Transcript: Rethinking income inequality, with Chris Giles

This is an audio transcript of The Economics Show with Soumaya Keynes podcast episode: ‘Rethinking income inequality, with Chris Giles’

Soumaya Keynes
Before we begin, we would love to hear a bit more about you and what you like about this show. We are running a short survey, and anyone who takes part before August 29th will be entered into a prize draw for a pair of Bose QuietComfort 35 wireless headphones. You can find a link to the survey and terms and conditions for the prize draw in our show notes.

[MUSIC PLAYING]

Everyone knows that inequality in America has risen a lot over the past few decades, partly thanks to the work of Thomas Piketty in his economic bestseller, Capital in the Twenty-First Century. But does everyone know that? A recent bombshell of a paper by researchers Gerald Auten at the US Treasury and David Splinter at the Joint Committee on Taxation, argues maybe not. They take aim at some of the basic assumptions made by Piketty and his collaborators, and they argue that US inequality did not rise as much as previously thought. Who’s right? How much has income inequality in America risen?

[MUSIC PLAYING]

This is The Economics Show. I’m Soumaya Keynes in London, and I’m joined by FT economics commentator Chris Giles, who has taken a special interest in this topic over the years. So, Chris, where were you when Capital in the Twenty-First Century by Thomas Piketty was first published? Did you queue outside the bookshops waiting for the doors to open?

Chris Giles
I’m a journalist, so I was quite lucky. I got sent a review copy, ahead of time even. And yeah, picked it up, looked at it, thought, hey, there’s a lot of good data in this. Then, you know, blew up in a good way, became an absolute bestseller. And, the next time I picked it up, to be totally honest, was, when I thought I was doing something on UK wealth inequality, and I thought, I know I’m going to just put this in the context of other countries, I’ll go to myself, get the reference out and just say how it compares with the US, France, Germany and other countries he had looked at.

Soumaya Keynes
OK. Ominous. I think I know where this is going. What did you find exactly?

Chris Giles
Well, I just wanted to have one number that was the same. And I looked at the current UK wealth inequality, and I can’t remember exactly what the number was. It was X and Piketty’s number was X plus 25 percentage points. Not just a little bit different. It was enormously different. I just thought, wow, this is something. When I’ve got at least a few minutes to look at, why are we getting these differences? And I found I couldn’t justify pretty much any or many of Piketty’s numbers on wealth inequality, not just for the UK, but also when you looked at other countries as well. When he went back through his spreadsheets to the original sources, the numbers didn’t quite match up. And so I wrote a piece which was basically very rude, saying he plucked the numbers out of thin air. They didn’t make any sense. And it blew up. It hit the headlines, particularly in the US, because he had been so big. And, the response from Piketty was that I was an idiot. It was all fine. And then a few years later, he quietly changed his numbers.

Soumaya Keynes
Right. OK. So there was this earlier debate looking at the UK numbers. But the debate we want to talk about today is looking at the US. Now, there’s a kind of parallel discussion going on here because not a journalist, but two economists have gone in to look at the data and said, hang on, some of these assumptions are a bit questionable.

Chris Giles
And absolutely to be nerdy for a second, we’re now looking at income inequality, not wealth inequality. So it’s a different concept. But this is all about incomes and at people’s income they get annually. And we have two tax nerds working in the US government but they are not political appointees. They are both officials, one in the Joint Committee of Taxation in the US Congress and one in the Treasury department. They’ve basically said that if you look at income inequality in the US really carefully, in fact, from about 1960 to 2020, it hasn’t really risen at all. And that is quite mind-blowing because everyone accepts pretty much that income inequality did rise.

Soumaya Keynes
Yeah. I mean, it was incredibly important finding for the narrative, right? You know, we are the 99% protests, you know, discussions about wealth taxes. I don’t know how close any of those ever got to being put in place. But, you know, the idea that there was something deeply wrong with society that these systems that allowed inequality to rise, these results were pretty core to that narrative.

I think, OK, we should talk about who these economists are, because, you know, it wasn’t just Thomas Piketty, right? He had co-authors. You had essentially three French economists arguing that inequality had risen. You had Emmanuel Saez at the University of Berkeley, and also that sort of protégé, Gabriel Zucman, who’s also at Berkeley and affiliated with the Paris School of Economics and Zucman’s also done lots of research looking at tax evasion by big corporations. It all fits into this same kind of there’s something wrong here narrative.

And then on the other side, as you said, we had these two American economists, almost officials saying, actually, no, there’s something wrong here. It’s not the story that you’re telling. We’ve got Gerald Auten of the Office of Tax Analysis. And then there’s David Splinter of the Joint Committee on Taxation. So they are basically publishing these papers, I think starting in 2017, pointing out these methodological flaws. And then you get this huge back and forth, this battle of these economists making incredibly technical, nerdy arguments about the best way to measure this. So, Chris, why don’t you start by walking us through what they actually did, right, because both of them share some methodology, right?

Chris Giles
Yes, absolutely. So the good thing here is that what they’re doing is essentially the same thing. They’re using the same data. They’re using the same measure of inequality brackets. Not a very good one, because what they’re looking at is the share of total income that goes to the top 1 per cent of the population.

Soumaya Keynes
Yeah, that seems like a really narrow measure of inequality, right? We surely want to think about more than that.

Chris Giles
Yeah. So 99 per cent of the population are excluded effectively from this but just I’ll say this right up front, it doesn’t really change the results hugely if you had a better measure of inequality, so let’s just leave it there and we’ll go with it.

And just those are the very, very basic facts just very generally. And it does get complicated. If you look at the raw data, coming out of tax returns. So this all comes out of Americans filling in their tax returns and assuming they are, for the time being, are true that they haven’t lied on their tax returns, then the top 1 per cent in 1960 had about 10 per cent of the total income, and in 2020 had about 20 per cent of the total income. So if you just look at those facts coming out of tax returns, that is what you get. And so you see this very big increase in inequality, just in the sort of broad facts. Both the French group and the US group agree on that. That’s not under dispute. So everything that’s under dispute is how you turn that basic fact into something that tells you something about society, because there are lots of reasons why just looking at how people fill in tax returns might give some quite odd results and you need to adjust for those. And so this is all a very nerdy debate about adjusting tax return data into something that might say something more about society.

Soumaya Keynes
OK, so if you just look at the numbers filed by people on tax returns, it looks like the top 1 per cent of income rose by ten percentage points over those 60 years. But actually, those numbers are not a complete picture of your income. There are lots of other things that could be going on that determine how well you’ve done that year.

And so essentially, you need to make lots of choices about how you estimate those other things, how you go from these numbers you’ve got on the tax return to income inequality, which is what we actually care about. And so this battle is over those choices. Essentially, Piketty says, and Zucman disagreed with Auten and Splinter over those choices. Actually, on that note, this is a lot of names to be bandying around. Can we think of a shorthand? We could have the French versus the Americans? Maybe that’s a bit too general.

Chris Giles
Yeah. How about the academics versus the officials? Something like that. So we got these big cheeses in universities and then these two sort of nerdy guys in the US government.

Soumaya Keynes
OK. All right. The academics thinking inequality has risen, the officials are questioning that. OK. And the other point, I suppose, just to reinforce is that these measurement issues apply to everyone, right? It’s not just the top 1 per cent, it’s also the rest of the 99 per cent. You need to know what’s happened to them, if you want to calculate the share of income going to the top 1 per cent. So this is a pretty general measurement question. So Chris, take us through this. What are the big methodological disagreements here?

Chris Giles
So remember that what we’re trying to do is go from what people fill in on their tax returns to something that means something for society. And so the six big ones, I think, are looking at the different ways and the different income levels of married couples and single people, and how that’s changed over time. The second one is looking at a very big tax reform in the mid-80s, in 1986, and the effect that had on how people filled in their tax returns. The third one is a conceptual issue about, should you care about capital gains? Is this important for income? The fourth one is the big issue of are people lying on their tax returns? Are they just evading tax or are these, is this good data in the first place? The fifth one is quite a technical thing about retirement accounts. There’s something weird going on tax returns and so we just mismeasuring stuff. And the sixth one is, well hang on, there’s lots of things that aren’t on tax return. So some of the most important things are the benefits people get from government, in benefits, in healthcare and all these other stuff.

Soumaya Keynes
OK. So my understanding is that of those six, the first three are actually pretty uncontroversial. It’s pretty obvious who is right. And in some cases they’ve actually changed the methodology to reflect that. So shall we go over those three first and then dive into the more controversial ones?

Chris Giles
Yeah, absolutely. So let’s do marriage first. So it is uncontroversial that in the US over this 60-year period, richer people have stayed together much more than poorer people. So the marriage rate for really rich people has gone from 90 per cent down to 85 per cent. And for middle and poor people from 67 per cent to 37 per cent. A big, big decline. And in the US, you fill in your tax return as a couple. So if you’re doing it over a very long period and you’re looking at the number of tax returns, not the number of people, it means that you suddenly have a period that now you have many more people right at the top and they all got more income, so they get a higher share. And that’s just silly. Everyone has accepted that that has given a silly result. And so that explains about 15 per cent to 1.5 percentage points of this ten percentage point difference. That’s basically been accepted.

Soumaya Keynes
Fine. So it looked like inequality had risen, but actually it was just the way that people were filing their taxes had changed so that you had more single tax units filing them. It wasn’t a real change in their circumstances.

Chris Giles
Yeah, absolutely. That’s exactly what was going on there. Now the second one is this big tax reform in the mid-80s. What that essentially did was because it was cutting tax rates, it made people more willing to declare the income from companies they owned on their tax returns and less, and they didn’t any longer want to just roll up the income in the company and just pretend they’re completely separate.

All of that was completely legal, but it meant that it was not a real change. Again, it was just a change in the behaviour of people’s filing of their tax returns and not anything to do with inequality. It has a big impact on the academic measures in that period between 1984 and 1987. So that basically goes away and has pretty much been conceded as well.

Soumaya Keynes
OK. Yeah. I mean, I do remember it just being really striking how much of the increase in inequality of those original series happened around that time. But really it was just a change in the filing incentives. Again, it wasn’t real changes in people’s circumstances. OK. Number three.

Chris Giles
Number three is a conceptual issue. And this is where I think we can have a disagreement. So the academics, they tend to include capital gains. So where you’ve made money and sold shares at a higher price. Has that improved your circumstances? Whereas the officials say well what we’re trying to do is match GDP or national income. And in GDP we only look we don’t look at things that have changed in value because we look at the flows of money. So capital gains don’t come into it and they don’t, they just don’t go into that at all, which I think you and I might agree that your access to resources has improved quite a lot if you have capital gains, and particularly if you’ve realised capital gains. And so the officials just basically ignore that, whereas that’s another 1.5 percentage points. Everyone agrees on the numbers here, but it’s just about whether you include it or don’t.

Soumaya Keynes
Yeah I mean call me a raving lefty, but I think if someone has a massive capital gain then yeah that’s income. There was a really interesting debate, I think I remember a few years ago about whether, you know, the Elon Musks of this world should be counted as having received a lot of income in one year if their share prices went up a lot. Right? The argument being, well, that’s not necessarily liquid. They can’t access it. It shouldn’t count. But a lot of people would like, no, that’s income. That they’ve got a lot better off there.

Chris Giles
Or even if it’s not income, what’s something they can use right now or spend, they’re definitely better off. And so it’s one of those things that I think in these sorts of statistics, you probably want to include it. And rather than just say, oh, well, this is, rather than taking a really purist line and say, this isn’t national income and therefore we can’t do it.

Soumaya Keynes
Yeah. OK. So I think that’s a conceptual disagreement. Fine. Perhaps we’ll say that the academics have it. Let’s go on to the more controversial ones now. So let’s talk about tax evasion. This got really heated I think looking at the back and forth between these two groups. How would you characterise the debate?

Chris Giles
Yeah, this is where they really like to knock socks off each other. And they really can’t find any agreement on this whatsoever. This is what’s generated, this inequality denial accusation from the academics towards the officials. And the officials just say, well, you’re idiots. You don’t really understand how taxes work.

So it’s all about things that aren’t on tax returns, because obviously we understand that people lie and some people evade taxes and don’t actually write it down on their tax returns. So if you want to know what’s really going on, you can’t just assume tax returns are correct. And so what both sides do is they say, well, we can know from other pieces of evidence that there’s an amount of money we have to allocate, which is money that’s not on tax returns but should be. And the question is who does it go to. Does it go all at the top end, in which case, which is where the academics tend to put it. Or does it go towards the bottom end which is where the officials put it? So they really just have a completely different assumption on where this money goes. And the broad arguments are that the academics are saying, well, let’s just allocate it with the money we can see, because that’s something we know on tax return. So we’ll just allocate it like that. Whereas the officials say, oh but we’ve done some audits. Our tax offices have done some audits. And we think the people who are dodging their taxes more are right at the bottom.

Soumaya Keynes
Right. And then the argument against the audit studies is that particularly rich people might also be particularly good at not being picked up by those audit studies. And so that would be the kind of riposte to that. I mean, who do you think is right?

Chris Giles
I think this one’s really hard. I mean, this is I think this gets to the absolute crux of what we’re trying to do is we’ve got this data from taxes and we’re trying to say something with it, and we don’t actually know. I think, you know, in some ways I would want to split the difference here, but then I don’t have a lot of evidence for why I would do that. But it feels to me that you can’t just ignore the audit studies. But it does seem odd that it all goes pretty much to people who’ve got no income. So that feels slightly odd to me.

Soumaya Keynes
I did see a Brookings analysis of this, and they sort of tried to be almost a neutral arbiter, and they actually came down on the side of the academics saying, actually, our evidence suggests that more of this hidden income should be allocated to richer people in favour of the argument that inequality has risen. But I suppose this is just one thing that people are going to fight over for some time.

Chris Giles
Because we don’t know, because there isn’t data. So what we’re trying to do is we’ve got some data we know is wrong in some way, but we don’t know where to allocate it. And there isn’t an obviously truth. If there’s something true out there, then the tax authorities would be on it.

Soumaya Keynes
Right. Let’s move on to the next controversial one, retirement accounts. What’s going on here?

Chris Giles
So on some people’s tax returns, there will be some very, very big income that will apparently be received because they’ve just taken their income as retirement. But when you look more closely at it and this is what the officials have done, they’ve said in many cases, not in all cases, but in many cases, all of this is is a rolling over of one retirement account into another. And I think that’s basically agreed that some of it is definitively rolling one to another. And the difference is how much you need to get really involved in the micro differences of how people report on tax returns. I think the officials probably are right here because this is something they really know about, but I think that’s where this is going. But it’s definitely, it’s quite, it’s statistically important. So it’s a big number. So we have to include it in our list of differences.

Soumaya Keynes
OK. Well, that sounds very boring and technical. So I want to move on to the final spiciest controversial topic, which is the one of government transfers. This is just really conceptually interesting, right. So when you think about defence spending, right, how do you allocate that income across different people due to allocate out evenly. And then what about health spending. I mean benefits, you know, transfers, that’s pretty easy. People just perceive benefits or not. But with all of those other public services allocating, it just seems much trickier.

Chris Giles
It’s really hard. It’s conceptually difficult and it’s practically difficult. And what the officials say is that at least half of the benefit of health expenditure goes per person. For example, and defence expenditure, everyone gets the same value from it. But this is a philosophical issue. If you have an operation and it is quite expensive and it saves your life, does that benefit you? Clearly does. Does it also benefit me? Because we live in a society that I now know provides this sort of insurance, and that’s a really great thing for me to have as well.

Soumaya Keynes
Or that you’re just really happy that I’m alive.

Chris Giles
Obviously, I’m super happy that you’re alive, right? But these are really conceptual and philosophical issues. And essentially what the two sides do, the officials give more of it, quite a lot of it per person. And that says that poorer and middle-income people get a lot of that value from public spending. The academics, they basically say that it goes without their income. It’s no different from any other piece of income. So you get it on an income basis, which feels a little bit strange to me. I’ve got to say, the rich people get more out of defence and poor people, maybe they’ve got more to lose, but that is the conceptual argument there.

Soumaya Keynes
OK? I mean, there seems to be a pattern here, right? Which is the academics, the people who are arguing the inequality had risen a lot. All of their choices seem to be ones that would tend to push that estimate of the amount inequality had increased up, right? So they’re all going in the same direction. But on the other hand, the officials who are the ones arguing that inequality hasn’t risen very much, all of their choices are going the other direction, right? And so maybe you need a mix.

Chris Giles
I think you’ve got that absolutely right. So we know what’s going on here. Choices made by the academics make the rising inequality really big. Choices made by the officials make it basically go away. So we need a scoresheet about who’s right and who’s wrong. So of the six we’ve got marriage I think that goes in the officials’ favour. I think we’ve got the mid 80s tax reform again in the officials’ favour. We’ve got the capital gains conceptual issue. We can have an argument about that. But for me it goes pretty much in the academics’ favour. We’ve got tax evasion. I would split the difference slightly and we’ve got health which I think goes in the officials’ favour for me and retirement also in the officials’ favour. So for me it goes more in the favour of the officials, of Auten and Splinter, but not the whole direction, but definitively not the whole direction. I don’t think you can legitimately say that inequality hasn’t risen, but I think you can say it hasn’t risen as much as some of the big and stark doubling or whatever you had in the debate over the last ten years or so.

Soumaya Keynes
Yeah. I mean, I want us to go to what this should mean for policy, but in some sense, there is something really weird going on here, right? If inequality hasn’t risen at the same time, you’ve got inequality of other outcomes that has risen, right? We’ve got inequality in health that’s gone up in mortality. We’ve got inequality in wealth that looks like it’s risen. It’s a bit mysterious if inequality hasn’t gone up, why do these other things have? Isn’t that useful evidence?

Chris Giles
I think that makes really good questions. So one thing we know for certain is people aren’t very happy about the structure of society in the US, and quite often it’s just going, oh, it’s inequality. And I think this allows us to have better questions. So what is it about the US healthcare system that, even though it spends a lot of money, does redistribute quite a lot, doesn’t affect US health inequalities, which has definitively gone up, so something’s going wrong there. And what is it that, you know, you have all these other aspects of things that governments do to stop inequality rising, but isn’t making society feel a better place? I think those are better questions than just saying, oh, it’s just all the rich running off with all the spoils, and if we could just tax them, society would be a better place because I think that will end up with the wrong answers. Whereas if this is really, really useful work because it makes us think more carefully about what’s really going on in American society.

Soumaya Keynes
I mean just from other countries’ perspective, there’s potentially a lesson which is that, you know, the US economy has done super well over the last few decades, and it always seemed like, oh, there’s a trade-off here, but you can do really well, but you have these costs in times of higher inequality. And if you start challenging the result that actually there was higher inequality, then it looks like you can have your cake and eat it.

Chris Giles
Oh, I don’t think so because we don’t think that the US is very happy. So this again gets us back to why is the US not happy? So what else has gone on? Is it not just inequality, income inequality that’s the issue. So I think it says that the US can grow fast, can have maybe not as big rising inequality as we thought, but still things aren’t quite working perfectly. And that’s where we need to be thinking about: what does that tell us for what we do in the US, or what the US does and what we do in other countries?

Soumaya Keynes
OK, so there are still trade-offs, just not with income inequality still.

Chris Giles
It’s not one big trade-off, there’s lots of trade-offs.

Soumaya Keynes
Things are complicated. How do I know this is an economics podcast? OK, well, before we go, can we just talk briefly about how applicable all of these debates are to other countries, right? I mean, we started off with your big spat over the Piketty numbers. Do any of these arguments point in a particular direction for the UK?

Chris Giles
I think for the UK it doesn’t have a huge amount of overlap because we don’t really do income inequality measurement in the same way. So we don’t start with tax returns. Maybe we should. We have the same sort of data. It is different, but lots of the issues like the mid-80s tax reform, not the same, some issues about marriage, not the same because we don’t have a tax system that works on marriage in the UK. So I think you’ve got to be super careful about reading this across from the US to the UK, then say that the massive increase in income inequality all the UK data says, existed between the mid-80s and the early 90s, and then it was quite flat before then. It was quite flat after that, but it was huge in the second half of the Thatcher period, a huge income inequality increase. I don’t think this tells us that that didn’t happen. I think it’s pretty clear that that did happen in the UK.

Soumaya Keynes
OK. Final question. How well do economists come out of all of this? You know, on the one hand you’ve got, you know, schoolchildren bickering. On the other hand, you know, academic scholars getting close to the truth. Is this a good sign of the profession’s health or a bad sign of its bad temper?

Chris Giles
I think it makes things quite easy for critics because it says that oh, look at all these economists, they can’t agree. You have these nerdy officials who are batting back against a bunch of academics who don’t really understand it as well as the officials, and it makes economists look really bad. That’s not how I would see this. I’d see this as we are in a search for truth. It’s complicated. We have imperfect data, highly imperfect data. We want to say something about society. We need a good punch up to try and get closer to the truth. And although this is messy, this is a really, really valuable exercise. And I think it’s just in that sense, it’s a great paper that we’ve been talking about because it allows us to think deeply about data because quite often with data, we just say, oh, this is what the figures are, this is what the facts are. And it just shows us that when you’re dealing with data, it’s messy and difficult and you’ve got to think about conceptual issues. And that is something that we should all learn and keep close to our chests.

Soumaya Keynes
Yeah. And maybe if there was a bit more mess in other countries’ debates, that would be a good thing. OK, Chris, we’re going to take a break now, but I want you to think about this question, which is, what is it like being in a fight with the world’s most famous economics rock star? How are relations now?

[UNHEDGED PODCAST TRAILER PLAYING]

Soumaya Keynes
We’re back. Chris, what was it like being in a fight with the world’s most famous economic rockstar?

Chris Giles
Well, I’m going to show my complete naivety here because I found these things that I thought were really wrong with his numbers. And I got really excited, did loads of work, and felt like I was in an academic environment again. And I really went for it. I really wrote a very nasty article and blog, told him that he was pulling figures out of thin air and lots of other quite big accusations, and I didn’t realise that doing that with the FT name behind it was going to blow up far more than if it was just me. And so it ended up being quite fun, quite intense, and a massive fight, because partly there was no other US news at that time because it was a public holiday. And so it was the only thing around. And all these things which are luck rather than planned seem to come together to make it a massive blow-up.

Soumaya Keynes
Right. But I’ve seen you on social media, Chris. You’re not you’re not exactly reluctant to enter into a bit of a conflict. I mean, could you rank that against your other top fights?

Chris Giles
Oh, I think that was, the largest, the biggest fight. But the longest fight I’ve certainly had is with the Office for National Statistics in the UK over the retail price index.

Soumaya Keynes
Yes. I knew you were going to say that one.

Chris Giles
Yeah, because it’s a terrible index, but I’ve won that one now, so I’m very pleased about that. I had a big spat with the International Monetary Fund over some of their analysis in about 2012, which was quite big as well. And I had a big row with Gordon Brown and the Treasury in the late 2000, because the way he was presenting some of the figures was not really an accurate way of just the amount of trouble the public finances were in.

Soumaya Keynes
Truth to power. It’s here. Chris, thank you so much.

Chris Giles
It’s been an absolute pleasure, Soumaya.

[MUSIC PLAYING]

Soumaya Keynes
That is all for this week. You have been listening to The Economics Show with Soumaya Keynes. This episode was produced by Edith Rousselot with original music from Breen Turner. It is edited by Brian Urstadt. Our executive producer is Manuela Saragosa. Cheryl Brumley is the FT’s global head of audio. I’m Soumaya Keynes. Thanks for listening.