The Failings of the Economic Freedom Index
For free market economists, the issue of economic freedom is very important. No country in the world is consistently ruled according to libertarian ideals and no country is completely socialist, but there are clearly great differences among countries in their degree of capitalism and socialism or statism in general.
Clearly there is a great difference between the degree of economic freedom in the traditional capitalist bastion of Hong Kong and in Stalinist North Korea. But beyond such clear-cut cases how do we determine which countries are the most free and the least free?
This is not an easy question because often it is not as clear as the comparison between Hong Kong and North Korea, where Hong Kong is arguably much freer in all aspects. Often it is the case that one country is freer in one aspect while being less free in another.
For example, take Sweden and China. Which of these are the most free? On the one hand, Sweden has an enormous welfare state spending more than 55% of GDP and also has extremely powerful unions which have been given the power by the state to force companies operating in Sweden to obey their command, while China has public spending of only about 20% of GDP and has no unions with any real power. Moreover, labor and environmental laws are far more intrusive in Sweden than in China.
On the other hand, despite great progress in recent years in reducing tariffs and non-tariff trade barriers, China is still more protectionist than Sweden. Also unlike Sweden, China still has capital flow controls and its state-owned companies still play a greater role (although that role is steadily declining due to the extremely high growth rate in the private sector) in the economy than in Sweden. And the court system and local authorities are clearly more arbitrary and corrupt than in Sweden. So, is the Swedish or the Chinese economy the most free?
To get to this problem, a need arises for an economic freedom index. But the problem remains: what items should be included and how much weight should be given to each item? Ideally, all forms of government intervention should be included and the weight given to each item should be in relation to just how much it restricts people's freedom to perform economic activities either in the form of taking away their money, regulating and restricting the ways in which they can use the money they get to keep as well as how these restrictions affect their behavior.
But in practice it is nearly impossible to measure just how much a certain tax or regulation restricts people's freedom. Taxes are the easiest to measure since we know how much money people pay to the government, but even this fails to completely account for the damage made because the disturbance of taxes goes beyond just the money people lose. We must also take into account just how much people have changed their behavior to avoid paying taxes and that is difficult to measure.
Still, while the uncertainty surrounding it means that one should take the exact scores in a freedom ranking with a grain of salt it should be possible to roughly estimate the importance of various forms of restrictions of economic freedom and accordingly the overall level of economic freedom.
Let's now analyze the recently released "Index of Economic Freedom 2005" from The Heritage Foundation. It may seem presumptuous of me to criticize the index without having made an alternative index myself, but this is merely meant to indicate how the index should be improved to better reflect the actual level of economic freedom.
One striking feature is the low ratings of almost all poor countries. Only Estonia and Chile make it into the top 19 countries. What is even more surprising is the high scores of notorius welfare-state countries like Sweden and Denmark. Denmark is considered slightly more free than the United States and Sweden only slightly less free than the US. Many other notorius welfare-state countries like Finland, Holland, Germany and Austria are also considered only slightly less free than America.
Looking at specific components the results are even more puzzling. Sweden and Denmark, which have the highest taxes and biggest welfare programs of all countries in the world, are considered to have a smaller "fiscal burden" than both the United States and China, even though public spending as a percentage of GDP is more than 20% points lower in the United States and more than 30% points lower in China.
To understand why these strange results have come about, let's analyze the index by the components it has—and the ones it doesn't have.
The first category, "Trade Policy," seems OK as it measures the degree of protectionism by both the average tariff level and the existence of various non-tariff restrictions on trade like quotas. Yet to get a true picture of just how protectionist a country is, indirect restrictions on imports should also be included. This includes the farm subsidies present in most rich countries, including the United States and the European Union, as well as loose monetary policies which contribute to a weaker exchange rate, something which has the same effect as a combined import tariff and export subsidy.
The second category, "Fiscal Burden," is perhaps the one that is most misleading. Again, it has the absurd result that Sweden and Denmark have a lower fiscal burden than the United States and China. The reason it comes up with such absurd results is its absurd methodology, although it seems in some cases based on them having their own peculiar ideas on how high taxes are. Danes would undoubtedly be very surprised if they found out from this index that they only have to pay 26.5% in income tax.
Unfortunately for the Danes, the Danish tax authorities disagree with the authors and their attributed source, The Economic Intelligence Unit, and says that Danes have to pay more than twice as much.
But again, the errors are mainly due not to incorrect data but to deeper methodological flaws. This category has three different components: change in government spending, top income tax rate and corporate tax. Looking at the first one, why should change in government spending be seen as a measure of the level of fiscal burden? Government spending of 20% of GDP is of course a far lower level of oppression than government spending of 50% of GDP, even if the first country saw its spending increased from 19% and the second saw it decreased from 51%. Clearly, this makes no sense, which is why the authors do not equate the change in freedom with the level in any other part of this index.
The two-thirds of the category that is composed by taxes is only somewhat less flawed. Here they at least use the level of taxation and not its change as the measure of the level of fiscal burden. Yet to use only the top income tax rate and the corporate tax rate as a measure of the burden of taxation is misleading for several reasons. First, because in most countries people with lower incomes have a lower marginal tax rate than the top rate. Second, because in some countries there is a much higher possibility of lowering your tax burden through various deductions than in others. And third and perhaps most importantly, because it completely ignores the burden from other taxes.
In a country like Sweden, it would be misleading to say that people with low or average income pay 31.5% in taxes while the rich pay 56.5% in taxes. That is because the product created by working is also taxed by a 33% payroll tax and a 25% value added tax. The total tax wedge, the difference between what the employer pays and what the employee gets to keep is almost 60% for most people and more than 70% for the rich.
Moreover, it completely neglects forced-savings schemes which some countries, most famously Singapore and Chile, have which act as a substitute for traditional pay-as-you go pension systems. This is basically the system proposed by the Bush administration and other advocates of social security "privatization" in America. In Singapore for example, the government forces people to save some 33% of their income. While one can perhaps argue that forced savings is preferable to pay-as you go pension systems as it is at least not redistributive, anything you are forced to do is clearly not compatible with economic freedom. The exclusion of this means that the level of economic freedom is exagerated in Singapore and some other countries.
Moreover, this category should be given a lot more weight than the others. An economy where the government confiscates and controls more than half of all resources cannot be characterized as free by any stretch of the imagination regardless of how much it regulates the economy with other means. High tax rates function as a form of internal tariff, reducing the rewards from productive activities and sharply reducing or eliminating the incentive for the division of labor, thus destroying the market signals on how resources should be allocated. This is enhanced by the increased disincentive to work created by high benefit levels in various forms of welfare and other transfer payments.
Moreover, it means that the control of how resources should be spent goes from the private sector to the government. Given its size in many economies today, government taxing and spending should be seen as the most important restriction of economic freedom. To say that all categories of restrictions of economic freedom must be given equal weight is completely arbitrary.
The next category is "government intervention," which is defined as the degree to which government directly controls spending. This category includes government consumption expenditure, that is, direct purchases or production of various services, like the military, the police and courts, education, infrastructure, health care, etc. In addition, it also includes government ownership of business. As far as it goes it seems fairly reasonable, although one can question whether received revenues are really the best indicator of the degree of government ownership and the degree to which that ownership restricts economic freedom. However, strangely absent from this category is the degree to which a country practices so-called industrial policy. That is the kind of selective support most famously practiced by Japan and its MITI department, but to a varying extent practiced by most countries including the United States.
The fourth category is "monetary policy." One would think that this category would measure the extent to which the government uses monetary policy to manipulate the economy. But according to the explanatory section, economic freedom in the monetary area is not inconsistent with government manipulation of money. That is because "There is no single, accepted theory of the right monetary institution for a free society." Well, left-liberals don't accept the argument that an extensive welfare state is inconsistent with economic freedom. Quite to the contrary they believe it liberates people from poverty and insecurity and liberates women from the patriarchial structures of the traditional family. Does this mean that a large welfare state should not be seen as inconsistent with economic freedom? Likewise, paleoconservatives don't accept the argument that a protectionist trade policy is inconsistent with economic freedom. Does this mean that the category "trade policy" should be erased from the index? Clearly, there can be no doubt that things like high taxes, protectionism and central banking are infringements of economic freedom.Of course, in today's world no country has a fully market-based monetary system but that should ideally mean that no country should be considered free and given a perfect score.
Even if one insists on the somewhat misleading requirement that the index should be purely relative and that some countries thus must be given the perfect score, average consumer price inflation during the last decade is hardly the best measure of relative monetary freedom. Relatively low but unexpected inflation can be more redistributive than relatively high but expected inflation. Moreover, monetary policy distorts not just domestic consumer prices but also asset prices, interest rates, and exchange rates. Any estimate of how much the economy is distorted through monetary policy should also include expected inflation versus actual inflation, money supply growth, asset price inflation, exchange rate fluctuations, and interest rate manipulation.
Moreover, the index should take into account the extent to which monetary policy is used to finance government and to actively manipulate the economy. Since no country has a market based monetary system all countries could be said to suffer from this, but some countries have no independent monetary policy. This includes countries with a currency board like Hong Kong and Estonia, as well as countries which have "dollarized" their economies like El Salvador and Ecuador, as well as countries which are part of a monetary union, most noteably the Euro zone countries. Since they are influenced by a foreign central bank, their monetary system cannot be considered free, but at least their governments lack the distrectionary power of an independent central bank to finance their operations and control their economy.
To estimate the extent to which monetary policy affects relative prices one could look at which sectors experience growth in credit.Admittedly,these indicators should also be taken with a grain of salt as they could be driven by real factors—just like the consumer price index could be driven by real factors—but at least they will indicate the other forms of distortion besides consumer price inflation that monetary policy could create.
The fifth category is "Capital flows and Foreign Investments," that is the degree to which these things are regulated by the government. In this area, the methodology seems basically sound.
The sixth category is "Banking and Finance," wherein they measure the extent to which government controls banks and financial markets. Their methodology here seems more or less sound apart from the fact that they strangely neglect how both banks and financial markets are manipulated by the government institution known as central banking.Commercial banks are allied with the central bank and strongly influenced by central bank policy which sets their rates, and central banks like the Federal Reserve strongly influence stock markets with their interest rate decisions. Even the slightest changes in wording in Fed statements and speeches from Alan Greenspan are often enough send stockmarkets into a tailspin. They often actively try to "save" the stock market from falling. There are several examples of when central banks like the Federal Reserve have done this.
Thus, this category should also take into account the degree to which central banks manipulate financial markets. Moreover, given the impact massive government lending and taxation on capital has on the financial markets, budget deficits and capital gains and dividend taxes should also be considered in this context.
The seventh category is "Wages and Prices." It describes to what extent government controls prices. Apparently, it only seems to include direct price and wage controls by the government apart from interest rates, the controlling of which this report consider consistent with the free market. But strangely, they exclude from the index the strong power of unions in many countries like Sweden, Denmark, and Finland to control the most important price of all in the economy (apart from the governmentally controlled interest rate): wages. Countries like Sweden and Denmark have no formally legislated minimum wage, but the unions have been given so much power by the government that they can and do force virtually all companies to bow down and obey their command, including their minimum wages. And the minimum wage levels they impose are far higher than in the United States where the federal minimum wage is less than a third of the average pay.
In Sweden, the union imposed minimum wages are usually twice that level. Although some states and cities in the United States impose a somewhat higher minimum wage than what federal law requires, none come even close to the levels in Sweden (Especially since these localites usually have a higher average pay than the U.S. average). While the strong union power in Sweden and many other countries are briefly mentioned in the text of these countries, it has no effect on their scores. Since a minimum wage at two thirds the average level restricts freedom far more than a minimum wage at one third the average level one would expect Sweden to score far worse, but strangely they score the same near-perfect "2." This is even more strange given that they also explicitly note in the text how high levels of unemployment benefits also act as an indirect minimum wage.
Lest anyone think that unions are simply a voluntary private entity, it must be emphasized that they derive their power from government regulation that prevents companies from firing union members and workers who are on strike. In Sweden and probably some other countries, they are also given direct and indirect financial support from the government. And given the importance of wages, they together with other direct and indirect (via unions) labor market regulations should be given a separate category.
The eighth category is property rights which measures the extent to which property rights and contractual agreements are considered secure. Apparently, this only seems to mean security from private crooks as well as from expropriation. The gross violation of property rights that high levels of taxation constitutes seems irrelevant for this issue. An exclusion which might have been justified if they had already accounted for it by giving the issue of taxation a far higher weighting than other categories but which now gives very misleading results.
The ninth category is "regulation." This is supposed to measure to what extent business is inhibited by various forms of regulation. One component in this is how easy it is to get a business license. The second is how tough regulations are after the business gets started. And the third is how consistently these rules are applied. The two first components both seem proper to include, although in the first one should also take into consideration the extent to which it is possible to operate a business without a formal business license. In third world countries that is generally easier to do than in rich countries.The third one is very questionable since it says that the more the government actually carries out their restrictions on freedom the more freedom people have. It does seem reasonable to say that given a certain level of actual restriction it is preferable that it is evenly applied since this decreases the risk of officials abusing their position to extort bribes.
But it is backwards to argue—as this index does—that given a certain form of formal freedom restricting regulation they should be applied as much as possible and accordingly freedom should be restricted as much as possible. Even if (which is not certain) this leads to officials abusing their position to extort bribes, this is preferable since the businesses—being willing to pay these bribes—suffer less from paying the bribes than to adhere to the unjust regulation. To take a rather drastic example: the German army and the SS were known for being effective, reliable and loyal to Hitler. Yet few people would consider these traits desirable in that context. Instead it would surely have been preferable for concentration camp guards to be unreliable and corrupt, accepting bribes from Jews who wanted to flee, than for them to be professional and unwilling to be bribed. Again, this might seem like a drastic example but it clearly illustrates the general principle: corruption is bad, but consistent application of unjust laws is worse.
Indeed this is noted in the next section where the authors correctly note: "In some circumstances, notes Harvard economist Robert Barro, corruption may be preferable to honest enforcement of bad rules. For example, outcomes may be worse if a regulation that prohibits some useful economic activity is thoroughly enforced rather than circumvented through bribes." Yet the authors take the directly oppositeapproach in this section.
The last category, "Informal markets," is the strangest of them all. It gives negative points for having a large informal market, also called an "underground economy." But since informal markets are markets where people do not have their freedom restricted by the state this should if anything be counted as something positive. The larger the informal market the greater chance people have to conduct their business without being taxed and regulated by government officials. In countries with a small informal sector it is far more difficult to find other people with whom you can do business and practice division of labor without having your freedom restricted by the state. Giving negative points in a freedom ranking for a large informal sector is outright Orwellian—"Freedom is slavery."
In their explanatory section, the authors try to justify this by saying that they do not really mean to consider informal markets per se as negative for freedom but that they want to use it as a kind of indirect measurement of how much freedom is restricted for businesses operating within the formal sector. The less economic freedom the bigger will the informal sector be, they argue. Yet there are several problems with this. First, even if it were a good indicator of the degree of restrictions on formal businesses it would only confirm what a correct index of economic freedom has already concluded. Second, while it is certainly correct to argue that ceteris paribus—other things being equal—lower economic freedom for businesses operating in the formal sector will increase the relative size of the informal sector, it would be wrong to assume that this is the only or even the most important factor determining the size of the informal sector.
The efficiency of the government in cracking down on the informal sector as well as cultural attitudes toward working in the informal sector are probably even more important. Tax evasion is usually more widespread in third world countries with relatively low tax rates than in high-tax Scandinavia because government is less powerful and less efficient in the third world countries than in Scandinavia. Third, the presence of a large informal sector means that the possibility to evade state control is greater and thus in itself constitutes freedom.
The informal sector category should therefore be eliminated.
So, how would my proposed changes affect rankings in The Economic Freedom Index? As this article is not intended to produce a new index, I will not produce a new ranking list. However, in general the increased weight on taxing and spending and union power as well as the elimination of the informal sector category and revision of the regulation category will probably lower the rank of Sweden, Denmark, and other welfare-state Western European countries. It will also improve the scores for most third world countries. Hong Kong will probably remain number one, while it is more dubious whether Singapore with its high level of forced savings and its use of "industrial policy" can remain in second place.
Note: The views expressed on Mises.org are not necessarily those of the Mises Institute.