The Arguments Against a Flat Tax

That's really the question, isn't it? With a highly progressive income tax structure, at what point (salary combined with rate) do we start doing more harm than good?

I don't know the answer to that question. But it seems to me that you can be making up to that amount and still have a very healthy lifestyle with a progressive tax structure.

I don't have time right now to rework a proposal with a 75% rate over 2 million in income. Maybe tonight, or this weekend, I'll work up a simple spreadsheet.

 

Yeah...because drug companies are taxed on revenue, not income after expenses.

Agree that the 90% is way overboard, but we are talking about personal income taxes (on taxable income).

 

Ed, I read your post, and to be honest it's one of those things where part of me wants to respond but I don't have the energy to really give it a proper response for a post of that length. It's not that I don't think it's not worthy of a response, I just don't have the time to give it one that would be worthy of the time and thought you put into it.

But to make a long story short, I fundamentally don't believe that the tax code should be an instrument of social engineering. It shouldn't be used to encourage or discourage certain behaviors. What people do with their money is their business. I just want a system that is most efficient, with an emphasis on both simplicity and minimal rent-seeking.

 

I, on the other hand, fundamentally believe that a tax code encourages and discourages behavior no matter how it is set up. You're just choosing to encourage (or fail to discourage) a different set of behaviors.

Fundamentally, we have fundamental differences.

I would, however, bet my fundamental theory wins against yours in a battle of economics professors. . .

 

Right, because under your system, econ professors are very important. Their studies and papers would determine tax policy.

But Colin's making a philosophical point, not an economic one.

 

Ok, then. I'll take our dueling theories over to the philosophy department then and let them slug it out. Either way, I'm confident that, in the words of that famous philosopher Neil Peart (as retold by Geddy Lee), "if you choose not to decide, you still have made a choice."

 

The Foosinho/Crocker Tax Plan: An Idea Whose Time has Come?

John, good points.

I. Ends and Means

It’s important that each person be able to choose among many possible actions to take. That one is able to do so requires certain external conditions. People sometimes use the word “freedoms” to refer to these conditions. It also inquires that one has certain internal abilities and traits. A good education is important. So is good physical health. Moreover, it is especially important that each person is able to decide what rules and policies govern the society that affects that person. Being able to have good relationships also is very important.

It seems as though the best way to achieve these ends is with the tax plan that Foosinho and I have been advocating. This is somewhat of an empirical claim, as I’m saying that, in fact, the Foosinho/Crocker plan is a part of the approach that is most apt to bring about these ends. But it is also a metaphysical, or philosophical, claim because it has to do with the issue of free will versus determinism. In other words, it’s important that people have a certain degree of free will – an ability to choose between various ends. I suppose it is not known that free will exists. But for reasons I cannot get into, we are justified in believing that it does.


II. The Rate of GDP Growth and Per Capita GDP: How Important are they?

By “GDP Growth” I will mean the difference in the amount of goods and services produced in a nation from time T to time T + 1. By “per capita GDP” I will mean the GDP of a nation divided by the number of people that live in that nation. Given present circumstances, it seems that the rate of GDP growth of nations is at least somewhat important to whether all the people in the nation can realize the ends I have mentioned. For example, the Great Depression -- a period in which GDP growth in the US actually was smaller at some periods of time than it was at previous periods of time -- contributed to a great deal of hardship.

However, some economists get the ends and the means backwards. This idea is similar to notion Jesus was getting at when he said, “The Sabbath was made for man, not man for the Sabbath.” A high per capita GDP is means to certain important ends. It is not an end in itself. For example, if a country has a high rate of growth and 90% of the people in the country die before the age of 18, that is not a good country. Meanwhile, if a country has a low rate of growth but is highly democratic, with freedom of expression and with everyone educated and living long, healthy lives, that is, prima facie, a good country.

Moreover, we know of some nations that have had a low rate of GDP growth and do well in terms of the valuable capabilities I have mentioned. Switzerland over the last couple years is an example. The province Kerela(sp?) is India is another example. Meanwhile, we know of nations with a high rate of growth that don’t do particularly well in terms of the criteria I’ve mentioned. The Soviet Union under Stalin and China under Mao are examples.

In addition, per capita GDP has limitations as a criterion for weighing well-being. Think of the example of when Bill Gates enters the bar, and the guy at one bar stool looks at the guy in the stool next to him and says, “The average income of the people in this bar just went up by $ 1 million dollars.”

Nevertheless, given the world as it is now, there is probably some degree of causal linkage between (1)per capita GDP and well-being and (2) rate of GDP growth and well-being. Moreover, that data is fairly easy to know. So, I will consider how the Foosinho/Crocker tax plan is likely to affect the rate of GDP growth and per capita GDP.

Some economists claim that there is reason to believe that the kind of approach that Foosinho and I are recommending would, if implemented, trigger a lower rate of GDP growth in the US than would a less robust tax plan. However, the correlations that I'm familiar with (and I want to learn more) do not support the hypothesis that the kind of tax plan that Foosinho and I are advocating -- as long as it is coupled with government spending in certain key areas -- would result in a lower rate of economic growth. At best this hypothesis is no more likely than not. First, there are so many variables. Second, it so hard to isolate variables. Finally, I have included some important data below.


III. Important Data on Taxes, Spending and Growth

A. Taxes, Spending and Growth in US History

1. In 1870-1912, federal taxes as a share of GDP were 3%. Average Top Income tax rate: 0%. Federal spending as a share of GDP: 2.7%. Annual Growth Rate of GDP per Capita: 2.2%

2. In 1947-1999, federal taxes as a share of GDP were 17.8%. Average Top Income tax rate: 66.3%. Federal spending as a share of GDP: 19.5%. Annual Growth Rate of GDP per Capita: 2.2%
-------

1. 1912-1929, federal taxes as a share of GDP were 3.9%. Average Top Income tax rate: 37.8%. Federal spending as a share of GDP: 5.1%. Annual Growth Rate of GDP per Capita: 1.2%

2. 1929-1941, federal taxes as a share of GDP were 5.2%. Average Top Income tax rate: 61.9%. Federal spending as a share of GDP: 8.0%. Annual Growth Rate of GDP per Capita: 2.0%

3. 1941-1947, federal taxes as a share of GDP were 15.2%. Average Top Income tax rate: 88.3%. Federal spending as a share of GDP: 29.3%. Annual Growth Rate of GDP per Capita: 3.2%

4. 1947-1973, federal taxes as a share of GDP were 17.3%. Average Top Income tax rate: 83.3%. Federal spending as a share of GDP: 17.8%. Annual Growth Rate of GDP per Capita: 2.4%

5. 1973-1992, federal taxes as a share of GDP were 18.1%. Average Top Income tax rate: 53%. Federal spending as a share of GDP: 21.5%. Annual Growth Rate of GDP per Capita: 1.7%

6. 1992-1999, federal taxes as a share of GDP were 18.7%. Average Top Income tax rate: 38.5%. Federal spending as a share of GDP: 20.4%. Annual Growth Rate of GDP per Capita: 2.7%

(Source: William G. Gale and Samara R. Potter: "An Economic Evaluation of the Economic Growth and Tax Relief Reconciliation Act of 2001")


B. US Worker Productivity (Productivity measured here as gross domestic product produced in the private business sector, per hour worked in that sector.)

1. In 1950-73, the average annual rate of real growth (percent) of GDP 3.9. Of GDP per capita: 2.4. Productivity: 3.1

2. In 1973-95, the average annual rate of real growth (percent) of GDP 2.5. Of GDP per capita: 1.5. Productivity: 1.3
-------------

1. In 1950-60, the average annual rate of real growth (percent) of GDP 3.3. Of GDP per capita: 1.6. Productivity: 2.8

2. In 1960-70, the average annual rate of real growth (percent) of GDP 4.1. Of GDP per capita: 2.8. Productivity: 3.3

3. In 1970-80, the average annual rate of real growth (percent) of GDP 3.1. Of GDP per capita: 2.1. Productivity: 1.9

4. In 1980-90, the average annual rate of real growth (percent) of GDP 2.9. Of GDP per capita: 1.9. Productivity: 1.4

5. In 1990-95, the average annual rate of real growth (percent) of GDP 1.9. Of GDP per capita: .9. Productivity: 1.2

(Source: Joel Slemrod's Taxing Ourselves: A Citizen's Guide to the Great Debate over Tax Reform, p. 97).


C. Cross-Country Studies on Taxing and Spending:

I cannot show the data right now, but, among developed countries, the data suggests that there is usually the cast that as the highest tax rates go up the rates of GDP growth per capita also go up.

Moreover, among developed countries, the data suggests that there is often a positive correlation between the extent of government spending and real income per capita.

See: Joel Slemrod's essay “What Do-Cross Country Studies Teach about Government Involvement, Prosperity and Economic Growth?”

Also, Slemrod’s Taxing Ourselves: A Citizen's Guide to the Great Debate over Tax Reform

Also, Amartay Sen's Development as Freedom

Also, for an interesting and clearly written, non-technical overview of these issues, see Why Economies Grow by Jeff Madrick.


IV. Some Perspective on GDP Rate of Growth

Even if there is some reason to believe that the Foosinho-Crocker Tax Plan would trigger a lower rate of growth than our current tax laws, the rate of growth is unlikely to be significantly lower. Moreover, our plan, coupled with spending in certain key is areas (e.g. education and health care), is likely to help people realize the values I mentioned at the top. Those values are more important than a higher rate of GDP growth.

My main point, though, is that something like the Foosinho-Crocker Plan is worth a try. It's not certain that it would make the world a better place. But it seems reasonable to believe that it would. For instance, key values were realized during FDR's New Deal, the period in US history with the tax system most similar to the kind that we are advocating.

On a different note, it would be great if one day in the future of our planet, rates of GDP growth and per capita GDP did not have any affect on the well-being of people. Unfortunately, we are not there yet.


V. Incremental Experimentalism

It would be better to implement the economic principles that we’ve recommended over the course of, say, 10 years rather than all at once. For example, it would be better to move the highest tax rates up to 60% first. And then we should assess the consequences. If, overall, the result are positive, then a year or two later, we should move that rate up to 70%. Then we assess. If the results are positive, then a year later move it up to 75%. Etc. This would enable us to get a decent understanding of the effects of different tax policies, and it would be less risky than doing it all at once. We can call this approach “incremental experimentalism” or “experimental incrementalism.”


VI. The Foosinho/Crocker Plan: Would it be Bad for Motivation?

Some people are concerned that our kind of tax policies would contribute to people engaging in less productive behavior than they would with a less progressive system. However, this notion is highly questionable. Based on my experience, few, if any, people would be less apt to engage in productive activities if every dollar over $500,000 were taxed 90% than if every dollar over $500,000 were taxed, say, 30%. Among all scientists I know, here, in order of importance, are the motivations for doing science:

1. They want to have the basic necessities of life, such a roof over their head and nutritious food on the table.
2. They cannot not do science.
3. The joy of discovery
4. This one is interesting: Scientists X wants good scientist to realize that X is a good scientist.
5. They want to make the world a better place, and they realize this is the way they can do it.

Being a millionaire is not high on the list. Moreover, under out system, people still can become multi-millionaires. But for some people it may not be as easy.

 

Re: The Foosinho/Crocker Tax Plan: An Idea Whose Time has Come?

Why are these years grouped together? Seems to me that if you want to make an effective study of the correlation between taxes and GDP growth that you would separate years with major changes in income tax rates. This does not do that. Instead, the 1970s which featured high top marginal tax rates are lumped in with the 1980s when top marginal rates were less than half of what they were in much of the previous decade.

So I'm not sure how useful these numbers are.

 

I understand.

One point of clarification. For a vast array of means by which people receive money, the nature of the means should be irrelevant to how much a person is taxed. For example, Maryland Governor Bob Ehrlich wants to put slot-machines in the Maryland horse race-tracks and heavily tax those who play the slots. This is not sensible. For one thing, lower-income people, in general, play the slots to a much greater extent than do higher-income people. So taxing the slots would affect lower-income people to a greater extent. Moreover, playing the slots is not such a bad thing. It does not hurt other people.

Similarly, some people say, “Prostitution should be legal and heavily taxed.” Part of this is sensible, part of it is not. Of course, prostitution should be legal. Moreover, that someone has earned his or her money be performing a sex act should be irrelevant to how much he or she is taxed. Prostitution doesn’t directly hurt anyone. And it helps some people.

However, it is very important that each person be able to choose among many possible actions to take. For example, it is very important that each person is able to decide what rules and policies govern the society that affects that person. Being able to have good relationships also is very important. Further, there is good reason to believe that the kind of tax principles that I have mentioned -- along with (1) government spending in certain key areas (2) certain election and campaign finance laws and (3) significant protections for expression -- would enable each person, or almost each person, to realize these ends.

The data I have presented should be taken with a grain of salt. And I agree that the way some of the data is presented is problematic. I am unable to lay it out in a better way. Moreover, I am interested in getting new and better data. However, a key point I want to make: It is highly inconclusive whether the Foosinho/Crocker tax system or our current tax system would contribute to higher rates of GDP growth and/or higher per capita GDP. Each human is so complex, and there are hundreds of millions of humans on our planet. Also, major acts of nature like floods and hurricanes are hard to predict.

However, it is highly likely that a good universal health plan would enable a higher percentage of US citizens to be healthy than would, say, abolishing Medicare and Medicaid and replacing them with nothing. Moreover, it is nearly certain that, in general, first graders will get a better education if there are only 18 kids in the class than if there are, say, 1,800 kids in the class. And there is pretty good reason to believe that it would be much easier for us to finance important programs with the kind of tax plan that I’m recommending than with a 15% flat income tax.

 

The above is vague and ambiguous. For instance, it's not clear whether I'm talking about per capita GDP or the rate of GDP growth. There is, of course, an important difference between the two. For example, MLS’s average-attendance for a given season is importantly different than the rate of change in average-attendance that MLS experiences from one season to the next.

I am unable in a clear way to present the numbers to which I have referred. But here is what Michigan economist Joel Slemrod has concluded on this issue in his article What Do Cross-Country Studies Teach about Government Involvement, Prosperity, and Economic Growth?:

"Over the last three decades, in all industrialized countries, there has been an enormous expansion of government involvement in the economy, as measured by the share of national income going to taxes or government expenditures. Figure 1 shows that, averaged over the OECD countries [Western Europe, the US, Canada, New Zealand and Australia], the ratio of either tax collections or government expenditures to GDP rose sharply between 1970 and 1990. Arguably it is this expansion of government that uniquely characterizes the post-World War II era.

"From the beginning, the growth in government has attracted critics who view this as an ominous development, endangering the political rights of the citizenry and economic prosperity. Leaving aside the issues of economic freedom, this paper critically evaluates the evidence about the influence of government tax and expenditures on economic prosperity and growth.

"It is worth pausing to reflect on what evidence would constitute support for the proposition that expanded government activity has been misguided. One option would be to assess the extent to which the goals of government expansion -- provision of public schools, maintenance of full employment, insurance against social risks, income maintenance, and adequate provision of certain basic goods and services such as food, shelter, and medical care to all -- have been achieved. Another would be to assess the cost, in terms of a lower average standard of living, of the programs designed to achieve this goal. Economists are a long way from consensus on measuring either the benefits or costs of government involvement. However, even a consensus on these two questions would not settle whether the big-government era has been a mistake, because weighing the benefits against the costs inevitably involves value judgments, about which economics is mute. This said, pinning down the cost is bound to be informative in the debate, because it can then, at least qualitatively, be stacked up against the benefits.

"...The empirical program, at least as it relates to the level, rather than the growth rate of prosperity, runs into an immediate snag that is disconcerting to anyone who is convinced that government has a substantial negative impact on prosperity. Both for a given country over time and across countries, there is often a positive correlation between real income per capita and the relative extent of government. Figure 2 shows the time-series relationship for the United States, plotting real GDP per capita and the ratio of taxes at all levels of government to GDP, for the period 1929-92. The strong positive association is clear. The unprecedented growth of government has occurred over the same period as the unprecedented growth of prosperity. The same story applies to all of the developed world" (emphasis added).
-------------------
Thus, Slemrod concludes that among developed countries there is a positive correlation between, on the one hand, per capita GDP and, on the other, the amount of government taxing and spending "as measured by the share of national income going to taxes or government expenditures." Moreover, Slemrod has not drawn any explicit conclusions concerning government taxing and spending and the rate of GDP growth.

I do think I have some information on how government taxing and spending is correlated with the rate of GDP growth among developed countries. Although I might be mistaken, I seem to recall that it shows no negative correlation between rate of growth and higher income-tax rates.

 

ASF, thanks for that information.

With the above, I did not write exactly what I wanted to get across. I want to focus on income-tax rates for high income-earners as opposed to general government expenditures as a % of GDP. If people earning $ 20,000 dollars per years or less were taxed 99% of their income, that would, under most circumstances, be bad for rates of GDP growth. However, what I wanted to get across -- and what I want to say now -- is the following: I am not familiar with any data showing that, among developed countries, there is correlation between higher income tax rates for high income-earners and lower rates of GDP growth.

This is important because it’s reasonable to believe that some people should be taxed less than they are. For instance, it seems that in the US, the first $ 10,000 dollars someone earns should be taxed 0%. And the more I think about it, the more it makes sense that each dollar a person earns between next $10,001 and $20,000 should only be tax 5%. However, every dollar a person earns over $ 500,000 should be taxed a high percentage. In a nutshell: The idea of high taxes for all is a bad idea, but the idea of high taxes for very high income-earners seems like a good idea.

Secondly, among developed countries, even if we look at tax ratio (regardless of which rates are taxed) as measured by the share of national income going to taxes, I doubt that, over the last 30 years, this is a correlation between lower tax ratios and higher rates of GDP growth. In front of me, I have data on about 20 countries from 1970 to 1990. I’ll give you a flavor of it. Japan’s GDP grew an average of like 3.3% per year and their “General government tax revenue/GDP” is about .25. Norway’s GDP grew an average of like 3.1% per year and their “General government tax revenue/GDP” is about .45. In other words, Norway had about the same rate of growth as Japan and had greater general government tax revenue than did Japan. The USA had about the same annual rate of growth as Sweden, and the former had less tax revenue relative to GDP than did the later.

ASF, you mentioned Ireland and South Korea. It is interesting that (among developed countries) during the 90s they have the highest rates of GDP growth and the lowest rates of general government expenditures as a % of GDP of X. I have always thought highly of Catholic schools. I wonder if that has anything to do with Ireland’s relatively high rate of GDP.

I disagree with you when you write the following, “the data seems to suggest that there is some correlation between smaller government and higher economic growth.” The numbers don’t show a nice, neat breakdown between rates of growth and “smaller government.” The numbers are all over the place. For instance, it isn’t as if during the 1990s countries with a smaller percentage of general government expenditures as a % of GDP always had a higher rate of annual GDP growth.

Incidentally, it is important to ask: What projects is the government spending money on? Some expenditures are probably better than others in terms of rates of GDP growth.

Moreover, if there is only one point I can get across, I want it to be this: A country’s annual rate of GDP growth is much less morally important than how well it does relative to the Developed Index that the United Nations Development Program uses. Information on the Index and the ranking of countries relative to the index can be found here:
http://hdr.undp.org/reports/global/2002/en/pdf/backone.pdf

This index includes a number of variables, including life expectancy at birth, literacy, infant mortality, access to health and % of primary and secondary school-aged students who are in school. It also factors in per capita GDP and GDP growth rates. But the latter two variables are just two variables among many. (And although I am unable to address the issue at the moment, one can make a strong case that those numbers should even be as important as they are.)

The “Development Index” is more morally relevant than merely the rate of GDP growth, because the variables it factors in are more closely tied to free will than is a country’s annual rate of GDP growth.

Moreover, according to the Development Index, Ireland and South Korea do less well relatively speaking than they do relative to annual rates of GDP growth. Here is the ranking:

1 Norway
2 Sweden
3 Canada
4 Belgium
5 Australia
6 United States
7 Iceland
8 Netherlands
9 Japan
10 Finland
11 Switzerland
12 France
13 United Kingdom
14 Denmark
15 Austria
16 Luxembourg
17 Germany
18 Ireland
19 New Zealand
20 Italy
21 Spain
22 Israel
23 Hong Kong, China (SAR)
24 Greece
25 Singapore
26 Cyprus
27 Korea, Rep. of
28 Portugal
29 Slovenia
30 Malta
31 Barbados
32 Brunei Darussalam
33 Czech Republic
34 Argentina
35 Hungary
36 Slovakia
37 Poland
38 Chile
39 Bahrain
40 Uruguay
41 Bahamas
42 Estonia
43 Costa Rica
44 Saint Kitts and Nevis
45 Kuwait
46 United Arab Emirates
47 Seychelles
48 Croatia
49 Lithuania
50 Trinidad and Tobago
51 Qatar
52 Antigua and Barbuda
53 Latvia
54 Mexico
55 Cuba
56 Belarus
57 Panama
58 Belize
59 Malaysia
60 Russian Federation
61 Dominica
62 Bulgaria
63 Romania
64 Libyan Arab Jamahiriya
65 Macedonia, TFYR
66 Saint Lucia
67 Mauritius
68 Colombia
69 Venezuela
70 Thailand
71 Saudi Arabia
72 Fiji
73 Brazil
74 Suriname
75 Lebanon
76 Armenia
77 Philippines
78 Oman
79 Kazakhstan
80 Ukraine
81 Georgia
82 Peru
83 Grenada
84 Maldives
85 Turkey
86 Jamaica
87 Turkmenistan
88 Azerbaijan
89 Sri Lanka
90 Paraguay
91 Saint Vincent and the Grenadines
92 Albania
93 Ecuador
94 Dominican Republic
95 Uzbekistan
96 China
97 Tunisia
98 Iran, Islamic Rep. of
99 Jordan
100 Cape Verde
101 Samoa (Western)
102 Kyrgyzstan
103 Guyana
104 El Salvador
105 Moldova, Rep. of
106 Algeria
107 South Africa
108 Syrian Arab Republic
109 Viet Nam
110 Indonesia
111 Equatorial Guinea
112 Tajikistan
113 Mongolia
114 Bolivia
115 Egypt
116 Honduras
117 Gabon
118 Nicaragua
119 Sao Tome and Principe
120 Guatemala
121 Solomon Islands
122 Namibia
123 Morocco
124 India
125 Swaziland
126 Botswana
127 Myanmar
128 Zimbabwe
129 Ghana
130 Cambodia
131 Vanuatu
132 Lesotho
133 Papua New Guinea
134 Kenya
135 Cameroon
136 Congo
137 Comoros
138 Pakistan
139 Sudan
140 Bhutan
141 Togo
142 Nepal
143 Lao People's Dem. Rep.
144 Yemen
145 Bangladesh
146 Haiti
147 Madagascar
148 Nigeria
149 Djibouti
150 Uganda
151 Tanzania, U. Rep. of
152 Mauritania
153 Zambia
154 Senegal
155 Congo, Dem. Rep. of the
156 Côte d'Ivoire
157 Eritrea
158 Benin
159 Guinea
160 Gambia
161 Angola
162 Rwanda
163 Malawi
164 Mali
165 Central African Republic
166 Chad
167 Guinea-Bissau
168 Ethiopia
169 Burkina Faso
170 Mozambique
171 Burundi
172 Niger
173 Sierra Leone