Popular in Social Sciences

Books

City Time: On Being Sentenced to Rikers Island

Independent Scholar David A. Campbell

Culture and Society

The City as a Growth Machine: Toward a Political Economy of Place

New York University Harvey Moltoch

Economics

Gender and Economics

University of California Santa Barbara Shelly Lundberg

American Politics

Keeping the Republic

Boston College Marc Landy

Banks and Policy Organisations

Greening the financial sector

European Investment Bank Matteo Ferrazzi

British Politics

Entrepreneurial thinking: the politics and practice of policy impact

University of Sheffield Matthew Flinders

Culture and Society

Carceral Entrapments: Views from the Prison/Street Interface in India

Indian Institute of Technology Delhi Mahuya Bandyopadhya

Top Wealth Shares in the U.S.: What is the Role of Family Firms?

View more insights in Social Sciences - Economics - Microeconomics

View more figures from University of California Berkeley

View the page of Andrew Atkeson

Read the Study

Andrew Atkeson explores the impact of family firms on wealth inequality in the US. Findings show significant mobility within the wealth distribution.

Image courtesy of interviewee. December 7, 2023

Generating Workspace

Export as PDF

This video video discusses extreme wealth concentration in capitalist economies, particularly in the context of the U.S. It highlights the stark contrast between the wealth of top billionaires, often exceeding hundreds of billions of dollars, and the median U.S. household net worth of approximately $200,000. The research led by the author and a graduate student named Magnus focuses on understanding the factors that contribute to this wealth disparity, specifically examining the role of family firms.

Four key facts motivate their research:
1. There is significant wealth concentration, resembling a Pareto distribution.
2. The shape of wealth distribution has changed over the last century, peaking in the 1920s, declining during the Great Depression and WWII, and rising again since the 1970s.
3. A notable portion of wealth is self-made, with two-thirds of the Forbes 400 coming from relatively middle-class backgrounds.
4. Many wealthy individuals invest heavily in a single firm, often one that they or their family founded.

The research employs a random growth model to analyze wealth accumulation dynamics, noting that a small percentage of families take significant risks in family firms, while the majority maintain diversified portfolios with steadier returns. The findings suggest that rapid changes in wealth distribution can occur due to the success of those taking high risks in entrepreneurship, linking individual choices and societal financial structures.

The results imply that wealth creation in capitalism is not merely a function of having large firms and entrepreneurs but involves deeper connections between finance, risk-taking, and policy, including the effects of venture capital and pension system reforms. The authors advocate for further exploration into how such financial structures and policies impact wealth concentration and distribution.

Language:

Export as PDF

X-TIMESTAMP-MAP:LOCAL=00:00:00.000,MPEGTS=900000

00:00.375 --> 00:03.235
One of the most striking features of capitalist economies,

00:03.335 --> 00:05.315
you know, around the world is

00:05.315 --> 00:07.555
that the very wealthy holds so much wealth.

00:08.495 --> 00:12.675
And, this is readily apparent.

00:12.895 --> 00:15.155
If you look at any list of top billionaires,

00:15.155 --> 00:18.555
you'll see people who these days have hundreds of billions

00:18.555 --> 00:20.195
of dollars at the very top.

00:20.975 --> 00:24.075
And this is in a world where the average US household,

00:24.115 --> 00:26.275
I think the median US household has about

00:26.275 --> 00:28.235
$200,000 in net worth.

00:28.735 --> 00:30.915
So the differences between the median

00:31.055 --> 00:33.955
and the very top are just, astronomical.

00:34.415 --> 00:38.755
And so this simple observation raises the question of,

00:38.855 --> 00:41.595
you know, what is it about capitalist economies

00:41.595 --> 00:44.035
that generates this extreme concentration of wealth?

00:44.535 --> 00:48.755
And so my research, you know, I'm working with Magnus, ie.

00:48.845 --> 00:50.995
Who's actually a graduate student coming out

00:50.995 --> 00:55.795
of London School of Economics in Finance, is part

00:55.795 --> 00:59.035
of a rapidly growing literature on that question of

00:59.095 --> 01:02.115
how it is that we generate this great concentration

01:02.115 --> 01:03.795
of wealth in capitalist economies.

01:04.535 --> 01:06.435
And, and not only that,

01:06.575 --> 01:09.275
but it's looking at how has the concentration

01:09.275 --> 01:11.915
of wealth changed over time in the last century in the

01:11.915 --> 01:14.155
United States and in many other countries.

01:14.935 --> 01:16.475
And so what Magnus

01:16.475 --> 01:18.995
and I are focused on is the role of

01:18.995 --> 01:22.875
what we call family firms in driving the accumulation

01:22.875 --> 01:24.635
of great wealth and allowing it

01:24.655 --> 01:26.555
to change dramatically over time.

01:27.215 --> 01:30.355
And before getting into the details of like what we do,

01:30.995 --> 01:34.675
I just wanted to review kind of four facts that motivate us

01:34.735 --> 01:35.915
to take the approach we do.

01:36.375 --> 01:38.795
So I've already discussed the first fact is

01:38.795 --> 01:43.035
that there's extreme concentration of wealth, and,

01:43.095 --> 01:46.955
and that the top of the wealth distribution looks like it's,

01:46.975 --> 01:49.315
has a Pareto distribution, you know, so it's,

01:49.415 --> 01:52.915
it many things in nature have a normal distribution

01:53.055 --> 01:54.635
or a log normal distribution,

01:54.735 --> 01:57.755
but it stands out for wealth at it's Pareto.

01:57.855 --> 02:00.115
And that's how you, you know, that's characteristic

02:00.115 --> 02:02.275
of this extreme right tail.

02:02.975 --> 02:06.195
But the second fact is the shape of the top

02:06.195 --> 02:09.915
of the wealth distribution has actually changed relatively

02:09.915 --> 02:12.995
rapidly over the last a hundred years in the United States.

02:14.075 --> 02:16.215
It was very high in the 1920s.

02:16.435 --> 02:19.535
You know, the, we, we, it fell dramatically

02:19.535 --> 02:21.575
during the Great Depression and World War ii.

02:22.195 --> 02:26.335
It stayed kinda low until the 1970s or 1980s,

02:26.675 --> 02:30.255
and has now the concentration has now climbed back up again.

02:30.255 --> 02:33.175
There's debate as to whether we're at the 1920s level,

02:33.275 --> 02:35.415
but we're certainly at a very high level

02:35.415 --> 02:36.535
of wealth concentration.

02:37.395 --> 02:40.375
But then the, the next two facts are really

02:40.375 --> 02:41.655
about individuals.

02:42.355 --> 02:46.815
So one striking feature of a capitalist economy is many

02:46.815 --> 02:48.935
of the top fortunes are self-made.

02:49.475 --> 02:53.495
So in the Forbes 400, you see roughly two thirds

02:53.495 --> 02:55.375
of the Forbes 400 came from,

02:55.725 --> 02:56.815
they might have been middle class

02:56.915 --> 02:58.695
or upper middle class, but now

02:58.695 --> 03:02.405
They're extremely wealthy, in, in a single lifetime.

03:02.505 --> 03:03.645
And I, I stress that

03:03.645 --> 03:06.645
because, you know, a traditional notion

03:06.645 --> 03:10.245
of aristocratic wealth, you know, persists for generations

03:10.465 --> 03:12.645
and takes a long time to accumulate it.

03:13.105 --> 03:18.005
And then the fourth fact that really struck us is

03:18.085 --> 03:21.845
that if you look at the portfolios, the wealth holdings

03:21.985 --> 03:25.805
of the very wealthy, the vast majority of them

03:26.955 --> 03:30.125
have their investments concentrated in the

03:30.145 --> 03:31.325
shares of a single firm.

03:31.595 --> 03:33.365
It's a firm that either they founded

03:33.545 --> 03:35.005
or family member founded.

03:35.945 --> 03:39.325
And, and that's what we're, you know, that was that last one

03:39.325 --> 03:41.925
that really got us thinking about family firms.

03:42.105 --> 03:45.365
And this decision of wealthy people to

03:45.955 --> 03:47.765
hold these very constant, you know,

03:48.145 --> 03:51.405
put all their eggs in one basket, so to speak, is what role

03:51.405 --> 03:53.605
that plays in driving, all of this.

03:54.265 --> 03:57.445
And so the premise of our research, you know, my research

03:57.445 --> 03:59.085
with Magnus is that all four

03:59.085 --> 04:01.845
of these facts are intimately connected.

04:02.475 --> 04:05.415
Uh, and, you kind of need all of them

04:05.435 --> 04:08.535
to be holding at the same time to understand,

04:08.795 --> 04:09.815
what's going on.

04:10.515 --> 04:13.295
Let me describe the model that we, we use.

04:13.795 --> 04:18.295
So, you know, before getting into our specific model,

04:18.685 --> 04:21.855
there's been research for a while now in economics to use

04:21.855 --> 04:23.655
what are called random growth models,

04:23.745 --> 04:25.095
which is something that we use.

04:25.675 --> 04:26.695
Uh, and,

04:26.695 --> 04:29.335
and so in our model, we'll have a random growth model

04:29.405 --> 04:31.695
with a, a starring role for family firms.

04:31.955 --> 04:35.455
But, you know, first, what is a random growth model?

04:35.555 --> 04:40.455
Why do we use it? So, as I mentioned, in nature,

04:41.885 --> 04:44.295
many outcomes have a normal distribution.

04:44.755 --> 04:48.215
The logic of this comes from the central limit theorem.

04:48.525 --> 04:49.735
This idea that you have

04:49.735 --> 04:52.255
to learn in an introductory statistics course,

04:52.325 --> 04:56.495
that if you add up a bunch of independent random variables

04:56.595 --> 05:00.295
and take the average, you know, as long as most of the time

05:00.875 --> 05:02.015
the distribution of

05:02.015 --> 05:04.895
that average is gonna be normally distributed.

05:05.555 --> 05:07.935
And so we see that with height or IQ

05:08.075 --> 05:09.855
or many natural phenomenon.

05:10.435 --> 05:12.775
And so wealth is different,

05:13.355 --> 05:16.575
but it's also interesting, you know, firm size is different.

05:16.635 --> 05:20.055
The distribution of firms, the biggest firms also are,

05:20.195 --> 05:22.215
you know, have a Pareto distribution

05:22.595 --> 05:25.535
and city sizes seems to have a Pareto distribution.

05:25.875 --> 05:28.815
So economists in a number of areas have been using

05:29.395 --> 05:31.455
random growth models as, you know,

05:31.455 --> 05:34.135
basically statistical processes that produce

05:34.755 --> 05:38.335
Pareto distributions as an, as a natural outcome

05:38.755 --> 05:41.815
for studying, you know, both the evolution of city sizes,

05:41.885 --> 05:44.175
firm sizes, the wealth distribution,

05:44.715 --> 05:45.935
and some other phenomenon.

05:45.955 --> 05:48.575
So we're in that class of research.

05:49.355 --> 05:52.975
Our specific model is that we have,

05:53.935 --> 05:57.885
families live forever, but they can be in one of two states.

05:58.985 --> 06:02.525
You know, at any point in time, 95%

06:02.525 --> 06:04.805
of the families don't have their own business.

06:04.955 --> 06:06.325
They, you know, work for a living

06:07.185 --> 06:10.365
and they invest their portfolio the way

06:10.365 --> 06:12.525
that a financial advisor would tell you

06:12.525 --> 06:13.645
to invest in a portfolio.

06:13.715 --> 06:16.845
They, they put it in a diversified set of securities.

06:17.505 --> 06:19.205
So we model the wealth process

06:19.425 --> 06:22.045
or the evolution of wealth for what, you know,

06:22.365 --> 06:25.845
families when they're diversified is having some mean rate

06:25.915 --> 06:28.045
that would be reflective of average rates

06:28.065 --> 06:29.805
of return on investments

06:30.025 --> 06:32.925
and, you know, the propensity of families

06:32.945 --> 06:34.925
to save rather than consume.

06:35.585 --> 06:39.445
But they have very small idiosyncratic, variants

06:39.465 --> 06:42.125
of their, returns, which will mean

06:42.125 --> 06:44.645
that when a family has a diversified portfolio,

06:45.315 --> 06:47.885
it's unlikely that it's gonna move very far up

06:47.885 --> 06:49.685
or down in the wealth distribution.

06:50.655 --> 06:54.045
Now, on the other hand, at any point in time,

06:54.185 --> 06:58.605
we model about 5% of families as having a family firm.

06:58.745 --> 07:01.165
So you envision somebody in your family comes up

07:01.165 --> 07:03.845
with a brilliant business idea, they're an entrepreneur,

07:03.845 --> 07:06.965
they start this, and the family puts basically all

07:06.965 --> 07:08.885
of their wealth in that venture.

07:09.855 --> 07:12.595
And what that means is they have all their eggs in one

07:12.595 --> 07:16.395
basket, and now they have a very high,

07:16.865 --> 07:19.275
idiosyncratic risk in their returns.

07:19.465 --> 07:21.515
This venture might pan out, it might not.

07:22.455 --> 07:25.715
And so, that's kind of the basics

07:25.715 --> 07:26.995
of the statistical model.

07:27.735 --> 07:32.155
And, then we have families transitioning from being,

07:32.155 --> 07:35.595
having a diversified portfolio to a family firm

07:35.695 --> 07:40.395
and back looking at, you know, how long does a family tend

07:40.395 --> 07:42.995
to run a firm on average there's some rate.

07:43.215 --> 07:45.555
And then they would, you know, maybe the, the, the,

07:46.335 --> 07:49.195
the firm fails or maybe they sell the firm

07:49.455 --> 07:51.155
and then they go on to go back

07:51.155 --> 07:52.835
to having a diversified portfolio.

07:53.695 --> 07:54.875
And, and so

07:55.305 --> 07:57.715
that parameter ends up being pretty important

07:57.785 --> 08:01.435
because, as I'll describe, so

08:01.435 --> 08:03.315
that's the basic structure of the model

08:03.315 --> 08:07.195
that we have a small set of families that are taking kind

08:07.235 --> 08:09.475
of extremely risky betts with their wealth.

08:09.735 --> 08:11.115
And then, you know, almost like

08:11.115 --> 08:12.275
they're buying lottery tickets.

08:13.895 --> 08:17.595
And then most families have, you know, very,

08:18.845 --> 08:19.895
very smooth returns.

08:19.895 --> 08:21.775
They don't have a lot of retic risk in their,

08:21.775 --> 08:22.855
in their returns to their wealth.

08:23.855 --> 08:24.915
Our first result,

08:24.975 --> 08:26.755
and perhaps the most important

08:27.575 --> 08:31.235
is just a mathematical one in the framework that,

08:32.395 --> 08:36.695
that for the model to allow the shape

08:36.755 --> 08:39.615
of the top of the wealth distribution to change rapidly

08:40.075 --> 08:42.175
to go, you know, say from the type

08:42.555 --> 08:45.655
of wealth concentration the US had in the 1960s

08:45.795 --> 08:48.775
and seventies to what it has now over the course of say,

08:49.045 --> 08:53.525
four or five decades, the model necessarily

08:54.025 --> 08:57.595
has to have, people

08:57.695 --> 09:00.395
who can move from the bottom of the wealth distribution

09:00.415 --> 09:02.635
to the top of the wealth distribution quickly,

09:03.345 --> 09:07.075
that those two mathematical properties of the model

09:07.615 --> 09:08.995
are necessarily linked.

09:09.455 --> 09:12.915
So in other words, if you set parameters of the model so

09:12.915 --> 09:16.325
that nobody is really taking big risks,

09:16.865 --> 09:18.845
you won't get self-made billionaires,

09:18.865 --> 09:20.205
but you'll get a very static

09:20.825 --> 09:23.965
or slow moving shape at the top of the wealth distribution.

09:24.425 --> 09:27.205
On the other hand, if you set parameters so that the,

09:27.905 --> 09:32.405
you do have rapid movement of those taking these risky betts

09:32.905 --> 09:36.045
and becoming self-made, then you can account

09:36.105 --> 09:37.845
for the rapid dynamics

09:37.865 --> 09:39.445
or the rapid changes in the shape

09:39.445 --> 09:41.165
of the wealth distribution that we've seen.

09:41.625 --> 09:44.445
So we think of that conceptually as kind of linking

09:44.505 --> 09:45.565
to observations

09:46.155 --> 09:48.125
that might have informally been put together,

09:48.185 --> 09:51.405
but saying they're mathematically necessarily related.

09:52.185 --> 09:55.325
And then the second is a kind of a more quantitative result.

09:55.945 --> 09:59.245
So, you know, what would you set the parameters

09:59.245 --> 10:00.965
of the model two there?

10:00.965 --> 10:03.445
We take as guidance, you know, to ask

10:03.585 --> 10:06.005
how much idiosyncratic risk are people

10:06.005 --> 10:07.325
with family firms taking?

10:08.185 --> 10:12.205
We use, data on the returns of individual stocks.

10:12.545 --> 10:15.645
So you could imagine if you as an investor just went

10:15.645 --> 10:18.565
and put all your money in Exxon or Tesla

10:18.825 --> 10:21.085
or any individual stock, what type

10:21.085 --> 10:23.325
of idiosyncratic return risk would you

10:23.325 --> 10:24.725
face that's very high.

10:24.795 --> 10:27.565
It's probably 40 to 50% standard deviation

10:27.565 --> 10:29.205
of returns in a given year.

10:29.985 --> 10:33.445
And then, and then there's also wealth data from

10:33.525 --> 10:35.405
Scandinavian tax registries.

10:35.405 --> 10:37.325
They have wealth taxes, or they had them

10:37.325 --> 10:38.805
for a while in Scandinavia,

10:39.145 --> 10:40.885
so they have more comprehensive data there.

10:40.915 --> 10:43.445
That also says that individuals

10:43.445 --> 10:47.365
who own their own firms typically have extremely volatile

10:47.835 --> 10:49.165
returns to their wealth.

10:49.785 --> 10:53.125
And so, you know, we then in the paper used the model

10:53.185 --> 10:56.765
to say, yes, with a very plausible set of parameters

10:56.765 --> 10:58.605
and amount of risk taking, we can account

10:58.625 --> 11:00.045
for the concentration of wealth

11:00.065 --> 11:02.005
and the changes in concentration of wealth

11:02.305 --> 11:04.925
and the portion of self-made, you know,

11:05.035 --> 11:08.085
very wealthy people, that we see in the data.

11:08.705 --> 11:13.285
So one, I might think at a, at an individual's level,

11:14.065 --> 11:18.075
if you were to be a young person who said,

11:18.275 --> 11:20.795
I really wanna become a billionaire, so

11:21.095 --> 11:22.435
so what's your best strategy?

11:23.095 --> 11:27.075
So we would say, you should ignore your financial advisor

11:27.855 --> 11:30.755
and put all of your wealth in one stock

11:31.015 --> 11:34.115
and hope that that pans out so that, you know, but,

11:34.135 --> 11:35.675
but that's tongue of tongue in cheek.

11:35.695 --> 11:37.235
That's, that's you, you know,

11:37.235 --> 11:39.435
ignore your financial advisor if you want to be wealthy.

11:39.735 --> 11:42.475
Uh, most of you won't make it, but somebody will.

11:43.135 --> 11:45.875
Um, no, but more seriously, we think one

11:45.875 --> 11:49.515
of the most important takeaways from the research is that

11:50.565 --> 11:54.645
there's a fundamental connection between,

11:55.975 --> 11:58.795
how we as a society, finance firms,

11:59.935 --> 12:01.395
and the creation of great wealth.

12:02.215 --> 12:04.475
So I've kind of stressed that it's odd

12:05.405 --> 12:08.855
that we see these families maintaining these concentrated,

12:09.235 --> 12:11.055
you know, investments in their firms

12:11.245 --> 12:14.135
because that flies in the face of the advice

12:14.205 --> 12:16.575
that you should have a diversified portfolio,

12:16.715 --> 12:18.575
and it means they're bearing quite a lot

12:18.575 --> 12:20.815
of idiosyncratic risk unnecessarily.

12:21.915 --> 12:24.855
And, you know, thinking more broadly in our economy,

12:24.995 --> 12:26.535
we have many very large firms

12:27.115 --> 12:28.695
and we have many entrepreneurs.

12:29.435 --> 12:31.135
And you could ask, does

12:31.135 --> 12:35.015
that necessarily mean we have great concentration of wealth?

12:35.715 --> 12:37.175
And, and we would say no.

12:37.835 --> 12:42.055
You, you see many large firms have a broad base of investors

12:42.285 --> 12:43.375
with no individual,

12:43.515 --> 12:46.535
no single individual having a large portion of the shares.

12:47.035 --> 12:49.135
And you know, they're probably owned by an s

12:49.135 --> 12:51.935
and p 500 index fund that then in turn is owned

12:52.035 --> 12:54.415
by many people with retirement accounts.

12:55.035 --> 12:58.015
And so you could certainly have large firms

12:59.165 --> 13:02.495
without any individual owner being, you know,

13:02.495 --> 13:03.575
fantastically wealthy.

13:04.475 --> 13:07.095
And on the other side with entrepreneurs, you think,

13:07.215 --> 13:09.095
you know, necessarily they have

13:09.095 --> 13:11.135
to invest their wealth in it when they get started.

13:11.955 --> 13:16.335
But there's a whole process for entrepreneurs to sell out

13:16.435 --> 13:19.375
of their business as it grows in the United States.

13:19.435 --> 13:21.535
We call this middle market investment banking.

13:21.685 --> 13:25.335
It's a whole industry to have, you know,

13:25.335 --> 13:28.295
investment bankers help you sell your firm when it gets

13:28.315 --> 13:30.415
to be worth a hundred million, 200 million,

13:30.505 --> 13:32.935
maybe 500 million in the middle market range.

13:33.555 --> 13:35.415
So you end up, you know, wealthy,

13:35.675 --> 13:38.815
but not with tens of billions of dollars.

13:39.475 --> 13:43.815
And so, you know, the, the question we think is,

13:44.035 --> 13:46.495
you know, the, the, the deeper question that needs

13:46.495 --> 13:50.585
to be explored is what, why do we find

13:51.095 --> 13:53.785
entrepreneurs who don't take advantage

13:54.045 --> 13:57.425
of the middle market opportunity to sell out, you know,

13:57.575 --> 14:01.185
sell out and diversify their firms, but instead hang on.

14:02.415 --> 14:04.355
You know, as the firm becomes gigantic.

14:04.655 --> 14:08.035
And, and, and many cases this passes down for generations.

14:08.235 --> 14:09.995
I mean, you see the Waltons doing it.

14:10.335 --> 14:12.595
The Pritzkers in the United States are the Hyatt.

14:12.785 --> 14:14.955
They, they made their money from Hyatt hotels.

14:15.375 --> 14:17.555
You know, you see this huge amount of wealth

14:17.795 --> 14:21.275
'cause they're following this and we think that

14:22.225 --> 14:25.235
this financial structure is amenable to policy.

14:25.975 --> 14:28.955
So Magnus in his, actually in his job market paper

14:28.955 --> 14:32.515
that he's coming out with, is focusing on what's the role

14:32.655 --> 14:34.315
of the rise of venture capital

14:34.975 --> 14:37.875
in changing the wealth distribution, the concentration

14:37.875 --> 14:39.275
of top wealth in the United States.

14:40.415 --> 14:44.875
So the rise of venture capital is very much the result of,

14:45.655 --> 14:47.395
reforms

14:47.395 --> 14:50.755
of our pension system in the 1970s. So that's where

14:50.755 --> 14:53.405
Venture capitalists, they get it from pension funds is

14:53.405 --> 14:54.765
where they get the money to invest.

14:55.385 --> 14:57.325
And then of course, you know, we've seen

14:58.225 --> 15:03.005
all these tech firms that go from nothing to, you know,

15:03.085 --> 15:05.805
hundreds of billions of dollars in a short period of time,

15:06.335 --> 15:08.765
often backed by venture capital money.

15:09.345 --> 15:12.685
So, I mean, I, I don't have time to get into the details

15:12.745 --> 15:13.925
of Magnus results,

15:14.025 --> 15:16.765
but that's an example of a, of, of beginning

15:16.765 --> 15:18.765
to think more deeply about

15:18.825 --> 15:22.845
how policy changes the financial system that then changes,

15:23.625 --> 15:26.785
you know, do we create, you know, great fortunes

15:26.845 --> 15:28.465
and to what extent do we do so,

15:29.125 --> 15:31.425
so we certainly think there's a lot more work to be done on

15:31.425 --> 15:31.865
that question.

00:00 - One of the most striking features of capitalist economies, you know, around the world is that the very wealthy holds so much wealth. And, this is readily apparent. If you look at any list of top billionaires, you'll see people who these days have hundreds of billions

00:18 - of dollars at the very top. And this is in a world where the average US household, I think the median US household has about $200,000 in net worth. So the differences between the median and the very top are just, astronomical. And so this simple observation raises the question of, you know, what is it about capitalist economies that generates this extreme concentration of wealth? And so my research, you know, I'm working with Magnus, ie.

00:48 - Who's actually a graduate student coming out of London School of Economics in Finance, is part of a rapidly growing literature on that question of how it is that we generate this great concentration of wealth in capitalist economies. And, and not only that, but it's looking at how has the concentration of wealth changed over time in the last century in the United States and in many other countries. And so what Magnus

01:16 - and I are focused on is the role of what we call family firms in driving the accumulation of great wealth and allowing it to change dramatically over time. And before getting into the details of like what we do, I just wanted to review kind of four facts that motivate us to take the approach we do. So I've already discussed the first fact is that there's extreme concentration of wealth, and, and that the top of the wealth distribution looks like it's,

01:46 - has a Pareto distribution, you know, so it's, it many things in nature have a normal distribution or a log normal distribution, but it stands out for wealth at it's Pareto. And that's how you, you know, that's characteristic of this extreme right tail. But the second fact is the shape of the top of the wealth distribution has actually changed relatively rapidly over the last a hundred years in the United States. It was very high in the 1920s.

02:16 - You know, the, we, we, it fell dramatically during the Great Depression and World War ii. It stayed kinda low until the 1970s or 1980s, and has now the concentration has now climbed back up again. There's debate as to whether we're at the 1920s level, but we're certainly at a very high level of wealth concentration. But then the, the next two facts are really about individuals. So one striking feature of a capitalist economy is many

02:46 - of the top fortunes are self-made. So in the Forbes 400, you see roughly two thirds of the Forbes 400 came from, they might have been middle class or upper middle class, but now They're extremely wealthy, in, in a single lifetime. And I, I stress that because, you know, a traditional notion of aristocratic wealth, you know, persists for generations and takes a long time to accumulate it.

03:13 - And then the fourth fact that really struck us is that if you look at the portfolios, the wealth holdings of the very wealthy, the vast majority of them have their investments concentrated in the shares of a single firm. It's a firm that either they founded or family member founded. And, and that's what we're, you know, that was that last one that really got us thinking about family firms. And this decision of wealthy people to

03:45 - hold these very constant, you know, put all their eggs in one basket, so to speak, is what role that plays in driving, all of this. And so the premise of our research, you know, my research with Magnus is that all four of these facts are intimately connected. and, you kind of need all of them to be holding at the same time to understand, what's going on. Let me describe the model that we, we use.

04:13 - So, you know, before getting into our specific model, there's been research for a while now in economics to use what are called random growth models, which is something that we use. and, and so in our model, we'll have a random growth model with a, a starring role for family firms. But, you know, first, what is a random growth model? Why do we use it? So, as I mentioned, in nature, many outcomes have a normal distribution.

04:44 - The logic of this comes from the central limit theorem. This idea that you have to learn in an introductory statistics course, that if you add up a bunch of independent random variables and take the average, you know, as long as most of the time the distribution of that average is gonna be normally distributed. And so we see that with height or IQ or many natural phenomenon. And so wealth is different,

05:13 - but it's also interesting, you know, firm size is different. The distribution of firms, the biggest firms also are, you know, have a Pareto distribution and city sizes seems to have a Pareto distribution. So economists in a number of areas have been using random growth models as, you know, basically statistical processes that produce Pareto distributions as an, as a natural outcome for studying, you know, both the evolution of city sizes, firm sizes, the wealth distribution,

05:44 - and some other phenomenon. So we're in that class of research. Our specific model is that we have, families live forever, but they can be in one of two states. You know, at any point in time, 95% of the families don't have their own business. They, you know, work for a living and they invest their portfolio the way that a financial advisor would tell you to invest in a portfolio.

06:13 - They, they put it in a diversified set of securities. So we model the wealth process or the evolution of wealth for what, you know, families when they're diversified is having some mean rate that would be reflective of average rates of return on investments and, you know, the propensity of families to save rather than consume. But they have very small idiosyncratic, variants of their, returns, which will mean

06:42 - that when a family has a diversified portfolio, it's unlikely that it's gonna move very far up or down in the wealth distribution. Now, on the other hand, at any point in time, we model about 5% of families as having a family firm. So you envision somebody in your family comes up with a brilliant business idea, they're an entrepreneur, they start this, and the family puts basically all of their wealth in that venture. And what that means is they have all their eggs in one

07:12 - basket, and now they have a very high, idiosyncratic risk in their returns. This venture might pan out, it might not. And so, that's kind of the basics of the statistical model. And, then we have families transitioning from being, having a diversified portfolio to a family firm and back looking at, you know, how long does a family tend to run a firm on average there's some rate. And then they would, you know, maybe the, the, the,

07:46 - the firm fails or maybe they sell the firm and then they go on to go back to having a diversified portfolio. And, and so that parameter ends up being pretty important because, as I'll describe, so that's the basic structure of the model that we have a small set of families that are taking kind of extremely risky betts with their wealth. And then, you know, almost like

08:11 - they're buying lottery tickets. And then most families have, you know, very, very smooth returns. They don't have a lot of retic risk in their, in their returns to their wealth. Our first result, and perhaps the most important is just a mathematical one in the framework that, that for the model to allow the shape of the top of the wealth distribution to change rapidly

08:40 - to go, you know, say from the type of wealth concentration the US had in the 1960s and seventies to what it has now over the course of say, four or five decades, the model necessarily has to have, people who can move from the bottom of the wealth distribution to the top of the wealth distribution quickly, that those two mathematical properties of the model are necessarily linked. So in other words, if you set parameters of the model so

09:12 - that nobody is really taking big risks, you won't get self-made billionaires, but you'll get a very static or slow moving shape at the top of the wealth distribution. On the other hand, if you set parameters so that the, you do have rapid movement of those taking these risky betts and becoming self-made, then you can account for the rapid dynamics or the rapid changes in the shape of the wealth distribution that we've seen.

09:41 - So we think of that conceptually as kind of linking to observations that might have informally been put together, but saying they're mathematically necessarily related. And then the second is a kind of a more quantitative result. So, you know, what would you set the parameters of the model two there? We take as guidance, you know, to ask how much idiosyncratic risk are people with family firms taking?

10:08 - We use, data on the returns of individual stocks. So you could imagine if you as an investor just went and put all your money in Exxon or Tesla or any individual stock, what type of idiosyncratic return risk would you face that's very high. It's probably 40 to 50% standard deviation of returns in a given year. And then, and then there's also wealth data from Scandinavian tax registries.

10:35 - They have wealth taxes, or they had them for a while in Scandinavia, so they have more comprehensive data there. That also says that individuals who own their own firms typically have extremely volatile returns to their wealth. And so, you know, we then in the paper used the model to say, yes, with a very plausible set of parameters and amount of risk taking, we can account for the concentration of wealth

11:00 - and the changes in concentration of wealth and the portion of self-made, you know, very wealthy people, that we see in the data. So one, I might think at a, at an individual's level, if you were to be a young person who said, I really wanna become a billionaire, so so what's your best strategy? So we would say, you should ignore your financial advisor and put all of your wealth in one stock and hope that that pans out so that, you know, but,

11:34 - but that's tongue of tongue in cheek. That's, that's you, you know, ignore your financial advisor if you want to be wealthy. most of you won't make it, but somebody will. no, but more seriously, we think one of the most important takeaways from the research is that there's a fundamental connection between, how we as a society, finance firms, and the creation of great wealth. So I've kind of stressed that it's odd

12:05 - that we see these families maintaining these concentrated, you know, investments in their firms because that flies in the face of the advice that you should have a diversified portfolio, and it means they're bearing quite a lot of idiosyncratic risk unnecessarily. And, you know, thinking more broadly in our economy, we have many very large firms and we have many entrepreneurs. And you could ask, does

12:31 - that necessarily mean we have great concentration of wealth? And, and we would say no. You, you see many large firms have a broad base of investors with no individual, no single individual having a large portion of the shares. And you know, they're probably owned by an s and p 500 index fund that then in turn is owned by many people with retirement accounts. And so you could certainly have large firms without any individual owner being, you know,

13:02 - fantastically wealthy. And on the other side with entrepreneurs, you think, you know, necessarily they have to invest their wealth in it when they get started. But there's a whole process for entrepreneurs to sell out of their business as it grows in the United States. We call this middle market investment banking. It's a whole industry to have, you know, investment bankers help you sell your firm when it gets to be worth a hundred million, 200 million,

13:30 - maybe 500 million in the middle market range. So you end up, you know, wealthy, but not with tens of billions of dollars. And so, you know, the, the question we think is, you know, the, the, the deeper question that needs to be explored is what, why do we find entrepreneurs who don't take advantage of the middle market opportunity to sell out, you know, sell out and diversify their firms, but instead hang on. You know, as the firm becomes gigantic.

14:04 - And, and, and many cases this passes down for generations. I mean, you see the Waltons doing it. The Pritzkers in the United States are the Hyatt. They, they made their money from Hyatt hotels. You know, you see this huge amount of wealth 'cause they're following this and we think that this financial structure is amenable to policy. So Magnus in his, actually in his job market paper that he's coming out with, is focusing on what's the role of the rise of venture capital

14:34 - in changing the wealth distribution, the concentration of top wealth in the United States. So the rise of venture capital is very much the result of, reforms of our pension system in the 1970s. So that's where Venture capitalists, they get it from pension funds is where they get the money to invest. And then of course, you know, we've seen all these tech firms that go from nothing to, you know, hundreds of billions of dollars in a short period of time,

15:06 - often backed by venture capital money. So, I mean, I, I don't have time to get into the details of Magnus results, but that's an example of a, of, of beginning to think more deeply about how policy changes the financial system that then changes, you know, do we create, you know, great fortunes and to what extent do we do so, so we certainly think there's a lot more work to be done on that question.

Faculti AI can make mistakes