Why the rich paid less tax in the 1970s – despite 98% tax rates
taxpolicy.org.uk
taxpolicy.org.uk
When that is said, I whish we stopped using income as the measure of 'rich'. Clearly wealth is a better measure, although it's sometimes hard to get good numbers on it. And today's rich can to a large degree choose their income.
So basically taxing the 1% isn't why everyone could afford college and a house back then. But we are better at closing avoidance loopholes these days. With some very obvious exceptions.
Gary specifically does NOT advocate for an income tax. He advocates for a wealth tax. This piece does nothing to refute that. It's knocking down a strawman, i.e. that we should go back to the exact same tax policies of 1950s-70s. Gary has never said that. He simply pointed out that achieving financial security was significantly easier for average folk back then, and clearly the answer lies at least partially in our attitudes towards taxing rich people back then, compared to today.
The data in the article proves Gary's point. Not only is income tax brought in less now, but more taxes are VAT (sales taxes) which are regressive. In other words, the poorer people in society are taking on a bigger share of taxes. They are transferring them from a progressive tax system (income and wealth taxes) to a regressive one (sales taxes and VAT).
Talk about missing the whole point of Gary Stevenson's argument. The point was that inequality was less back then and the tax system was one reason why. Now inequality is high and keeps getting higher. So yes, we want to go back to the tax system of the 50s, 60s, and 70s not because of getting extra taxes, but because it slows down and may reverse the rise in inequality.
Does the article data show that? Looking at the tax % graph the ratio between Income+NI and VAT+Stamp duties has been pretty constant (apart from a brief time in the 70s)
https://taxpolicy.org.uk/wp-content/uploads/2025/05/image-1....
> In other words, the poorer people in society are taking on a bigger share of taxes.
Seems unlikely. The article shows that high earners are paying are far greater share of income tax. Poorer people would need to be buying a crazy amount more stuff to offset that with VAT.
The UK is in a fairly unique position where its GDP per capita has been stagnant for decades. Which means it's slowly sinking under inflation. It's been going on for so long that the people at the bottom of the boat are now drowning (which I agree is deeply unfair). But no amount of redistributing people between the decks is going to actually fix the problem.
No it's not. The charts showing "stagnant" GDP have already adjusted for inflation, which means it's not "slowly sinking under inflation".
https://ourworldindata.org/grapher/gdp-per-capita-worldbank?...
Gramm et al's book, or at least the first couple of chapters, are worth reading on this topic. Oft-quoted official statistics neglect taxes and transfer payments. There is definitely a large uptick at the top 5% but the spread across the lower 80% isn't as strong as it often is stated, according to Gramm and his co-authors.
https://en.wikipedia.org/wiki/The_Myth_of_American_Inequalit...
Is there another book making the same argument that doesn't have this baggage?
> A simultaneous rise in rewards for work and a growing welfare state that allows more people to live without working virtually guarantees increasing inequality and earnings and incomes, when many of the welfare state benefits are received in kind rather than in money, such a subsidized housing or subsidized medical care, since those benefits are not counted in income statistics.
If you earn $5 and the state gives you $45 in assistance, you practically have $50. If I earn $200 and the state taxes me $100, then I practically have $100. The income inequality is 40x but the practical inequality is 2x.
Current common metrics would report 40x, in this contrived example, is the point Gramm and Sowell both make. Their observation that transfers and taxes are neglected when measuring income inequality is self-evidently true and frequently overlooked. Be careful with income inequality conclusions.
The first chapter of Gramm's book calls out that difference between the hours worked in the various income quantiles is large. Sowell makes the point that the age demographics vary considerably across the quantiles with very young and very old people working less and therefore frequently in the lower quantiles. Sowell further points out that different US demographic groups have different mean ages so that one can explain some seemingly-demographic-related inequality by age differences.
Both books are recommended.
1) We should expect that the share of income paid by the top to increase as GDP increases. As GDP / Capita increases the amount of people moving into a higher taxed bracket increases so the overall revenue increases.
2) The period with a higher tax rate also has a drastic decrease in Debt to GDP [1]. It seems prudent to cargo-cult your policies after ones that historically worked.
3) Instead of taxing the rich, they're borrowing money from the rich which further exacerbates the deficit.
[1]: https://en.wikipedia.org/wiki/United_Kingdom_national_debt#/...
Since then we have had a top income tax rate range from 28% to 90%, but the actual amount of taxes collected is really not impacted by this.
Highest tax rate by year:
https://taxpolicycenter.org/statistics/historical-highest-ma...
Tax receipts as percentage of GDP:
https://fred.stlouisfed.org/series/FYFRGDA188S
There are lots of reasons for this. The main one, however, is that people react to different tax rates and change their behavior.
And very rich people have a huge amount of leeway. If you decide to tax a 1% at 99% they might just decide to simply stop earning altogether. They will continue to be rich for the rest of their lives and their children and their grand children.
I think a lot of people think you can just depend on people's blind greed so they keep on maximizing their earnings regardless of how much they actually earn. However this is not even how it works a little bit.
The economy can be seen as little more then essentially random giant feedback mechanism. Everything you do to it has a effect and it is impossible to predict it accurately. People's behavior changes, their choices change, their goals change, etc. etc. based on conditions and incentives forced on them by the government.
This behavior is described in the so-called "Laffer Curve"... which is not a mathmatical formula for maximizing revenue, it is just a way to illustrate the sort of behavior you deal with when raising taxes.
Nowadays most governments are going to have the _effective_ tax rates maxed out. Governments hire plenty of accountants and economists that have clear understanding on how all of this works.
Which means, very generally, that massively raising tax rates will yield short term gains until the economy adjusts... then chances are reasonably high they will actually collect LESS in taxes then otherwise would be.
Which means that if the government really wants to raise money they need to go after people with little flexibility in their behavior... which is going to be poor and middle class people who live mostly paycheck to paycheck.
No matter how much you tax the middle class or poor they have to work. They don't have choice. Which means that you can actually increase taxes collected by targeting those groups.
They can't target these people directly with taxes because we live mostly in at least semi-functioning democracies and any politician voting on doubling or tripling taxes on those people are not gong to be politicians for very long.
So they resort to inflationary policies like deficit spending, manipulating interest rates, and "money printing".
These policies are actually a sort of "shadow tax" or "indirect taxation". They extract wealth out of the economy for the government to use, which then increases the costs for poor and middle class people.
The whole thing is very messy, but since most people don't really understand how inflation works or where it comes from it is a effective way to short circuit any political accountability.
All of this means that if you want to raise government income without massive negative effects your only choice is to grow the economy.
Don't forget all the hidden taxes like payroll taxes ("It's totally the company paying the tax!") and sin taxes ("This is for your own benefit!"). These are all ways to "hide" how much taxes regular people have to pay.
I think one of the most interesting charts to look at is "government spending as a percentage of GDP". Seeing it grow from about 10% to 55% from 1900 to today (in France) [0] makes you think about government/social spending and its future. It certainly cannot 'improve' the way it has over the past century -we're hitting diminishing returns after getting to 50%.
Think about it: government spending going from 10% of GDP to 20% of GDP doubles the money available for the government, but people retain 80% instead of 90% of their money. Meanwhile going from 60% government spending of GDP to 70% only gives the government 16% more budget.
Also, as an aside whenever you see US numbers in charts like this you have to figure out whether they're using federal government spending or federal + state, because it's only the latter that's comparable to most other countries (and it's typically about 50% more than federal spending).
Right, that's not how it works.
Most people do not try to maximize their revenue and it's why when a sector (say Tech) starts paying very high salaries it takes a long time (if-at-all) for other sector's salaries to increase despite the opportunity cost of working in those sectors being higher now.
I really do not buy an argument that Alexander Bell (and the at least 4 other people who simultaneous invented the telephone) would've given up and done subsistence farming if the tax rate was "too high". You will still have (and possible even more of it given how well DARPA works) technological advanced with higher top-bracket tax rates.
> This behavior is described in the so-called "Laffer Curve"... which is not a mathmatical formula for maximizing revenue, it is just a way to illustrate the sort of behavior you deal with when raising taxes.
> Nowadays most governments are going to have the _effective_ tax rates maxed out. Governments hire plenty of accountants and economists that have clear understanding on how all of this works.
This part is all false.
The main criticism of the Laffer Curve argument is that in practice governments are below the maximum revenue point [1].
---
You've also completely ignored my initial point of 1) which is that we would've seen government receipts increase as a percent of GDP had it not been for the lowering of the tax brackets as more money was within those brackets as GDP goes up. Which would of course result in a lower deficit into a surplus (as it was during Clinton's years before his successor cut taxes).
>Right, that's not how it works.
>Most people do not try to maximize their revenue and it's why when a sector (say Tech) starts paying very high salaries it takes a long time (if-at-all) for other sector's salaries to increase despite the opportunity cost of working in those sectors being higher now.
This makes no sense. I don't see how the observation that people don't immediately flood into programing disproves the claim that high earners will work less if taxes were higher. The two aren't at all comparable. Not everyone is cut out to be a programmer, and even if they are, it takes a non-trivial amount of time to train to become a programmer. In other words, it takes a long time for people to switch to higher paid industries. On the other hand working less costs nothing, and anyone can do that at a moment's notice.
>I really do not buy an argument that Alexander Bell (and the at least 4 other people who simultaneous invented the telephone) would've given up and done subsistence farming if the tax rate was "too high". You will still have (and possible even more of it given how well DARPA works) technological advanced with higher top-bracket tax rates.
No, they'll just kick back and retire. That might be subsistence farming (aka. cottagecore), or not.
>The main criticism of the Laffer Curve argument is that in practice governments are below the maximum revenue point [1].
Given the that the actual curve can only be estimated, and the theoretical curve levels off near the top, and the right side is steeper than the right side, doesn't it make sense to be below the maximum point? If you overestimated the peak and overshot it, the effects would be devastating. Even if you're to the left of the peak, tax raises are unpopular and cost political capital, so it makes sense that politicians aren't trying to perfectly optimize their tax rates.
Very peculiar.
Economic theory is showing itself to be more a dogma than a science.