Journal Issue

Recentralization in China

International Monetary Fund. External Relations Dept.
Published Date:
March 2003
  • ShareShare
Show Summary Details

Interview with Ehtisham Ahmad: Recentralization in China?

IMF Survey: The title of your working paper—“Recentralization in China?”—suggests several interpretations. Should it, can it, will it recentralize? Why is China weighing such a move?

Ahmad: Our title is a bit provocative. People tend to think of China as centrally planned with a monolithic administration. In fact, until recently, China has never really had a central tax administration. The center has traditionally relied on revenues transferred up from the provinces.

In the late 1970s and early 1980s, Deng Xiaoping initiated a series of reforms, including in the way provinces used retained revenues. These reforms initially gave local governments a lot of flexibility to invest resources, and some researchers credit this, in part, with spurring China’s phenomenal growth. In aggregate terms, though, China saw a precipitous drop in the overall ratio of tax revenues to GDP from the early 1980s until the mid-1990s. Local governments had incentives to hang on to the money they collected. But the center realized that this severely compromised its ability to redistribute revenue among provinces. It also recognized that it had to develop levers to control the macroeconomic situation. So, in 1994, the country undertook major reforms that, for the first time, established a central tax administration. The reforms dramatically raised the center’s share of revenues—to around 55 percent from 35 percent—and reversed the decline in the overall revenue-GDP ratio.

With its entry into the WTO, China will have to take an urgent look at its productionbased VAT, which does not provide credits for the taxation of capital goods.—Ehtisham Ahmad

IMF Survey: But your paper also suggests that the 1994 reforms left a number of issues unsettled.

Ahmad: The 1994 reforms did not go far enough. They helped stem the decline in total tax revenues but were essentially incremental. As part of a compromise, the center agreed that a significant, though declining, portion of the revenues would be returned to the provinces that generated them, and this solution left very little scope for real redistribution.

Also, significant expenditure responsibilities remained at the local levels. Traditionally, state-owned enterprises provided cradle-to-grave benefits, but as China modernized, responsibility for many of these services shifted to local governments. Most of China’s educational and health care services, for example, are now delivered at the city and county levels, but the local governments have very little flexibility to raise additional revenues legally. They do not have the authority, for example, to increase tax rates within a specified range.

In addition, local governments—in most cases, cities and counties—are responsible for providing pensions and unemployment insurance. China does not have a national social security system, and there are significant risks in having a county try to provide unemployment insurance—especially if it has a concentration of enterprises that need to be restructured.

It makes greater economic sense for pension and unemployment insurance needs to be pooled at a wider level—at least at the provincial level. But if China is going to redistribute among provinces, it will also have to reassess appropriate expenditure responsibilities for the provinces. The coastal provinces in particular will want to keep their revenues if they retain responsibility for pensions.

IMF Survey: Does the argument for centralizing pensions and unemployment insurance hold for the delivery of health care and education?

Ahmad: Decentralized service delivery is appropriate, but the problem is ensuring an adequate level and quality of services when local governments lack the ability to collect needed additional revenues. Some governments have attempted to impose illegal nuisance fees and charges. These have led to considerable unrest in some rural areas and have spurred central government attempts to reform local taxation.

IMF Survey: In terms of tax reform, what do you think the government’s approach should be?

Ahmad: Clearly, major decisions would be involved in reforming China’s tax structure. The incoming administration will need time to think through these issues. Individual tax reforms would face difficulties because they would have a differential impact on the provinces. A more ambitious and well-sequenced reform may be needed to strengthen the country’s investment climate—something that is critical to providing for the welfare of 1.3 billion people over the medium term but also to ensuring sufficient resources for redistribution, meeting the minimum needs of pensioners, and offering provinces reasonably similar access to education and health care.

In 1994, the Chinese authorities did a very good job of reforming the revenue-sharing and transfer systems, but the center did not have the fiscal space to make regional redistribution work effectively. If China is to gain that fiscal space, it may need a more ambitious reform.

IMF Survey: Where does China start?

Ahmad: The key early decisions will be on the tax side. With its entry into the WTO [World Trade Organization], China will have to take an urgent look at its production-based VAT [value-added tax], which does not provide credits for the taxation of capital goods. This type of VAT places domestic producers at a disadvantage as the country complies with WTO requirements and removes tariff and nontariff barriers on competing goods. Moving to a consumption-based VAT is therefore a critical reform, but it will entail revenue losses in aggregate, with uneven revenue implications for the provinces.

China can opt to extend the VAT to cover services. This would raise revenue in the aggregate, although some provinces would gain and some lose as the business tax is removed. The increases in central revenue, however, should be sufficient to compensate the losers.

IMF Survey: Your paper also cites the need to rethink personal income and property taxes.

AHMAD: The 1994 reforms assigned the personal income tax to the provinces. The authorities have now recognized that the personal income tax is a potential major source of revenue, given rising incomes. In fact, since the first version of this paper was presented at the IMF’s Conference on Fiscal Decentralization in 2000, the Chinese authorities have decided to share revenue from the personal income tax with the local governments.

However, sharing does not provide the provinces with any control, at the margin, over tax rates. Control is crucial if local governments are to become more accountable for the services they provide. It might be possible in the short term to allow the provinces to piggyback on the central tax or have some control over a portion of the total amounts. That would make them less dependent on either illegal fees or transfers from the center, which may or may not rise, and they would not find themselves with unfunded mandates that might lead to arrears or indirect borrowing.

Similarly, property taxes levied on a certain percentage of a properly assessed lease value could raise fairly substantial revenues, particularly for the large cities in the coastal region, and thus reduce the demand for revenue returns.

Such measures could serve a dual purpose—provide a little more fiscal space for the center to redistribute revenues to needy regions and individuals, and give local governments an incentive to be accountable for service delivery. This is the essence of effective decentralization.

Of course, accountability will also mean strengthening budget and treasury systems. China, like many transition economies, currently has a big program under way to set up a treasury system. That will take time, but the need is clearly recognized. In fact, since 1999, there has been a major shifting of gears. Chinese policymakers want assurances that public moneys are being spent for their intended purposes and that revenues collected are actually going into the treasury. This has been a major thrust of the outgoing administration. Zhu Rongji had been very tough on embezzling and cheating, but China has not had the instruments—the budget and treasury systems—to track the funds to spending agencies. This is beginning to be done through an array of well-thought-through and focused reforms. The bases for greater transparency and better governance are being laid now.

IMF Survey: And how does China begin to establish a national social security system?

Ahmad: China will have to make the link between social security reform and all the other intergovernmental reforms. Pooling pension funds at the provincial level is a start, but one that does not address the liabilities that Shanghai and the other rich provinces face. Many coastal provinces already have aging populations, and some provinces are now running arrears. A lot of work will need to be done, but some long-term efforts are already under way. For example, China has taken some of the proceeds from the sales of state-owned enterprises and reinvested them for future social security payments.

The authorities will still have to tackle structural reforms relating to expenditure assignments and taxation as well as transfer design, so that they can begin reforming the system of intergovernmental relations. There are linkages between several reforms, and while it is clear that China cannot move on all fronts simultaneously, it will be important for it to keep a medium-term goal in sight.

IMF Survey: Are there pitfalls that China will need to be mindful of as it moves forward?

Ahmad: The Chinese authorities have always carefully thought through their reform strategies. Any major reform has winners and losers, and balancing these will be the politically difficult part. But the interconnectedness of these measures suggests a sequencing of reforms together with some flexibility in the way that transfers are designed, so that there will be no major disruptions in incomes or production patterns.

The difficulty over the past few years has been impatience in the use of equalization transfers. There have been transfers for wage adjustments, and for western and underdeveloped regions. China could have better results putting everything through its overall equalization system, together with judiciously designed and monitorable special purpose grants. The issue remains whether the center will have the fiscal space to undertake redistribution. That may be feasible only if the country moves along several fronts simultaneously.

IMF Survey: What has the IMF’s role been in this reform process?

Ahmad: The IMF has been seen as an impartial and valued advisor. It does not dictate policies but works with the authorities to assess particular issues and describe the pros and cons of alternate solutions.

In particular, relations between the IMF’s Fiscal Affairs Department and its counterparts in the Chinese government have been exceedingly good. In the recent past, the department offered its advice in the context of a program that has provided technical assistance on tax policy and tax administration, budget and treasury, and intergovernmental fiscal relations. This program, financed jointly by UNDP and DfID, is coming to an end this calendar year, but there is a good chance it may be extended for a new term. In my view, compared with similar programs in other parts of the world, it’s been phenomenally successful.

IMF Survey:You are optimistic then?

Ahmad: I am. China has some very wise and competent people.

Copies of IMF Working Paper No. 02/168, “Recentralization in China?” by Ehtisham Ahmad, Li Keping, Thomas Richardson, and Raju Singh, are available for $15.00 each from IMF Publication Services. See page 93 for ordering information. The full text is also available on the IMF’s website (

Laura Wallace


Sheila Meehan

Senior Editor

Elisa Diehl

Assistant Editor

Christine Ebrahim-zadeh

Assistant Editor

Natalie Hairfield

Assistant Editor

Maureen Burke

Editorial Assistant

Kelley McCollum

Editorial Assistant

Philip Torsani

Art Editor/Graphic Artist

Julio Prego

Graphic Artist

Prakash Loungani

Contributing Editor

The IMF Survey (ISSN 0047-083X) is published in English, French, and Spanish by the IMF 22 times a year, plus an annual Supplement on the IMF and an annual index. Opinions and materials in the IMF Survey do not necessarily reflect official views of the IMF. Any maps used are for the convenience of readers, based on National Geographic’s Atlas of the World, Sixth Edition; the denominations used and the boundaries shown do not imply any judgment by the IMF on the legal status of any territory or any endorsement or acceptance of such boundaries. Text from the IMF Survey may be reprinted, with due credit given, but photographs and illustrations cannot be reproduced in any form. Address editorial correspondence to Current Publications Division, Room IS7-1100, IMF, Washington, DC 20431 U.S.A. Tel.: (202) 6238585; or e-mail any comments to The IMF Survey is mailed first class in Canada, Mexico, and the United States, and by airspeed elsewhere. Private firms and individuals are charged $79.00 annually. Apply for subscriptions to Publication Services, Box X2003, IMF, Washington, DC 20431 U.S.A. Tel.: (202) 623-7430; fax: (202) 623-7201; e-mail:

With its entry into the WTO, China will have to take an urgent look at its production-based VAT, which does not provide credits for the taxation of capital goods.

Other Resources Citing This Publication