RBA Annual Conference – 2001 Discussion
1. Gordon de Brouwer
There have been quite a few conferences and seminars on monetary policy in east Asia over the past few years and there still appears to be no end in sight. This is no bad thing because there is still a lot of unfinished business.
The financial crisis has had a profound effect on thinking about monetary policy and the frameworks, institutions, and markets needed to support it. There is deep dissatisfaction in east Asia with the way things have been run. In some countries, the aims and targets of policy have been too unclear, diffuse or inconsistent. In some countries, institutional structures – of the central bank, commercial banks, and prudential supervisor – and markets have been too weak. And across the board, there is deep concern about how to balance flexibility in the real exchange rate with exchange-rate instability and excess volatility.
Everyone recognises the need to deal with these issues comprehensively and consistently.[1] But political and national interests, institutional inertia, lack of human and financial resources, competing intellectual paradigms, and other factors make it a difficult and piecemeal process. There is also now more serious consideration of coordinated regional policy responses and institutions, which complicates things even further.
Bob's paper neatly highlights these complexities. Typical of Bob, it is a paper broad in coverage and rich in detail. Also typical of Bob, it is subtle and objective. In my comments, I will take up just a few of the items on his menu.
The objective of monetary policy
Policy-makers face a choice in terms of objectives, and of the tools and institutions, formal or otherwise, to realise them. But for all the objectives that we talk about, it is essential to keep the real aim in mind: maximising the economic welfare of the people. The perennial issue is how to do this. Invariably, to my mind, some flexibility in the policy framework and its application is essential.
For example, as much as price stability is important, inflation targeting should be seen as a flexible discipline on policy-makers, and should not be applied without regard to the variability of output and employment or the stability of the financial system. The danger is that when countries adopt inflation targeting, they become obsessed with proving themselves and hence become vulnerable to being too focused on keeping inflation on target in the short run. This is relevant to east Asia because as more countries in the region adopt some form of inflation targeting, they may feel compelled to over-deliver on their commitment to the target in order to establish their credibility (including among their central bank peers). This can prove unnecessarily costly in terms of lost output, as shown by the New Zealand experiment.
Similarly, central bank independence (CBI), transparency and accountability – the accoutrements of the modern central bank – are means to an end and never aims in themselves. They are not sacrosanct. The aim is constructive CBI, transparency and accountability. The criterion is a practical one: how, in the case at hand, do they improve well-being?
Providing more and more information may not improve policy-making. There is a perennial debate, for example, about whether central banks should release the minutes of their policy meetings. Does it improve policy-making? Providing the reasons for why a monetary-policy decision was made the way it was is important information for private decision-makers and helps keep the monetary authority accountable. But there is such a thing as too much information. For example, providing full details of discussions is not helpful if it makes participants reluctant to debate the issues for fear of looking stupid or being wrong.[2]
Making the central bank report to parliament may not increase accountability and improve policy-making if the parliamentarians are purely partisan or don't know anything about economics. Similarly making a central bank independent will generally help overcome the political bias in monetary policy but it can have downsides. For example, it may weaken the coordination of monetary, fiscal and wages policies. It can also be costly if the central bank does not have the necessary human resources or skill to make ‘good’ decisions or if it is more worried about inflation being above target than below. The gains (if any) from independence need to be compared to the costs (if any). The focus needs to be on substance not form.
More generally, this should be applied to all the policy proposals that we discuss here today: on balance, does a policy add to, or detract from, stability? And in so doing, we are more likely to beware of overstating and overgeneralising the benefits of the particular policies we advocate.
The exchange rate
Policy-makers and commentators in east Asia are deeply concerned about excessive exchange-rate variability. This is so for a number of reasons: concern that the exchange rate will overshoot if it is set by the market, especially in times of crisis (which is just when there is a premium on stabilising forces); concern that exchange rate movements can create uncertainty and, in the case of appreciations, hurt export competitiveness; and concern about emerging east Asia's exports remaining competitive in the United States, Japan and the European Union when there are big movements between the major currencies.
This is a rich field for analysis, and I will defer most of my comments to the general discussion on John Williamson's paper. But let me make two observations. First, volatility is not necessarily the issue. It is really how stabilising the exchange rate is to the shocks that hit the economy, and whether the interaction of the exchange rate system, policy-making and financial-market behaviours (like herding) create instabilities and shocks of their own. Volatility can arise for any number of reasons and is typically symptomatic of something else in the economy. It arises for ‘good’ or ‘bad’ reasons and its effect can be stabilising or otherwise.
Second, there is some concern that countries are reverting to implicit dollar pegging, which, given the diverse export dependence of most of non-Japan east Asia, is potentially costly if there are large swings in major currencies. My friend Eiji Ogawa (2000), for example, argues that correlations between daily movements in some regional currencies and the US dollar have strengthened since the crisis. If this is so, the argument for basket pegs is strengthened. But Bob questions this and I suspect he is right.
Figure 1 plots the baht, rupiah and won against the US dollar on a monthly basis for a two-year period before the crisis, from mid 1995 to mid 1997, and after the crisis, from 1999 to 2000.[3] The axes are scaled so that the percentage movement for each currency is the same in both periods. The won has been more stable against the dollar in the post-crisis period, but the baht and rupiah have certainly not been, making it difficult to accept as a general proposition that there has been a return to implicit dollar pegging in east Asia in the past few years. If the time period is restricted to 2000, then none of the three currencies appear stable against the dollar (relative to periods of pre-crisis implicit dollar pegging).
I would go even further and say that even if there were stability of regional currencies against the US dollar, this would not necessarily be a reliable indicator of implicit dollar targeting. The crisis-affected economies have successfully sought to build up their foreign exchange reserves in the past few years, largely in the form of dollars. In practice, they will be buying dollars when they are cheap – that is when their own currency is strong – and holding back their purchases when the dollar is expensive. This will tend to limit currency movement against the dollar but it is not right to characterise this as implicit dollar targeting.
It is also not clear which bilateral rate is being targeted. The yen/US dollar exchange rate has been relatively stable in the past few years, and it is not clear whether the crisis-affected economies are targeting the dollar or the yen. In Korea's case, in particular, there now appears to be a greater focus on bilateral stability with the yen than before the crisis.[4] Won/US dollar stability may simply be masking won/yen stability.
Capital controls
Concern about exchange-rate volatility has led a number of governments in east Asia to limit access by non-residents to domestic currency swap facilities as a means to control offshore speculation in their foreign exchange market. This is meant to head-off an excessive build-up of short positions.
As the experience of Singapore suggests, this can be an effective tool for reducing exchange-rate volatility. The Singapore on/offshore interest differential widened substantially when speculative activity in east Asia in 1997 and 1998 was most intense (Figure 2), indicating the success of Singapore's controls on local currency swap and forward markets.
I have no problem with Singapore's approach. Despite its vulnerability in the east Asian crisis, Singapore's financial prices moved substantially less than those of its neighbours, due not just to its good fundamentals but also its system of effective financial controls.[5] But it is not clear that its approach is as useful a tool in limiting exchange-rate volatility for that many other economies, especially developing ones.
In the first place, limiting offshore access to swap markets depends on two conditions being satisfied: there must not be a substantial offshore market which can provide an alternative source of funding, and the controls must be strictly and effectively enforced.
If a substantial offshore market in the local currency exists, then speculators can use it to fund their short speculative positions, bypassing the onshore market. The Malaysian authorities, for example, limited swap funding in the onshore market to non-residents in August 1997 but this did not stop speculative activity because a large offshore market existed in Singapore. For this reason, Malaysia revoked the convertibility of ringgit located offshore in September 1998, effectively destroying the offshore ringgit market.
The strategy of limiting swap access also needs to be strictly enforced by banks if it is to work. South Africa, for example, had swap limits in place for non-residents in 1998 but these were completely ineffective, not just because of the large offshore rand market but also because the laws were simply ignored by local banks and not enforced by the central bank. By way of contrast, the same distinction on swap funding exists in Malaysia and Singapore but banks in both countries strictly enforce the distinction because they deeply fear the penalties that their respective monetary authorities may impose in case of breach.[6] Frankly, few other central banks in the region have the credibility, means, or willingness to do so. (And in some cases, like Indonesia, imposing limits on offshore access to swap financing is not the answer since the onshore players, not the offshore players, are the ones doing the selling.)
Moreover, limiting the development of offshore markets is a two-edged sword. While it might reduce the vulnerability of a currency to speculative attack and contagion,[7] it stops firms and banks from passing foreign exchange and other risk to offshore parties who want that risk. This can be a serious disadvantage for countries which may want to borrow internationally in their own currency. This is not a problem for a rich city-state with excess savings like Singapore, but it is (or should be) for countries like Korea, Indonesia, and Thailand.
When banks are unable to pass on foreign exchange rate risk, they do one of two things. Either they have to bear that risk themselves, in which case they are more vulnerable to currency shocks and more likely to need bail-outs by the prudential and monetary authorities. Indonesia is an obvious case (Pangestu and Habir 2001). Or else they force local firms to bear the risk, in which case the corporate sector is more vulnerable to currency shocks. Ultimately, the taxpayer or consumer pays. It is up to policy-makers to decide how to deal with this trade-off, but we can't kid ourselves that it is not there.
Asian regionalism
In the past few years there has been a tectonic shift in east Asia toward developing regional policy responses and frameworks. It is most obvious in the formation of the ASEAN+3 group, the Chiang Mai Initiative, and the focus on bilateral trade arrangements. While there are certainly legitimate questions about particular elements of these initiatives, the process of regionalism is well and truly under way and is an important and welcome development.
Regional cooperation cannot help but have an effect on the moulding of national policies and institutions in the region. For example, the basket-peg proposal of John Williamson at this conference, and made by many others in the region,[8] involves changes to regional exchange-rate arrangements, which go to the heart of a country's monetary policy arrangements. There is also renewed focus on expanding regional financial support and policy-making, which, if it works, means sharing financial resources and information and letting others have a say in domestic policy formation.
These developments are a fundamental challenge to autonomous decision-making and action driven by narrowly defined national interest. It is certainly within the scope of east Asia to make this shift but the challenge that it poses should not be underestimated. There is plenty of fine rhetoric about regionalism but substantive action is much harder to come by.
In this, there is no need for the region to simply copy what other countries and regions have done. The European experience, for example, can inform east Asia but it is not the only model and it is not necessarily the one that most enhances the economic and social well-being of people in the region. It would be useful during this conference to highlight practical ways that regional developments could be used to build up and reinforce effective and welfare-improving national institutions and policies in east Asia. The strengthening of regional policy discussion is one way to do this. Indeed, the strengthening of regional policy dialogue and institutions is an important goal in its own right.
In closing, I would like to congratulate and thank Bob for his comprehensive, deep and useful analysis.
Footnotes
See Nellor (2001) for a discussion. [1]
This is how a long-standing insider to the FOMC described to me the effect of the Gonzales Bill, which requires the release of detailed minutes from FOMC meetings after five years. This person said that meetings have become more rigid and formal as a result, with FOMC members more inclined to make prepared presentations and less willing to enter discussion. Some of the discussion shifts away from the formal meeting to informal chats, effectively boosting the control and power of the Chairman of the Fed and subverting the intention of the legislature. [2]
See de Brouwer (2001a). [3]
See Wang, Kim and Ryou (2000) and de Brouwer (2001a). [4]
See de Brouwer (2001b). [5]
See the report of the FSF Working Group on HLIs (2000). [6]
See Rankin (1999). [7]
See Ito, Ogawa and Sasaki (1998), Ogawa and Ito (2000), Dornbusch and Park (1999), Murase (2000), Kawai and Akiyama (2000), Kawai and Takagi (2000) and Yoshino, Koji and Suzuki (2000). On a regional currency unit, see Moon, Rhee and Yoon (2000) and Moon and Rhee (forthcoming). [8]
References
de Brouwer GJ (2001a), ‘Debating Financial Markets and Policies in East Asia’, in GJ de Brouwer (ed), Financial Markets and Policies in East Asia, Routledge, London, pp 286–314.
de Brouwer GJ (2001b), Hedge Funds in Emerging Markets, Cambridge University Press, Cambridge.
Dornbusch R and YC Park (1999), ‘Flexibility or Nominal Anchors?’, in S Collignon, J Pisani-Ferry and YC Park (eds), Exchange Rate Policies in Emerging Asian Countries, Routledge, London, pp 3–34.
FSF Working Group on Highly Leveraged Institutions (HLIs) (2000), Report, available at <http://www.fsforum.org/Reports/RepHL1.html>.
Ito T, E Ogawa and YN Sasaki (1998), ‘How Did the Dollar Peg Fail in Asia?’, NBER Working Paper No 6729.
Kawai M and S Akiyama (2000), ‘Implications of the Currency Crisis for Exchange Rate Arrangements in Emerging East Asia’, World Bank Policy Research Working Paper No 2502.
Kawai M and S Takagi (2000), ‘The Strategy for a Regional Exchange Rate Arrangement in Post-Crisis East Asia: Analysis, Review and Proposal’, World Bank Policy Research Working Paper No 2503.
Moon WS and YS Rhee (forthcoming), ‘Foreign Exchange Market Liberalization Policies in Korea: Past Assessment and Future Options’, in KT Lee (ed), Globalization in the New Millenium, Routledge, London.
Moon WS, YS Rhee and DR Yoon (2000), ‘Asian Monetary Cooperation: A Search for Regional Monetary Stability in the Post Euro and the Post Asian Crisis Era’, Bank of Korea Economic Papers, 3(1), pp 159–193.
Murase T (2000), Ajia Antei Tsuukaken: Yuroo ni Manabu Yen no Yakuwari (The Asian Zone of Monetary Stability: Lessons from the Euro and Role of the Yen), Keiso Shobo, Tokyo.
Nellor D (2001), ‘Securing Consistency in Macroeconomic Policy’, in GJ de Brouwer (ed), Financial Markets and Policies in East Asia, Routledge, London, pp 169–199.
Ogawa E (2000), ‘East Asian Countries Return to Dollar Peg Again?’, Hitotsubashi University, mimeo.
Ogawa E and T Ito (2000), ‘On the Desirability of a Regional Basket Currency Arrangement’, NBER Working Paper No 8002.
Pangestu M and M Habir (2001), ‘The Boom, Bust and Restructuring of Indonesian Banks’, in GJ de Brouwer (ed), Financial Markets and Policies in East Asia, Routledge, London, pp 122–156.
Rankin RW (1999), ‘The Impact of Hedge Funds on Financial Markets: Lessons from the Experience of Australia’, in D Gruen and L Gower (eds), Capital Flows and the International Financial System, Proceedings of a Conference, Reserve Bank of Australia, Sydney, pp 151–163.
Wang Y, TJ Kim and JW Ryou (2000), ‘Regional Arrangements to Borrow: A Scheme for Preventing Future Asian Liquidity Crises’, Korea Institute for International Economic Policy (KIEP) Policy Analyses No 00-01.
Yoshino N, S Koji and A Suzuki (2000), ‘Basket Peg, Dollar Peg and Floating: A Comparative Analysis of Exchange Rate Regimes’, Keio University, mimeo.
2. General Discussion
While covering many areas the main issue of discussion was the relationship between interest-rate volatility, exchange-rate volatility and openness portrayed in Figures 1 and 2 (and Appendix B) of Bob McCauley's paper. Some participants supported the graphs as highlighting a real relationship while others felt that there were problems with the construction of the measures and interpretation of the relationship. One pointed out that Japan, with zero interest rates, would confound the volatility measure. Others suggested that using the short interest rate could also be misleading. While monetary policy operates through the short rate it is the long rate that is of more relevance for the economy. Thus, the interest-rate volatility measure might be more illuminating if it made use of a long interest rate rather than a cash rate. Other participants indicated a desire to see some formal econometric testing of the proposed negative relationship between volatility and openness. (Which McCauley has included in the revised version of the paper in this volume.)
Some participants questioned whether it was even sensible to talk about a trade-off between foreign-exchange volatility and interest-rate volatility. It was suggested that the relationship really revealed the nature of shocks hitting the various countries while the reduced form evidence of the graphs could not determine whether it was a conscious choice or a difference in the shocks hitting each country that led to the observed variation.
On exchange-rate regimes, it was argued that the forces that pushed countries towards floating exchange rates – huge capital flows – were no longer present in many countries. In this case countries could engage in some management without triggering massive speculative attacks. The point was made that the evidence in the paper does not rule out the possibility that countries are now managing their exchange rates more than they used to. In a related point, one participant noted that it is difficult to evaluate the performance of Asian floating rate regimes after so few years. While Australia is clearly regarded as a freely floating country, it took almost ten years for operating procedures to evolve to their current form. This participant argued that during that period it would have been possible to question the exact nature of the Australian regime – for example, the extent of management taking place.
Some participants pursued the question of how much freedom countries had to choose their exchange-rate regime. It was felt that many countries had the regime forced upon them. For example, because of massive capital flows, some countries were forced to float; in this situation an obvious nominal anchor was an inflation target. Similarly, participants referred to the experience of South American countries where some countries may have been forced to fix their exchange rate to combat inflation by importing the monetary credibility of some other country, typically the US.
One participant questioned whether the measures of central bank independence were still relevant. The Cukierman index had not been updated beyond 1989 while much has changed in the central banking world since then. Concern was also expressed at how useful these indices were in practice. Finally, a participant was interested in whether capital controls could be effective in the long run. He suggested that, while controls can work for a while, they are eventually subverted as people find ways around them, and hence would not be effective in the long run.