Republic of Korea -- Concluding Statement of the 2004 IMF Article IV Consultation Mission

October 28, 2004

Describes the preliminary findings of IMF staff at the conclusion of certain missions (official staff visits, in most cases to member countries). Missions are undertaken as part of regular (usually annual) consultations under Article IV of the IMF's Articles of Agreement, in the context of a request to use IMF resources (borrow from the IMF), as part of discussions of staff monitored programs, and as part of other staff reviews of economic developments.


After a long post-crisis boom, Korea's economy has begun to sputter.  Growth suddenly stopped in the first half of 2003, spurted ahead in the second half of the year, but then slowed to an annualized rate of just 2 3/4 percent in the first half of 2004. What explains this pattern of stop-and-go growth? And what can be done to put the economy back on the path of sustained, rapid expansion? These are the key policy questions facing Korea at this current juncture. Consequently, these are the questions the IMF mission has tried to address.

I. What is the problem?

1. To shed some light on the problem, it is useful to examine the composition of the economy's recent growth. When this is done, it can be seen that the sharp acceleration in the second half of 2003 was attributable to a spurt in exports, as the pace of global economic activity quickened, especially in the information technology industry. Then, as the world economy decelerated to a more sustainable rate earlier this year, exports began to slow and Korean growth subsided again.

2. Meanwhile,  throughout the past two years, domestic demand has been stagnating. In large part, this reflects the after-effects of the 2001-2002 credit boom. Just as the rapid expansion of credit to households and SMEs boosted growth in the earlier period, growth is now being depressed by the subsequent credit "bust". Households are now trying to pare back their debts from the exceptionally high level of 62 percent of GDP reached during the boom years, toward a more normal pre-crisis level of 45 percent of GDP. (This is not to say that household debt will need to fall to 45 percent of GDP. Most likely, the equilibrium level of debt has increased, since households have much greater access to bank credit than they did before 1998. Still, experience in other countries which have undergone credit boom-and-busts suggests that debt will need to be reduced significantly from the current level.) To do this, they are increasing their saving and reducing their consumption. SMEs, too, are struggling with their own heavy debt burdens, making it difficult for them to invest. Meanwhile, larger corporations are flush with cash, but are reluctant to invest. Indeed, overall sentiment is weak, much weaker than can be explained by the actual state of the economy.

3. Reviving the economy will consequently require restoring confidence and restarting the domestic engines of growth—spending by households, SMEs, and larger corporations. To a certain extent, after two years of adjustment, these engines are now ready to restart on their own. Experience from other countries that have experienced credit boom-and-busts suggests that the increase in household savings rates should begin to taper off in 2005, allowing consumption to recover. Moreover, the broader economic environment is also conducive to recovery, since large corporations are highly profitable, the banks are financially sound, and considerable progress has been made since the Asian crisis in establishing a robust market-based framework for economic activity.

4. For these reasons, the mission is convinced that the economy is headed for better times. The precise trajectory of the recovery is difficult to forecast, and we will revisit our forecasts once the third quarter GDP numbers are released and more details on the size and timing of the construction stimulus package are available. Still, the overall direction seems clear: economic growth is likely to accelerate over the course of 2005.  

5. How can the government ensure that this economic recovery is strong and sustainable? After extensive discussions with officials, the private sector, and academics, the IMF mission has come up with a number of recommendations. Some of these can be implemented in the short term. Others will take some time to realize fully, but could at least  be initiated quickly, providing an immediate boost to confidence. Strikingly, many of these measures are already part of the government's policy program. And it is this fact that government policies are moving in the right direction that makes us fundamentally optimistic about Korea's economy.

II. What needs to be done?

A. Providing A Macroeconomic Spark

6. The first step toward reviving the domestic engines of growth would be to provide a macroeconomic spark, by implementing stimulative fiscal and monetary policies. The 2005 budget submitted to the National Assembly contains some welcome measures that could help provide such a spark. In particular, the draft budget calls for reductions in a variety of taxes, including personal income taxes, which could quickly raise the disposable incomes of households, allowing them to increase their spending. There are also welcome plans to front-load fiscal spending. Nevertheless, the budget as a whole is essentially economically neutral, since the deficit is targeted to remain at about the same level as this year. (Since the output gap is difficult to estimate, calculations of fiscal stimulus are necessarily subject to a range of uncertainty.)

7. This is why the mission was pleased to learn of the plans to supplement the budgeted spending with an additional program to support the construction sector.  The mission understands that this plan would involve new spending of perhaps as much as 1 percent of GDP, targeted on infrastructure projects such as expanding the road network, renovating schools, and developing new cities. Interestingly, much of these plans would involve private sector participation or funding from the National Pension Fund and other non-budget sources.

8. This is a promising approach, but it poses several challenges, in the mission's view. First, it will be a challenge to make these arrangements as commercial as possible, ensuring on the one hand that entities providing the financing (such as the National Pension Fund) get an adequate cash rate of return, while simultaneously ensuring that the government does not provide them with excessive guarantees. Second, it would be important to ensure fiscal transparency: to the extent there are public sector outlays, these should show up in the consolidated central government accounts. Third, the projects may take time to materialize, especially as complex agreements will need to be negotiated among the various interested parties. For this reason, it might still be necessary to provide stimulus, as in previous years, by adopting a supplementary budget.

9. In the meantime, the task of macroeconomic management will fall mainly on  monetary policy. Of course, the primary task of monetary policy is to preserve price stability. And it is a concern that headline inflation is well above comfortable levels; indeed, based on this measure, the Bank of Korea's target interest rate is now negative in real terms. But monetary policy is not based on current levels of inflation. Rather, it is forward looking, geared to affecting the level of inflation six to eight quarters ahead, the time it takes for monetary policy to have its intended effect.

10. Viewed over this longer horizon, inflation is less of a concern. The increases in food, oil, and other import prices do not seem to be feeding through into more generalized inflationary pressures. To the contrary, core inflation has remained within the target range, wage growth has been slowing, while longer-term interest rates have fallen to unusually low levels, suggesting that inflationary expectations remain firmly under control. Housing prices have stabilized, and thus should not be a constraint on monetary policy. Finally, in the mission's estimation, the slow growth of the economy has kept output below its potential level, which will put downward pressure on inflation over the months ahead. For all these reasons, the mission believes that there is scope for further easing in monetary policy, to provide support to the real economy.

11. With respect to the exchange rate, the IMF strongly supports the official policy of allowing the won's external value to be determined in the market, with intervention limited to smoothing operations.

B. Reviving Household Spending

12. While macroeconomic stimulus can provide a spark for the economy,  achieving a self-sustaining recovery will require addressing the underlying structural difficulties. This will require, first and foremost, addressing the household delinquency problem. Evidence from around the world suggests that the speed of recovery following the collapse of a household credit boom can vary significantly. In some cases, recovery has been swift and sharp; in others, it has been slow and weak. What this means is that much depends on policies: well-designed plans can promote recovery, while indecisive measures can prolong it.

13. The mission strongly supports the government's basic policy approach toward debt delinquencies. We commend the decision to stand firm against a mandated debt amnesty scheme, which would only generate moral hazard, make financial institutions more reluctant to lend to households in the future, and trigger calls for government bail-outs in other sectors. More broadly, by standing firm in this area, the government has sent a strong signal of its determination to reduce its intervention in the economy.

14. Rather than becoming directly involved, government efforts have appropriately focused on establishing a framework within which debtors and creditors can work to resolve their problems. Already, a number of programs have been created, including the Credit Counseling and Recovery Service, the Bad Bank, and the Personal Debt Rehabilitation Program. All of these have proved helpful in advancing the resolution process. Even so, only about 15 percent of delinquencies have been resolved to date.

15. One problem is that despite all the workout schemes there is a gap in the current framework. All of the existing schemes are tailored to the better-off borrowers, those whose debt burdens are relatively light and whose incomes are relatively secure. They leave out perhaps the largest group of delinquents: those who have little chance of repaying their debts in full, and whose incomes are, at best, difficult to verify. In other countries, this group would seek refuge in bankruptcy courts. But such a course is less likely in Korea, partly because of the large stigma attached and partly because the terms of bankruptcy are relatively onerous, with debtors having to live on minimum incomes for 8 years, compared to 3-5 years in the United States.

16. How can this group's needs be met? One possible way might be to modify the design of the Personal Debt Rehabilitation Program. Applications under this program have been very modest so far, partly because it is new, but mainly because it is perceived as complex, costly, with very high documentation requirements. Currently, plans are underway to simplify the program; we would propose more generous asset exemptions and shortening the repayment period from 8 to 5 years. In addition, we would suggest looking at Hong Kong's Individual Voluntary Arrangements (IVAs), as an example of how procedures can be further streamlined. IVAs are informal agreements that resolve disputes quickly, at minimum cost to the parties involved. Under this program, debtors and creditors submit their debt restructuring proposals to arbitrators, usually accountants, who decide what plans would be appropriate. Then, to give these plans legal status and afford creditors some protection against any relapses into delinquency, they are ratified by the courts. Thus, those debtors who subsequently default could be sent into bankruptcy for a final resolution.

17. At the same time as efforts are made to accelerate debt resolution, it would be important to lay the groundwork for households, to regain access to financial credit. For without such access, households will be forced to continue to build up their savings, since this would be the only way to ensure they could handle unforeseen expenses. In such circumstances, it would be difficult for consumption to revive.

18. But how can banks be convinced to lend again to households, after the severe  problems encountered by their credit card affiliates? The answer lies in building up positive credit bureaus. If banks had comprehensive information about their borrowers, not just about their delinquencies but about their positive repayment records as well, they would be able to make more sophisticated calculations than simple yes/no lending decisions. They could begin to create credit scores for individuals, as is done in other countries. Lending could then be tailored to those scores, enabling those with lower ratings to still get access to credit, as long as they were willing to pay higher interest rates.

19. Establishing such a system will take time. But precisely for that reason it is important to start quickly. In this effort, a key objective will be to make sure that credit assessment by lenders is as efficient as possible. Currently, financial institutions do not widely share positive credit information (that is, information about loans that have been serviced on time, as opposed to negative information about delinquencies) with credit bureaus, making it very difficult for lenders to obtain a complete picture of their loan applicants' credit standing. Ideally, this information should be made readily available to, and processed by, the credit bureaus, which would then provide comprehensive reports instantly, whenever financial institutions need them. This is what is done in other advanced countries.

20. Building such an efficient system can be done with only two basic regulations, as Mexico discovered. First, financial institutions submitting information to credit bureaus could be required to submit the same information, at the same price, to other licensed credit bureaus. Second, financial institutions could be encouraged to obtain credit reports, for example by requiring them to provision more heavily on loans for which reports were not obtained. Of course, in return for obtaining these reports, the institutions would need to provide the information that the credit bureaus need. Consequently, with these two rules, credit bureaus could have comprehensive information, and could compete on the basis of their ability to process this information into the formats that creditors need. But the rules would need to be put in place soon: it is important to establish the framework before the new credit bureau, currently being set up by several large banks and credit card companies, is created—rather than try to change the system after such an investment is made.

21. Beyond restoring households' access to credit, it will also be important to address the problems on the asset side of their balance sheets. This will involve housing policy. The government's measures of October 2003 have succeeded in stabilizing housing prices. But they have also effectively "frozen" the housing market, with the number of transactions dropping sharply, especially in zones designated as speculative.  As a result, the main asset of households has been rendered less liquid, giving them further reason to build up their precautionary savings, rather than consume. One way to address this problem is to rebalance the tax burden on property by reducing transactions taxes (including the capital gains tax) at the time when the new wealth tax on property is introduced in 2005. In this way, the costs of selling would be reduced, restoring liquidity to the housing market, while taxes on property holdings are increased, thereby reducing demand for such assets and discouraging speculation.

C. Revitalizing the Small and Medium Enterprises

22. In addition to addressing the difficulties of households, Korea also needs to resolve a growing problem in the SME sector. Around one fifth of manufacturing SMEs are now incurring losses, with more than one third failing to generate enough operating income to cover their interest expenses. Moreover, these figures likely understate the extent of the sector's problems because the bulk of SMEs are in services, where firms' financial difficulties are worse. Service sector firms are now facing heavy debt burdens, since they received the bulk of the SME credit during the boom period.

23. To a certain extent, the problems in the sector are cyclical. Many firms have been hit hard by the slowdown in domestic demand, and as demand recovers, so too will their financial position. But the problems in this sector go beyond mere cyclical difficulties: there are profound structural problems, as well. While the profitability of SMEs improved in the early 1990s, it has been declining steadily since the 1998 crisis, in marked contrast to the experience of large firms. Part of the reason is that SMEs have made much less progress than large firms in operational restructuring, as can be seen from their lagging productivity growth.

24. In contrast, consider the situation in Taiwan POC. That country has an SME sector similar in size and importance, equally (if not more so) exposed to intense competition from China. Yet its enterprises are significantly more profitable, with net income in recent years around 2 percentage points higher than their Korean counterparts. Moreover, Taiwan's SME sector is very dynamic, with high entry and exit rates. In other words, Taiwanese entrepreneurs are quick to create new firms when opportunities arise, and quick to close them when market conditions change.

25. What explains this difference? A key factor is government loan guarantees. In Taiwan, such guarantees are relatively small, amounting to only 1 percent of GDP, one fourth the level in Korea. As a result, banks cannot rely on them to guide their lending: they instead need to do the difficult work of evaluating credit and taking risks. By contrast, in Korea, resources flow to firms that can secure guarantees, which are overwhelmingly well-established firms. Moreover, even though most guarantees are given for only one year, the average length of KCGF guarantees is around five years, since they tend to be rolled over again and again. The effect is to limit the dynamism of the SME sector, preserving existing firms making traditional products, even as times and tastes change.

26. The government has already recognized this problem, and has formulated a plan to deal with it. Under this plan, guarantees would be directed away from well-established firms toward start-ups and new technology companies. In addition, a government-run credit bureau for SMEs will be established, to provide banks with the information they need to undertake risk-based—rather than guarantee-based—lending.

27. The mission welcomes these steps, and recommends that the government go further in this direction. For example, to encourage banks to do more credit analysis, the proportion of a loan covered by a guarantee could gradually be reduced from the current average of around 85 percent to around 50 percent. And to limit the risk of firms "monopolizing" guarantees, the fee structure could be altered to impose higher rates on firms rolling over existing loans. Fundamentally, though, what is needed is to wind back the levels of the guarantees themselves. This reduction could be gradual, perhaps by 1 percent of GDP per year for the next five years. It could also be preannounced, so that banks and enterprises can make plans to adjust. But it should also start now, because postponing the necessary task of restructuring will only increase the eventual costs.

28. The point can be put in a different way. Right now, SMEs are not investing because they are not in a financial position to do so. Reviving investment therefore requires creditors to force them to accelerate—or begin—the process of restructuring.

29. One possible objection to such a plan is that the mere announcement of changes to the loan guarantee program would trigger a credit crunch, making the situation worse in the short run. In the mission's view, however, the risks of such a reaction would be small. To begin with, the unwinding of guarantees would only be gradual. Moreover, even if banks feared that current guarantees would not be rolled over in the future, they could still remain reasonably confident in their current lending, since even in the worst case they would still be reimbursed 85 percent of any loan that is able to secure a guarantee today.

30. At the same time, measures could be taken to provide direct help to SMEs, especially start-ups.  One way to do this would be to relax the rules on venture capital firms, to allow them to acquire majority voting rights or take over management, when appropriate. These practices are quite standard in the United States. Indeed, since start-up entrepreneurs are typically long on ideas and short on practical experience, many young firms in the US are attracted to venture capital firms precisely because they can provide needed management expertise. At the same time, to make it easier for SMEs to borrow on the basis of their receivables and other non-property collateral, the Civil Code could be amended to allow for the registration of property rights and security interests on a wider range of assets.

31. A further useful step would be to reduce the costs of starting businesses. The minimum capital requirement for incorporating a business, at W30 million or more than twice per capita income, is relatively large compared with other OECD countries, where costs average less than half the per capita income. Similarly, the large number of land use laws—the OECD counted 112 of them—create complexity and uncertainty for investors; perhaps a public information service or even a "one-stop shop" for approvals could be established to guide investors through this maze.

32. Finally, to provide a more efficient way for SMEs to reorganize or exit, it would be useful to finally push through the unified insolvency bill.  This bill, which has been languishing in the National Assembly, would introduce a debtor-in-possession system for firms' assets, thereby allowing management to continue running companies in bankruptcy, as long as this has creditor approval.

D. Unleashing Corporate Investment

33.  In addition to reviving the household and SME sectors, Korea also needs to unleash corporate investment. As in the other sectors, solving this problem will require fundamental reform. For the decline in investment is not just cyclical: facility investment has been declining for more than a decade, falling from a peak of 15 percent in 1992 to a projected 9 percent of GDP this year. In part, this decline has been a natural consequence of moving from a system that encouraged over-investment toward one that places more emphasis on profits. But the main reason why so few investments are now profitable in Korea is that structural problems are raising the cost of capital, and the projects themselves. These problems including poor corporate governance and a relatively rigid labor market have become particularly pressing in recent years, as Korea's economy has begun to integrate with that of China, forcing firms to consider carefully what production should take place domestically and what should be out sourced.

34. There are some who argue that the government's efforts to improve corporate governance are not a solution, but a source of the problem. They claim, in particular, that the FTC's plans to tighten restrictions on the voting rights of chaebol's financial affiliates will leave industrial firms open to hostile takeovers, thereby discouraging investment. They further argue that investment is also discouraged by the ceiling on total shareholdings of other domestic companies.

35. The mission is not convinced by these arguments. To begin with, as noted above, the reluctance to invest is not new: facility investment has been falling for some time, even before the FTC's recent proposals were made. Moreover, the risk of hostile take-overs is actually quite small. For example, foreign portfolio investors rarely initiate take-overs. In fact, hostile take-overs of any kind are rare outside the United States, and for a good reason: unless the new owners can secure the loyalty of mid-level management and workers, it will be impossible for them to run taken-over firms successfully. Indeed, there are cases in Asia of owners who purchased firms after the 1998 crisis, only to return them after they have found they could not manage them (e.g., Bank Bali in Indonesia). As for the risks arising from the ceiling on total shareholdings of other domestic companies, these too are minimal, since there are numerous exemptions in the current regulations, to allow for new joint ventures and other worthwhile projects.

36. In the mission's view, better corporate governance is much more likely to promote investment than it is to hinder it. This is because improvements in governance will reduce the "Korea discount" on equity, thereby lowering the cost of raising capital. For this reason, the mission strongly supports the FTC bill currently before the National Assembly, which aims to tighten regulations on chaebol with poor corporate governance, while exempting those that bring their governance up to adequate levels.

37. Indeed, the government could go further by encouraging the development of more market-based mechanisms for enforcing corporate discipline. The introduction of class action lawsuits for securities violations is a welcome step in this direction. In addition, last year's mission recommended that the government encourage the Korea Stock Exchange to adopt a "comply or explain" regulation, which would set out a short list of good governance practices and ask listed companies either to comply with each individual measure or explain why they haven't done so.

38. At the same time, it will be important to remove the most important impediment to investment: the rigidities in Korea's labor market. Over the past year, the government's stance of refraining from intervention in disputes, apart from enforcing labor laws, has had remarkable success. Strikes, though still numerous, are being settled relatively quickly, with little disruption to the overall economy. Wage settlements have decelerated, as both workers and management have taken into account the weakness of the economy in their negotiations.

39. Now, the government needs to take the next step, of addressing the fundamental labor market problem: the high levels of employment protection for regular workers. With the legal criteria for dismissal being difficult to satisfy, firms—especially large firms in unionized sectors—have been forced to offer very high voluntary severance payments to those workers they wish to dismiss. For example, the severance pay for financial service workers is twice as high as in other financial centers, 22 months of salary for employees who have worked for 15 years, compared to 11 months in Hong Kong and 12 months in London.

40. Such high severance payments reduce the prospects for turning Korea into a regional hub. More generally, such inflexibility is incompatible with Korea's growing role as a manufacturer of advanced products and a provider of high technology services. These functions require a high degree of labor flexibility, so that firms can adapt to rapidly changing opportunities and technologies. As long as costs of adjusting the labor force remain high, companies will continue to be deterred from making risky investments.

41. High severance payments have led to an additional significant problem: the growing number of non-regular workers.  This has created a dualistic labor market, a trend which is not sustainable socially or even economically, because it generates considerable uncertainty for non-regular workers, discouraging them from consuming and leading them to build up their precautionary saving.

42. But how can the dual labor markets be unified? It is tempting to think that unification can be achieved by raising the pay and working conditions of non-regular workers up to the level of their regular counterparts. Tempting, but almost certainly wrong. During the 1990s, Spain tried this very approach and found that it did not work: employers responded by reducing their hiring of all workers, and unemployment increased. To reduce the risk of a similar outcome in Korea, the bill to prohibit discrimination against non-regular workers needs to be designed carefully, to make it clear to the courts that the government does not intend to require equal pay or working conditions.

43. In the mission's view, the only real way to unify the labor market is by reducing the protection given to regular workers. During the previous Article IV consultation, the mission proposed such a strategy, the one eventually followed by Spain. Essentially, the country "grandfathered" existing contracts for workers who had already secured regular employment. Meanwhile, those applying for jobs were offered a new contract, with less protection than that enjoyed by the grandfathered regular workers, but considerably more than that of non-regular workers. This contract has proved enormously successful in generating jobs, helping to unify the labor market, and promoting flexibility. For this reason, the mission is pleased to learn that the Tripartite Commission is now considering this example, among others, to see if it can be applied to Korea. The mission also welcomes the plans to lay the groundwork for increased labor market flexibility by expanding the social safety net, especially eligibility for unemployment insurance and basic livelihood protection.

III. Conclusion

44. In sum, the mission remains optimistic about the Korean economy. In our view, a recovery is likely to take hold in 2005, with domestic demand accelerating as the year progresses. The real question is how strong this recovery is likely to be. Past history gives some hope in this regard: Korean recoveries have tended to be sharp and swift.

45. But past history is not always a guide to future performance. A far better assurance can be provided by implementing timely policies to address the underlying difficulties. In the current circumstances, and in the mission's view, such policies would include providing macroeconomic stimulus, accelerating the resolution of household delinquencies, restructuring the SME sector, and modernizing the labor market.

46. There are those who fear that undertaking structural reforms at a time of economic weakness could aggravate the downturn. The mission takes a different view. In fact, international experience demonstrates the precise opposite: it is only by offering and implementing convincing plans for addressing the structural difficulties that governments can restore confidence. This is especially important in Korea, for it is confidence that is the missing, vital, ingredient for the country's economic recovery.

47. In fact, the government already has well-designed plans for addressing the country's underlying problems. It is in the process of implementing them. And this is why we remain optimistic about Korea's economic outlook.


Public Affairs    Media Relations
E-mail: E-mail:
Fax: 202-623-6278 Phone: 202-623-7100