EIGHT LESSONS FOR PUBLIC FINANCE IN DEVELOPING ECONOMIES by Michael J. Boskin T. M. Friedman Professor of Economics and Senior Fellow, Hoover Institution, Stanford University Public Finance, Economic Development And Social Harmony DEVELOPMENT RESEARCH CENTER OF THE STATE COUNCIL (CABINET) CHINA DEVELOPMENT FORUM, 2006 “CHINA: PROMOTING SOCIAL HARMONY AND DEVELOPMENT” Diaoyutai State Guesthouse Beijing, China March 19-20, 2006 INTRODUCTION Ministers, participants and guests, it is a pleasure to be back at the China Development Forum, so well timed following on the release of the 11th Five year Plan, with its emphasis on continued rapid growth and social harmony, inequality and the environment. The appropriate role of the public sector in promoting economic development and overall societal well being has been the subject of some of the most important writings of philosophers, economists and statesmen over the ages in virtually all cultures and societies. Principles of public finance are deeply embedded in everything from constitutions and laws to public attitudes. For example, in my own country in the late 18th century, the Declaration of Independence and Constitution incorporated such principles as central themes. Questions such as how large and pervasive should the public sector be? How should it finance its spending? What taxes should be used? Who should pay them? For what purposes should debt, as opposed to tax, finance be used? What should be done at the national vs. subnational level? What should be the relationship between the national government and subnational governments? These are questions to which all societies must not only find answers, but must continuously evolve answers in response to changing demography, economic development, domestic political considerations, international competition and best practice. 1 Economists attempt to frame an answer to such questions with primary emphasis on various measures of economic performance. To be sure, other considerations may be important, but tend to lend themselves less to the kinds of analysis that are economists’ natural comparative advantage. Not surprisingly, many generations of academic economists have focused on one or more of these issues, bringing to bear successively more rigorous analytical methods, empirical information, and practical historical and international real-world case study and comparative evidence. What follows is what I consider the main lessons that economists have learned from these analyses, empirical studies, and evaluations of real-world case studies on these various issues. To be sure, the extent to which such lessons are relevant to today’s China would require much more thorough and deeper analysis of a wide range of specific current and prospective Chinese tax, spending, regulation, intergovernmental, and debt policies. But I believe it is useful to keep this summary in mind and for policy-makers, in all countries, constantly to try to evaluate particular fiscal programs or policies and to test them against these lessons. Certainly in the real world, various compromises will have to be made for numerous reasons. But to the extent the fiscal program can, as a rough approximation, reflect these principles as opposed to contradict them, it will lead to better economic performance and enhanced social harmony. 2 II. EIGHT LESSONS OF PUBLIC FINANCE Lesson #1: Relative prices should reflect true relative scarcity values, to the maximum extent possible. Production efficiency is desirable in virtually all circumstances. Relative prices of different goods and services and of inputs should be allowed to reflect their relative marginal scarcity value, or marginal cost. This is a fundamental analytical result in modern economics. It also accords with the historically well-documented difficulties and inefficiencies that tend to result from sustained substantial deviations from production efficiency. Differential taxes or subsidies, and price supports or ceilings, for example, create serious distortions, embed powerful constituencies to retain them, and generate an eventual, more costly unwinding. China’s impending move to liberalize pricing of energy would be an important policy improvement in this context. Keeping the price of some essential goods and services artificially low may seem like a very appealing method to help relatively low-income people, but it is a very inefficient way to do so. If the price of a commodity is subsidized at the margin at well below market rates, everyone pays the lower price: rich, middle income and poor alike. The net subsidy to the poor occurs only to the extent their propensity to consume the good differs substantially from others’ consumption propensity. It is more efficient to have a targeted program, to transfer resources directly to people in need rather than to subsidize their 3 purchases via a much larger, more expensive program that broadly subsidizes purchases. A major policy advance increasingly used in advanced economies is peakload pricing – charging higher prices during times of peak congestion, lower prices at other times. In sectors such as energy and transportation, peak-load pricing properly used can decrease the need for expensive extra capacity and help internalize the external congestion costs. Most societies, however, do subsidize the production and/or consumption of various goods and services, from public education to food, electricity, and health care. If these policies are used, it is best to restrict them to what economists call inframarginal prices, or the price on an initial subset of consumption. Prime examples include low prices on a minimum electricity usage or on lifeline rates for minimal telecom services, e.g. the first few kilowatt hours or minutes, respectively, with the price rising to the full market price for all consumption thereafter. Sometimes, price subsidies end up creating problems that have long-lasting and socially expensive consequences. For example, in the United States, subsidies to health care financing through government provision, the tax system and third party payment divorce most medical decisions from any relationship to their cost at the point of purchase. This encourages consumption to the point where the marginal value is zero, not the opportunity cost of the resources, and thus greatly increases medical spending. The United States is in 4 an early-stage, politically contentious reform effort to have pricing reflect costs over a much wider range of medical care purchasing decisions. Also in the U.S., the cross subsidies among consumers in telecommunications pricing have been at the core of difficulties in regulatory reform. And, of course, the agricultural subsidies in many developed countries – including the U.S. but especially Western Europe – are inefficient, expensive and grossly distort world trade. Lesson #2. Subsidize Only the Lowest-Income People, Not Special Groups or Commodities. Help People Invest in Their own Skills and Future Income To the maximum extent possible, rely on individuals’ and families’ decision-making for their economic well being, with two notable exceptions: to assist those who do not have some minimally decent standard of living, relative to the society at the time; and to ensure against catastrophes, such as immense medical costs, for which a sufficient private market may not exist. Do not try to micromanage the redistribution of income; concentrate on those with the very lowest skills and those suffering economically catastrophic emergencies. Avoid large transfer programs that risk establishing a permanent underclass, worse yet, that may become intergenerational. The extensive European welfare state is a large part of the explanation for Europe’s slow growth and high unemployment. In the U.S., our 1996 welfare reform, with its emphasis on temporary welfare, moving people back to work, etc., has been a great success. 5 Transfer payments should be designed in such a way as to minimize the disincentive to work effort, and should emphasize relief from extreme destitution or catastrophic events and investment in human capital. In this regard, China’s emphasis on universal mandatory rural education is an outstanding example of sound policy. Studies show very high returns to investment in education. As the old saying goes, “Give someone a fish and you feed him today; teach him to fish and you help feed him for a lifetime.” It is also important to manage expectations in developing programs of assistance to the poor and/or poor provinces. When I was Chairman of President George H. W. Bush’s Council of Economic Advisers, he asked me to advise Chancellor Kohl on the economic issues in the reunification of Germany. Recall, at the time, many thought that uniting East and West Germany would be politically impossible. Chancellor Kohl wanted to exchange one Ostmark for one Deutschemark – a huge overvaluation. When I said this would be an economic disaster (Bundesbank President Karl Otto Poehl said so as well), Chancellor Kohl’s response was, “I have to do this for history”. I leave it to others to decide if the Chancellor was correct in his viewpoint. Unfortunately for Germany, my analysis has proved to be correct. The prospects for the East German economy were damaged, the huge transfers from the West have become permanent and a major drag on West Germany, and the (unrealistic) initial expectations fostered in East Germans of quickly catching up with their counterparts in the West have 6 been dashed, leaving deep resentment. Thus, policy makers should manage expectations and should be careful not to radically over-promise what can reasonably be expected to be delivered. Lesson #3. Keep the Hand of the Government in the Economy as Light as Possible. Keep Taxes and Regulation to the Minimum Necessary While it is important for government to finance various types of spending, e.g. for national defense, and to provide a general social safety net, experience indicates that, after some stage of development and after some point, the value of a larger government quickly hits diminishing returns and becomes inimical to growth. Keeping tax rates as low as possible is vitally important. As the chart shows, there is a negative correlation between the size of government, as measured by spending and tax shares of GDP, and economic growth rates. When government grows large, it becomes involved in too many things which might be left to the private sector; powerful political constituencies develop for rent-seeking to try to get through the political process what they cannot achieve through private activity. Taxes get too high. Incentives to work, invent and innovate are weakened and economic performance is seriously damaged. The United States has demonstrated that if most of the economy is governed by market forces, an overall-federal-state-local-tax share of almost 30% is consistent with strong economic growth and low unemployment. The large, 7 complex economies of continental Europe – France, Italy and Germany – have demonstrated that tax burdens approaching 50% of GDP are not consistent with a vital, growing economy. Figure 1 Relation Between Economic Growth And Tax Burden (OECD Countries) 7.0 6.0 average GDP growth rate, 1991-2004 5.0 outlay regression line 4.0 3.0 2.0 receipt regression line 1.0 0.0 20.0 25.0 30.0 35.0 40.0 45.0 50.0 55.0 60.0 government receipts and outlays as percentage of GDP, 2004 To the extent regulation is necessary, use flexible market mechanisms to achieve sensible regulatory goals, not command and control regulation. On the environment, for example, emissions trading regimes are preferable to cumbersome command systems. When I was in the White House, we put in a cap and trade system of emissions allowances for sulfur dioxide emissions from coal burning power plants that reduced the cost of compliance by two-thirds. Indeed, a futures market in the permits was established virtually simultaneously 8 with the spot market. This has been heralded by all sides in the environmental debate as a spectacular success. Lesson #4. Use Comprehensive Broad Base, Low Rate Taxes. The most efficient taxes are general, flat-rate consumption taxes. These will be the least inimical to long-run economic development and future standards of living. Even a flat rate income tax will doubly tax saving, retard capital formation, and ultimately result in a lower standard of living than would otherwise have occurred. A progressive rate structure potentially creates four main problems: 1) The top rates get too high and impede incentives, thereby limiting future incomes. 2) Unstable government revenues surge in booms, collapse in busts. The government must then respond to the collapse with counterproductive higher taxes or politically difficult spending control. This leads to tremendous social disharmony, as happened in my own state of California, and eventually led to a recall of the Governor; 3) It tends to create vast differences and disharmony over spending; those paying high taxes want it limited, those paying little or nothing want more; 4) Perhaps most important, as I demonstrated 30 years ago, a progressive-rate income tax retards individual human capital investment, as the returns to such investment would be reduced by people driving themselves into higher tax brackets. Thus, the main attempt to redistribute income ought to be on the transfer payment side of the ledger, not on the tax side of the ledger. However, if issues of fairness and social harmony make it impossible to sustain public support for flat rate taxes, it is vital that any 9 progressive income (or consumed income) tax structure keeps the highest rate as low as possible, certainly no higher than 30%. Tax rules and budgeting should be as transparent as possible. There should be as much long-term stability in the rules and laws as possible to facilitate long-range planning by households and firms. This may be particularly vexing for an economy in rapid transition, such as China. But it is important to avoid the substance, and the appearance, of arbitrarily changing fiscal rules. In all societies, people believe that other people are getting a better deal from the tax system than they are themselves. People would like their taxes to be lower, and for other people to pay taxes to finance the things they want from the government. A relatively stable, transparent tax system will give people more confidence that it is being administered fairly, and it will result in greater tax compliance and in less social disruption and bitter debate over taxes. The definition of what is a tax should be thought of very broadly. When the government requires people or firms to do something and pay for it themselves, that is very close to taxing them and using the proceeds to finance government spending. But in the former case, it does not show up formally as part of taxes and spending in the budget. Such administrative or regulatory taxes are substantial in all societies and it is important to try to keep them to a minimum and rely as much as possible on explicit taxes and transparent spending. 10 High tax rates eventually become counter-productive, not only lowering economic performance but eventually generating little, if any, additional revenue. Taxes are especially harmful on mobile factors of production, e.g., (eventually in China) capital. It is useful to have the tax system as neutral, with respect to the types of investment and to the decision about whether to consume or invest, as possible. The most effective way to do this is through a broad general flat-rate consumption tax or consumed-income tax. A comprehensive low rate or rates income tax with broad personal exemptions could integrate the various taxes currently levied in China. This would allow, for example, a widow receiving modest dividends to be taxed more lightly – or not at all, if below the exempt amount -- than a wealthy dividend recipient. Over time, capital income such as dividend and interest receipts and capital gains will become more important for a larger and larger fraction of the Chinese population. It is important to keep any taxes on them light and to equalize them as much as possible. A flat rate income tax above a reasonable exemption level that allowed deductibility of saving would be the ideally efficient progressive consumed-income tax. Further, double taxation should be avoided, by integrating business and personal taxes. Taxing returns to capital first at the business level and again at the personal level is both inefficient and unfair. It also encourages debt paying tax deductible interest, rather than equity paying taxable dividends. Given the still high leverage (despite recent improvement) in 11 many Chinese companies, a level playing field between debt and equity would help reduce instability in Chinese firms and the banking system. LESSON #5: Keep a comprehensive measure of the ratio of government debt to GDP under control, by both limiting liabilities and policies promoting economic growth Traditional explicit government debt in China is modest, but large contingent and potential government liabilities in the banking system and social insurance need to be managed. Sound debt management, however, does not necessarily require a nominally balanced budget, even over the business cycle, because inflation erodes the real value of previously issued debt and because it may make sense to debt-finance particularly large net (of depreciation) public investment, (so long as it passes stringent cost-benefit tests – see Lesson 6). A small nominal deficit is not a problem if the real operating deficit is under control. With economic growth, a modest deficit can still be consistent with a stable or declining debt/GDP ratio. For example, with modest (by China’s standards) nominal growth of 6% and a debt/GDP ratio of 50%, an annual nominal deficit of 3% of GDP is consistent with a stable debt/GDP ratio; a smaller deficit would reduce the debt/GDP ratio. Further, because the private saving rate is so high in China, the potential for deficits to crowd out private investment is not as serious as if the saving rate were much lower. Further, it is important to keep tax rates stable or smooth as well as low. In times of temporarily large spending, debt finance may be appropriate; conversely, when spending is temporarily low, debt 12 should be retired. Tax levels should be geared to the normal or average spending level. To be sure, China must work through its bad loan problems and develop its social insurance system. A combination of strong growth, discipline in new lending, closing down the most financially insolvent borrowers and selective capital injection should strengthen the banking system considerably over the next few years. Combined with financial market reforms, that should gradually allow the PBOC to use traditional monetary policy rather than administrative measures to stabilize the price level and the economy. It would also provide a necessary ingredient in any `substantial future move to exchange rate flexibility and capital account convertibility. It should be noted that stable low inflation should be a key goal of economic policy. High inflation will distort relative price signals, increase uncertainty, retard investment, and risk recession to bring it under control. Outright deflation would radically worsen the debt problems for all debtors, including government, as the real value of debt rises when prices fall. Further, outright deflation creates immense difficulty for traditional monetary policy because of the zero lower bound on nominal interest rates, as the Bank of Japan has experienced for many years and which the Federal Reserve was deeply concerned might occur in the U.S. earlier in this decade. 13 As the social insurance system replaces the various piecemeal systems, it is likely that China’s private saving rate will decline somewhat. Since China’s growth will eventually slow (although nobody can be sure when and how much) and the ratio of retirees to workers will rise rapidly and predictably, the social insurance system should not rely exclusively on a pay-as-you-go defined benefit program. What appears affordable today will be a huge problem in the future. Virtually every advanced economy has or will confront this problem. It would be far better to minimize it by adopting a mixed defined benefit and defined contribution system. Finally, anticipating a future of slower growth and more burdensome demography, China has a window of time – several years to a decade perhaps – to address these problems. But if it gets to the end of that window to a future of slower growth and more difficult demography without having both worked through the bad loan problem and deployed a sound social insurance system, the economic and social disruption will become progressively more severe. Lesson #6. Apply Rigorous Social Cost-Benefit Tests to All Spending and Regulation Decisions Continuously implement rigorous, professional cost-benefit analysis to justify public expenditures. In the modern polity, there will be a tendency for the diffuse interests of the taxpayers to be subordinated to the narrower, concentrated interests of those benefiting from the spending. It is very hard to 14 eliminate programs once they have become entrenched, as the post-World War II histories of the United States, Western Europe and Japan suggest. Therefore it is vital to have very rigorous standards for enacting them in the first place. And, to the extent possible, to continually re-evaluate, reform and restructure them, as makes sense over time. That may be especially true for China, as eventually slower growth and demographic pressure will make some programs, such as defined benefit pensions, difficult to sustain. There are two distinctively different episodes when a large expansion of social spending tends to occur. First is in times of great hardship, when the need is so apparent. For example, the New Deal of President Franklin Delano Roosevelt, which created Social Security and many other programs, was launched during the Great Depression of the 1930s. The second is in times of rapid growth when resources are plentiful and everything seems affordable. But it is important to see beyond the present and to test a program for robustness in the coming era of slower growth and/or more pressing demography. The best advice I can give on the latter point is to adopt a no-regrets strategy: Do the things that make sense for governments to do in any event, such as investments in public infrastructure and education that meet stringent cost-benefit tests in the poorer provinces. But be careful not to create an openended, never-ending, growing entitlement. The lesson of Japan in the last fifteen years is instructive. As part of an economic stimulus fiscal program to combat 15 recession, the Japanese embarked on a program of huge public infrastructure spending. No doubt, as anyone who has been in Japan is aware, the public infrastructure certainly needed improvement. But the combination of the political process and the power of the construction companies has made it very difficult to slow the public capital spending, even as the highest priority projects have been worked off, diminishing returns has set in, and the original anti-recession goal has long since vanished. Lesson #7. Align Responsibilities and Resources Among Levels of Government Spending – whether on public infrastructure or social safety net – should be done and financed at the level of government closest to the decisions as possible, with due regard to the extent to which the benefits and costs will affect a broad population. It is important to target intergovernmental grants to activities, not general revenue. It is desirable that they contain a matching feature, a nontrivial co-payment, so that local effort guarantees they will have considerable value to the local residents. General transfers of resources to local governments in non-short-run emergency situations will not necessarily be spent on assisting the neediest in the poorest places. The United States tried such a program in the Nixon Administration, in what we called “General Revenue-Sharing” The idea was that the federal government would collect taxes, send the revenues out to state and local 16 governments, disproportionately going to governments in poorer states, for example, Alabama. However, the idea that these funds would flow through to the poorest residents in the poorest places proved naive. They were partly used to cut local taxes or to spend on goods and services that benefited non-poor people. Worse yet, from the standpoint of the national elected officials, who were getting all the blame for the [high national] taxes, the local officials, by cutting local taxes or increasing popular local spending, generated a constituency and ran against the national officials in subsequent elections. Not surprisingly, General Revenue Sharing is the only large federal government program that was completely abolished in the post-World War II era in the United States. Finally, local jurisdictions should be given room to experiment, because they will learn from, and adapt, each other’s successful policy experiments and avoid the failures. In the United States, a famous Supreme Court Justice labeled this “the states as laboratories”. China appears to make extensive use of this approach, for example, trying out Social Security reform in a couple of northeastern provinces, and more such approaches should be encouraged. Lesson #8. Rapid Economic Development Is, on Balance, a Contributor, to Social Harmony, Not a Detriment to it It is important to emphasize that economic development and social harmony go hand in hand. Slow or no growth will result in a bitter debate over division of a stagnant or shrinking pie. While the benefits of growth tend initially to be unevenly distributed -- inevitable in any event but heightened in a time of 17 remarkably rapid economic growth such as in contemporary China – rapid development is necessary both to generate the resources from which a general social safety net and basic human capital investment can be financed, and to generate opportunities for those who have not yet made it up the economic ladder to have a better chance at upward economic mobility. A comparison of the results of market reform in Russia and China is instructive in this regard. The major reason China is a less equal society today than thirty years ago is the vast improvement in the well-being of 400 million Chinese citizens. This is a very different outcome from market reform than in Russia, where very few benefited much in the early years of reform and many were made worse off initially, as demonstrated in the following chart. Results of Market Reform CHINA VERY RICH MUCH BETTER OFF SAME WORSE OFF RUSSIA Few Small Many Many Few Many Most Few Conclusion I have taken this opportunity to summarize what I believe are some useful practical lessons for any government fiscal program in any society at any point in its development. These derive from rigorous economic analysis, statistical evidence, historical-international case studies and comparisons, and my practical experience in, and as an advisor to, governments for several decades. 18 Exactly how these principles relate to China’s current and likely future economic development and social harmony requires deeper analysis of each potential policy in China’s own context. From an overall strategic viewpoint, China will be much better off if its fiscal programs broadly reflect and are consistent with these principles than if they are far removed from them. We all grapple with these issues, whether in or advising government. China has an opportunity to learn from what has gone right and what has gone wrong in the fiscal programs of many other countries, recently and historically, and even prospectively. Indeed, a major reason why the economic performance of the United States has been so much better than that of Europe for many years, to the point where the standard of living is about one-third higher in the United States, is that we saw the much earlier development of the extensive welfare state, high regulation, rigid markets and high tax regimes in Europe. In France, for example, that combination resulted in economic stagnation, high unemployment and socioeconomic ossification. That window on our own future resulted in a series of changes in United States fiscal policy, starting in about 1981, that have led the role of government in the American economy, while still substantial, to stop well short of European levels, about one-third compared to one-half. This smaller role of government in the economy is certainly a substantial part of the explanation for the superior economic performance of the United States relative to Europe. 19 Finally, China should be justifiably proud of its recent economic development. As near as anyone can tell, there has never been a time in human history where so many people, or such a large fraction of the population, have been lifted out of abject poverty into a substantially improved standard of living in so short a time frame. That is a remarkable achievement. But the many Chinese who have not yet had that opportunity and made that improvement, remain a source of concern, as does the environment. These are welcome emphases in the new (11th) Five Year Plan. They should neither be ignored nor overemphasized. The biggest blight on the environment would be if the 400 million people lifted out of poverty in the last quarter-century were still desperately poor. Likewise on the environment, growth will generate the resources to switch to cleaner technologies. As mentioned above, use prices, not command and control regulation, to internalize environmental externalities. Avoid the excesses, e.g. in Western Europe, of environmental and inequality policies sapping initiative, vitality, employment and growth from the economy. While a rising tide will not lift all boats, it will lift by far the most boats. It is important to continue strong economic growth to lift as many as possible and to assist those who have not or cannot make it from some part of the extra resources generated from the strong growth. That is a far more sensible strategy than one which sets in motion, even if long-delayed, a process that will leave so large a role of government that eventually it becomes inimical to economic growth and hence also to social harmony. 20