oil crises of the 1970s - which were triggered by a relative-price shock – reminds that the combination of high oil prices, high inflation, high unemployment and high nominal interest rates, in an environment where inflation-fighting credibility is not well established, can be much more than a theoretical curiosity. Accordingly, in a global economy, where changes to relative prices are significant and dynamic, a clear definition of price stability from the central bank provides concrete benefits to the public: it creates a credible reference point for guiding price and wage formation mechanisms, it establishes a stable anchor for inflation expectations embedded in medium and long-term interest rates, and it ensures that the purchasing power of our currency will not be eroded unexpectedly.
Ability to maintain price stability in a global economic environment
There are good reasons for a strong and credible commitment to maintaining price stability, even in the face of globalisation. But can central banks honour this commitment? Are they able to achieve and maintain price stability in a highly integrated global economy? Let me try to answer these questions from two different but complementary perspectives. First, I will very briefly address to what extent the monetary policy transmission mechanisms have been affected by globalisation. And second, I will analyse what the policy implications are of the changes that globalisation triggers in the Phillips curve.
The transmission mechanisms are the channels through which monetary policy decisions affect the economy . Probably, the best-known of them is the interest rate channel. While the monetary policy instrument is a very short-term interest rate, savings and investment decisions are more influenced by long-term rates. However, in recent years and coinciding with the strengthening of globalisation (and perhaps not only coinciding according to the “savings glut” view), the response of long-term rates to changes in the policy instrument has been in relative terms more muted. For instance, while the ECB has hiked its policy rate by 175 basis points since December 2005, the yield on the five-year German bonds has increased by around 80 basis points only, with the yield on the ten-year bond increasing by just around 50 basis points . To the extent that the link between policy rates and longer-term rates has loosened, it could at first sight be argued that central banks may have a harder time trying to preserve price stability in a global environment. However, we should recall that, according to the “expectation hypothesis of the term structure”, long term interest rates are also linked to the policy instrument through the expectations of market participants, and that these expectations are to a considerable extent shaped by the monetary policy framework. Thus, by committing to a transparent and credible monetary policy strategy and consistently implementing it (while also openly communicating the assessment underlying its decisions), a central bank can provide guidance to market participants on how future policy rates are likely to respond to economic developments, while also anchoring inflation expectation. In doing so, it can retain substantial leverage on long term interest rates.
Nevertheless, this is not the end of the story, because globalisation may also affect other transmission channels. In a global economic environment, more and more investments are placed overseas, and an increasing share of domestic savings is invested in international financial markets. Indeed, globalisation has granted households and firms easier access to international financial markets. As a result, an increasing share of household wealth and company balance sheets has become relatively more sensitive to fluctuations in asset prices. In other words, the wealth channel and the balance sheet channel may have become more important in the transmission of monetary policy over the last few years.
Overall, an analysis of the transmission channels does not suggest to me that, in a global environment, central banks cannot honour their commitment to price stability. What seems to be true is that our job as central bankers has become even more demanding, because monetary policy needs to be conducted at a time when the relative importance of various transmission channels may be changing. In this respect, more research on how monetary policy feeds through the economy in a global environment is definitely needed.
A complementary way of exploring whether the ability of central banks to stabilise prices has been impaired by globalisation is to look at the Phillips curve, which is the link between domestic inflation and domestic demand conditions. Central banks influence domestic inflation mainly through domestic demand. Therefore, the analysis of the slope of the Phillips curve is very relevant to investigate if central banks are able to deliver price stability in a global economy. While parallel shifts of the Phillips curve are also very likely effects of globalisation, the ability of central banks to influence inflation in the short to medium term largely depends on the slope of the Phillips curve, and this is where I will focus now.
Recalling briefly what I said earlier, it has been argued that increased openness has made the Phillips curve flatter . In this context, a flatter Phillips curve would imply that the sacrifice ratio the central bank faces would be larger. Therefore, ceteris paribus, more aggressive and persistent adjustments in policy rates would be needed to keep inflation dynamics consistent with the central bank’s definition of price stability.
On the other hand, as I described earlier in more detail, more competition in product markets may increase the responsiveness of domestic prices to changes in production costs. For example, under the extreme assumption of perfect competition, prices need to track every change in marginal costs, and thereby inflation becomes extremely responsive to the factors that may have an influence on costs. Domestic demand is one of these factors: a positive shock to domestic demand may push marginal costs up because, as the capital stock is fixed in the short run, more energy and labour are needed to produce each of the additional units of goods and services demanded by the domestic economy. The resulting increased sensitivity of prices to domestic demand makes the Phillips curve steeper, the sacrifice ratio falls and changes in policy rates and money supply are transmitted more forcefully and quickly to inflation.
Additionally, I have also stated that the slope of the Phillips curve may not be totally exogenous to the monetary policy-making process. An increase in inflation-fighting credibility may lead to estimates of milder responses of inflation to domestic demand conditions, i.e. a flatter Phillips curve. Naturally, the central bank’s leverage on the inflation process does not decrease even when the Phillips curve becomes flatter, if an important driver of such a development is the high credibility of the monetary authority.
All in all, this tentative analysis of the Phillips curve confirms the outcome of the review of the transmission mechanism: central banks can honour their commitment to price stability in a global environment as well, although increased uncertainty on the slope of the Phillips curve makes the conduct of monetary policy more challenging.
Monetary policy response to the effects of globalisation on inflation
Once we have made it clear that a commitment to maintaining price stability is still of the essence in a global economic environment, and that central banks are in a position to honour such a commitment, how should a central bank which aims to preserve price stability over the medium term respond to the effects of globalisation on the dynamics of inflation?
Globalisation may affect inflation for two reasons. First, the economy is gradually moving from an equilibrium “without globalisation” to a new equilibrium “with globalisation”. This new equilibrium is likely to be characterised by a different set of relative prices and, therefore, changes in inflation rates may appear during the transition.
An example of changes in equilibrium-relative prices is the increase in the relative price of oil and other commodities I pointed out before. Strong economic growth in many emerging economies has boosted the demand for a limited endowment of oil and, as a consequence, inflation rates have been pushed upwards across oil-importing countries. How should central banks react? On the one hand, they should not react mechanistically to first-round effects from oil prices: if central banks attempted to stabilise overall inflation immediately, the rest of the economy could go through a spell of inefficiently low price increases, or even deflation. On the other hand, they must be ready to tighten the monetary reins as soon as significant risks of second-round effects appear, with the aim of keeping inflation expectations firmly anchored and preserving price stability over the medium term.
Another example is the potential decrease in price mark-ups in those sectors that have become more exposed to international competition after globalisation, as I mentioned at the beginning of my speech. This has driven down the relative price of domestically produced tradable goods and services, putting downward pressure on inflation rates. As before, central banks should not offset the first-round effect on prices from a decline in equilibrium mark-ups because, if they did, they could artificially delay the welfare gains stemming from a move towards a more efficient economy, with lower price levels and more output and employment . At the same time, the central bank must also be prepared to mitigate any risk of second-round effects, which might lead to widespread price cuts in the rest of the economy and could drive long-term inflation expectations to levels below those consistent with the definition of price stability.
Therefore, central banks should not obstruct efficient changes in relative prices triggered by a move towards a new equilibrium “with globalisation”, while they are well advised to prevent significant second-round effects from spreading to other parts of the economy.
But globalisation may also affect inflation for a second reason: the combination of globalisation and structural rigidities may move the economy away from its “equilibrium path” and lead to a build-up of inflationary pressures. Let me illustrate these disequilibrium phenomena and the appropriate policy response to them with an example.
Globalisation has enhanced competition in domestic markets. When I described the productivity channel, I remarked that more competition could trigger more innovation by domestic firms, helping to increase productivity and potential output in the domestic economy. If, however, some real rigidities led to sluggish changes in domestic demand, a negative output gap would open and downward pressures on domestic inflation would appear. In this scenario, the central bank may want to cut its policy rate to let demand catch up with supply and thus reduce the first-round effects on inflation from excess supply conditions.
This example shows that the combination of globalisation and structural rigidities may move the economy away from its equilibrium. Then, the monetary authority needs to fight the first-round effects on inflation arising from such disequilibrium developments.
It is easy to understand that, ideally, central banks would like to fully disentangle the effects on inflation coming from equilibrium and disequilibrium sources in order to implement the appropriate policy responses to each of them. We all know that, unfortunately, this is easier said than done. But the fact that we face a difficult task does not mean that we should give up. Instead, we must give it our best shot. For this reason, it is our responsibility as central bankers to continuously monitor all available indicators to gain the best possible understanding of how globalisation affects inflation dynamics.
Monetary policy, Globalisation and Asset Prices
Before concluding, let me briefly consider a topic that has also attracted substantial attention from economists and policy-makers in recent years, namely how monetary policy should react to asset prices. Globalisation does not only affect inflation dynamics. It impinges on almost every single part of our economies, including asset prices. As I did before when referring to inflation developments, I would like to distinguish between the effects of globalisation that influence the equilibrium level of asset prices or, as it is commonly said, its/their fundamental value, and the effects of globalisation that move asset prices away from their equilibrium value, away from the fundamentals.
Regarding the first group of effects, those that change fundamental asset values, let me mention two examples. First, globalisation has increased the pool of workers available for production, leading to a decline in the labour share in output . The other side of the coin is that the capital share in output may increase, supporting high corporate profits and boosting equilibrium stock prices.
A second example of fundamentals-based changes in asset prices is related to economies of scale. Globalisation has opened up new markets for domestically produced products. Hence, domestic firms may want to take positions in those new markets sooner rather than later. A quick way to do this is by buying a previously existing foreign firm that has a relatively strong position in the new market. Moreover, as I said before, the resulting multinational company would benefit more from economies of scale, reducing unit costs and increasing profits. Against this background, it is not surprising that globalisation is leading to a wave of mergers and acquisitions around the globe, which exerts additional upward pressure on equity prices.
When changes in asset prices are driven by fundamental factors, or, as in these examples, by a move towards a new equilibrium with globalisation, a stability-oriented central bank should not react to them over and above their effects on inflation and economic activity. In these examples, changes in asset prices are efficient market responses to transformations in the structure of the economy. The possibility that asset prices could move close to historical highs does not justify any monetary policy reaction to asset prices per se if the new equilibrium, which might have never been experienced before, validates them.
Despite this, I said before that globalisation might also contribute to driving asset prices away from their fundamentals. Non fundamental asset price moves can be caused by international spillovers from macroeconomic disturbances in combination with financial development asymmetries, whose effects are then magnified by globalisation. For example, some emerging countries have been keeping their currencies artificially low despite relatively strong economic performance. A weak currency favours the build-up of massive current account surpluses in those countries. These surpluses can then be used to purchase financial assets in developed economies due to the asymmetric nature of globalisation: since trade integration proceeds faster than financial development, the more advanced economies retain a comparative advantage in providing financial services, leading to a net inflow of capital from emerging economies . Consequently, prices of assets in developed economies, e.g. bonds, could rise, while their returns, which move inversely to bond prices, would tend to fall.
I am not arguing that exchange rate rigidities are the main factor behind low long-term interest rates in the euro area or the United States. I am just trying to provide an example of how globalisation may help to move domestic asset prices away from what the state of the domestic economy justifies. How should monetary policy be conducted in this scenario?  Well, it has been advocated that central banks should act when asset prices deviate from fundamentals, because a disorderly correction of such deviations may give rise to instability in the financial sector. Therefore, it is said that the central bank should lean against asset price deviations from fundamentals: implementing cautious increases in the policy rate aiming at mitigating the non-fundamental rise in asset prices and helping to bring longer-term interest rates to more normal levels.
I am aware of the risks to financial and macroeconomic stability posed by sudden asset price reversals. But I am not convinced that a policy of leaning against asset price increases would help to solve the problem I have described in my example. There, the non-fundamental deviation in asset prices comes from a current account surplus in some emerging economies and the relative attractiveness of medium to long-term bonds in developed economies. Interest rate increases in developed countries are not likely to mitigate any of these factors. Actually, they could accentuate them.
More generally, leaning against asset prices may be advisable under some circumstances : (1) a high estimated probability that asset prices are driven by non-fundamental forces; (2) a low probability of a sudden reversal in asset prices in the near future; (3) a high sensitivity of asset prices to modest policy rate adjustments, and (4) a high risk of instability in the financial system if asset prices collapse. When all these conditions are met, the monetary authority may want to lean against specific asset price developments. When none of these conditions is met, as in the example above, the central bank could do more harm than good if it attempts to stabilise asset prices.
Finally, let me briefly conclude. There are four major questions I addressed in this speech:
Has globalisation changed the environment in which monetary policy operates? More precisely, has it had an impact on domestic prices?
Is the objective of maintaining price stability still relevant in the globalised world?
Has monetary policy become less effective in globalised economies?
How should monetary policy be conducted under these new circumstances?
The direct and brief answers to the first three questions are: yes, yes and no.
According to economic theory, globalisation should have had a significant impact on domestic prices in recent decades although, as I said earlier, it is hard to assess the exact size of this impact on the basis of existing empirical evidence. A greater role has been found for the direct and indirect effects via the trade channel, and possibly some dampening impact on wage claims and employment. Globalisation may have potentially reduced the sensitivity of inflation to changes in the domestic output gap, while rendering it more sensitive to developments in the foreign output gap. But overall, empirical estimates fall in a wide range and reflect high uncertainty. I believe that there are at least three factors that may explain this. First, the effects of globalisation may be too recent to show up adequately in the data. Second, globalisation can be seen as a complex shock with both upward and downward pressures on prices which may offset each other. And finally, most of the tools used so far to evaluate the relationship between globalisation and prices are partial by nature. More complex macroeconomic models may show up a more significant impact of globalisation on prices.
Turning to monetary policy implications, I would like to emphasise that globalisation does not affect the central role and overriding responsibility of central banks to preserve price stability. On the contrary, the anchoring of expectations is even more essential in a world where adverse shocks and inflationary pressures in one country are more easily transmitted to others. I also believe that monetary policy still has tools effective enough to achieve the unaltered goal of price stability. What seems to be true is that our job as central bankers has become even more demanding, because monetary policy needs to be conducted at a time when the relative importance of various transmission channels may be changing in the wake of globalisation. In this respect, more research on how monetary policy feeds through the economy in a global environment is definitely needed. The bottom line is that our role and responsibilities as central bankers remain unchanged in the globalised economy, but the ways we achieve our goals may be becoming more complex and challenging.
 Rogoff conjectures that this consequence of globalisation on monopoly pricing power should be one of the main factors behind worldwide disinflation. This view finds empirical support for the EU case in the work of Chen and others. See K. Rogoff (2003), ‘Globalisation and global disinflation’, paper presented to the Federal Reserve Bank of Kansas 2003 Jackson Hole Conference; see also N. Chen, J. Imbs and A. Scott, ‘Competition, globalisation and the decline of inflation’, CEPR Discussion Paper Series No 4695, Oct. 2004.
 See IMF ‘How has globalisation affected inflation?’, World Economic Outlook, Chapter III, April 2006.
 See the box entitled ‘Effects of the rising trade integration of low-cost countries on euro area import prices’, ECB Monthly Bulletin August 2006.
 OECD (2006), ‘Globalisation and inflation in the OECD economies’, ECO/CPE/WPI(2006)14.
 In addition, the inflation differential component of the measure is suspected to be sensitive to factors other than globalisation (e.g. exchange rate developments, domestic cycles, administrative price regulations etc).
 Australia, Canada, France, Germany, Italy, Japan, the United Kingdom.
 For the US, Kohn estimated a downward impact of 0.5–1 percentage point for the US in the 2000–2005 period. See Kohn, D. L. (2006) ‘The Effects of Globalization on Inflation and Their Implications for Monetary Policy’, speech at the Federal Reserve Bank of Boston’s 51st Economic Conference, Chatham, Massachusetts.
 Freeman, R. (2006) ‘The Great Doubling: The Challenge of the New Global Labor Market’, August 2006.
 Borio, C. and Filardo, A (2006) ’Globalisation and inflation: new cross-country evidence on the global determinants of domestic inflation’, Bank for International Settlements, unpublished paper, Basel, March.
 The IMF estimated the slack coefficient to decline from 0.3 to 0.2 from 1980 to 2004 in the sample of eight developed countries.
 For instance, see the econometric results in N. Pain, I. Koske and M. Sollie (2006), “Globalisation and inflation in the OECD economies”, Economics Department Working Papers, No 524, ECO/WKP(2006)52, November.
 See, for instance, L. Ball (2006), “Has Globalization Changed Inflation?”, NBER Working Paper 12687.
 L. Papademos (2006), “Globalisation, Inflation, Imbalances and Monetary Policy”, speech delivered at the conference The Euro and the Dollar in a Globalised Economy, St. Louis, 25 May.
 In the spirit of R. Lucas (1973), “Some International Evidence on Output-Inflation Tradeoffs”, American Economic Review 63, pp. 326-334.
 For a detailed description of the monetary-policy transmission channels, see F. Mishkin (1995), “Symposium on the Monetary Transmission Mechanism”, Journal of Economic Perspectives 9, pp. 3-10.
 The same phenomenon is also apparent in the United States. See B.S. Bernanke (2007), “Globalization and Monetary Policy”, speech delivered at the Stanford Institute for Economic policy Research, Stanford, 2 March.
 For a recent analysis of the link between globalisation and the Phillips Curve see C. Borio and A. Filardo (2006), “Globalisation and Inflation: New Cross-Country Evidence on the Global Determinants of Domestic Inflation”, Bank for International Settlements.
 While potential output increases with competition and, as a consequence, inflation falls, the efficient level of output does not change, generating a trade-off for the central bank. See M. Woodford (2003), “Interest and Prices: Foundations of a Theory of Monetary Policy”, Princeton University Press.
 For recent international evidence on the impact of globalisation on the labour share see A. Guscina (2006), “Effects of Globalization on Labor’s Share in National Income”, IMF Working Paper 06/294.
 See Ch. Bean (2006), “Globalisation and inflation”, speech delivered at the LSE Economics Society, London, 24 October; and R. Caballero, E. Fahri and P.O. Gourinchas (2006), “An equilibrium model of global imbalances with low interest rates”, BIS Working Papers, 222, December.
 For a model-based analysis of why central banks should respond to asset-price deviations from fundamentals, see B. Dupor (2005), “Stabilizing Non-Fundamental Asset Price Movements under Discretion and Limited Information”, Journal of Monetary Economics 52, pp. 727-747.
 See J.C. Trichet (2005), “Asset Price Bubbles and Monetary Policy”, speech delivered for the Mas Lecture, Singapore, 8 June.