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

Reaching for yield and the housing market

View more insights in Social Sciences - Economics - Financial Economics

View more figures from Erasmus University Rotterdam

View the page of Matthijs Korevaar

Read the Study

Do investors reach for yield when interest rates are low and does this behavior affect the housing market? Matthijs Korevaar suggests the reach-for-yield behavior of wealthy investors resulted in a large boom and bust in house prices and major changes in rental yields.

Image courtesy of interviewee. January 31, 2024

Generating Workspace

Export as PDF

The speaker discusses their research interest in the effects of low interest rates on investment behaviors, particularly during the 2010s when interest rates were historically low. They explore how this environment encouraged investors, particularly those relying on income such as retirees, to shift their money from safer assets like government bonds to riskier assets like real estate.

The speaker references a historical case from 18th century Holland, where low interest rates (around 2.5% after tax) and a lack of government bond issuance coincided with a boom in housing prices. They analyze data from that period, including property transactions and estate taxes, to assess whether wealthy individuals altered their investments towards real estate in response to declining bond yields.

The findings indicate that as bond interest rates fell, wealthy investors began to favor real estate, driving up property prices significantly. This video video shift in investment patterns contributed to wealth concentration and affected home ownership rates. The research suggests that similar trends observed in historical data may be relevant to contemporary issues surrounding affordable housing and wealth inequality, especially given the prolonged low interest rate policies of recent years.

The overall implication is that low interest rates can lead to significant changes in investment behavior, disproportionately benefitting wealthier individuals at the expense of home ownership opportunities for others in society.

Language:

Export as PDF

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

00:00.245 --> 00:03.285
I started actually interested in being, in this topic,

00:03.625 --> 00:07.365
in the, in the 2000 tens when interest rates were

00:07.605 --> 00:09.725
reaching extremely low levels, partic

00:09.725 --> 00:11.725
we saw unconventional monetary policy,

00:11.865 --> 00:15.165
so central banks were buying up securities,

00:15.805 --> 00:17.725
interest rates were effectively close to zero.

00:18.305 --> 00:20.365
Uh, and there was sort of a question coming up,

00:20.365 --> 00:23.685
whether the fact that you effectively make no interest if

00:23.685 --> 00:25.325
you put your money in a government bond

00:25.665 --> 00:27.765
or if you put it on a savings account, was sort

00:27.765 --> 00:29.845
of leading investors,

00:30.105 --> 00:32.685
to shift their money towards risky assets.

00:33.065 --> 00:36.485
Um, and what you saw already in some sort of,

00:36.485 --> 00:37.845
some existing literature is

00:37.845 --> 00:39.285
that people looked at equity markets

00:39.425 --> 00:42.125
and there's some evidence that sort of investors,

00:42.125 --> 00:43.525
particularly sort of the type

00:43.525 --> 00:45.005
of investor that lives of income.

00:45.065 --> 00:48.165
So you can think of someone who is approaching retirement

00:48.185 --> 00:52.045
and wants to, effectively consume dividends.

00:52.045 --> 00:53.205
So that means you have enough money,

00:53.265 --> 00:56.845
you get a dividend payment every few months or,

00:57.305 --> 00:59.125
and you can use that money to live from.

00:59.185 --> 01:01.445
So they don't wanna sell their assets,

01:01.595 --> 01:03.845
they effectively wanna live off the dividends.

01:04.265 --> 01:06.685
And that's sort of what has been, have been established.

01:06.945 --> 01:08.965
Um, but I was sort of wondering

01:08.965 --> 01:11.125
because typically actually the most common investment

01:11.385 --> 01:12.845
how people do this is real estate

01:12.845 --> 01:14.845
because real estate is much more predictable.

01:14.845 --> 01:17.125
If you own a house, you can rent it out

01:17.145 --> 01:18.805
and you can cash the rent every month,

01:18.825 --> 01:21.405
and that's much more predictable than knowing whether a

01:21.405 --> 01:23.045
company will pay out you a dividend.

01:23.745 --> 01:25.165
Uh, but that's very difficult

01:25.165 --> 01:27.685
because housing markets are extremely complicated

01:28.625 --> 01:29.845
and there's all kinds of factors.

01:30.625 --> 01:34.285
Uh, and I was working, already with sort

01:34.285 --> 01:37.205
of long-term historical housing markets data,

01:37.225 --> 01:39.845
and there was effectively one episode,

01:39.895 --> 01:42.325
which is the Netherlands in the 18th century,

01:42.505 --> 01:45.485
or to be precise, I should say Holland, which was one

01:45.485 --> 01:48.365
of the big profits of the Netherlands, which, which sort

01:48.365 --> 01:51.245
of saw extremely low interest rates, at least relative

01:51.245 --> 01:53.325
to the time during the 18th century

01:53.345 --> 01:55.685
of about 2.5% after taxes.

01:55.985 --> 01:58.245
And, and just to, to tell that sort of the last,

01:58.465 --> 02:01.085
the next time that happened after the 18th century was

02:01.085 --> 02:02.085
after 2010.

02:02.105 --> 02:04.925
So it sort of took two, three centuries for interest rates

02:04.985 --> 02:06.085
to come to very low levels.

02:06.625 --> 02:08.925
And my, my, my interest was sort of to look, well,

02:08.925 --> 02:11.405
is there any evidence for, reaching

02:11.405 --> 02:12.845
for yield in that particular context?

02:12.845 --> 02:15.125
Because that's when interest rates were extremely low

02:15.385 --> 02:17.925
and I had all kinds of data that sort of could allow me

02:17.925 --> 02:19.125
to look whether that was going on.

02:19.545 --> 02:20.805
So what happened, established?

02:20.805 --> 02:22.805
So one of the first things I mentioned is there have been

02:22.805 --> 02:24.925
studies looking at sort of, living

02:24.985 --> 02:26.485
of income in the equity market.

02:26.745 --> 02:29.445
So, so, so, so, so, so trying to think, you know,

02:29.465 --> 02:33.405
do we see people that are older or follow this kind of rule?

02:33.405 --> 02:35.485
Because if you ask people, many people say,

02:35.565 --> 02:37.205
I just want a live of income when I'm old.

02:37.285 --> 02:38.365
I want to have saved enough

02:38.635 --> 02:40.565
that I can simply live of the dividends.

02:40.625 --> 02:41.925
And people have established that

02:41.925 --> 02:44.325
that is the case very much today in history.

02:45.105 --> 02:49.055
Um, and as a result of that, there,

02:49.055 --> 02:51.255
there were quite some papers establishing that already,

02:51.315 --> 02:53.775
but it's very difficult to figure out

02:53.775 --> 02:55.055
what the effect is on prices.

02:55.555 --> 02:57.335
Uh, and the reason is why is that?

02:57.335 --> 02:59.570
Because prices move for all kinds of reasons.

03:00.145 --> 03:02.485
And the second thing is there was very little in real

03:02.485 --> 03:05.725
estate, and there, there had been sort of some sense,

03:06.065 --> 03:09.605
I think, Isabel Schnabel, a board member

03:09.605 --> 03:11.725
of the ECB was sort of suggesting that sort

03:11.725 --> 03:12.725
of this low interest rates

03:13.105 --> 03:16.285
and volume that the ECB had created and this, and,

03:16.285 --> 03:19.205
and these purchases they were buying, could sort

03:19.205 --> 03:21.165
of have lead to, to higher house prices.

03:21.425 --> 03:23.885
But there was absolutely no literature on that at all.

03:24.505 --> 03:25.525
Um,

03:25.665 --> 03:27.725
and so that was something that I, that I wanted

03:27.725 --> 03:30.445
to contribute to specifically the institutional setting in

03:30.445 --> 03:32.685
the papers is mostly what has been published is looking at

03:32.685 --> 03:33.765
the United States, of course.

03:34.425 --> 03:36.445
And so that's where there has been a lot, a lot of work,

03:36.585 --> 03:39.765
and particularly also related to the fact that sort of,

03:39.765 --> 03:43.565
in the United States, today the United States are

03:43.565 --> 03:45.405
of the leading economy,

03:45.545 --> 03:46.685
and what that means is

03:46.685 --> 03:49.405
that United States government debt is also

03:49.405 --> 03:50.525
extremely high in demand.

03:51.265 --> 03:53.925
Uh, and debt might actually push down the interest rates

03:53.985 --> 03:56.165
or US government debt even lower than

03:56.165 --> 03:57.285
what they already were, right?

03:57.305 --> 04:00.005
So this is not only a story of the ECB, which I refer to,

04:00.005 --> 04:02.125
but also the United States, which is the biggest economy.

04:02.985 --> 04:05.245
Um, and that sort of special status

04:05.305 --> 04:07.205
of United States government, that was something

04:07.205 --> 04:08.525
that was also present in Holland.

04:08.705 --> 04:11.405
Of course, by then markets were not that international,

04:11.745 --> 04:14.565
but Holland debt was sort of perceived as sort of the best

04:14.585 --> 04:16.165
and safest in the world at the time,

04:16.185 --> 04:19.165
and it was sort of the main saving asset to invest in.

04:19.905 --> 04:22.645
And partially as a result of that, interest rates were also

04:22.645 --> 04:23.845
of course extremely low.

04:23.865 --> 04:25.845
But then when interest rates go down even more,

04:26.345 --> 04:29.125
you might get into this sort of reaching for yield type

04:29.125 --> 04:30.845
of behavior that, that I'm interested.

04:30.845 --> 04:32.845
And what I mean by that is sort of the idea

04:32.845 --> 04:34.565
that when interest rates are very low,

04:35.155 --> 04:38.245
that actually investors start putting, allocating more

04:38.245 --> 04:39.645
of their money to risky assets.

04:39.665 --> 04:42.045
And there's been one experiment, for example, in the review

04:42.045 --> 04:44.685
of financial studies where they looked at participants in a

04:44.685 --> 04:46.805
lab, are they going to allocate more

04:46.805 --> 04:49.045
of their wealth towards risky assets?

04:49.185 --> 04:51.325
So think of equities, think of,

04:51.325 --> 04:53.525
real estate when interest rates are low,

04:53.525 --> 04:55.765
and at least in the lab, people seem to be doing that,

04:55.825 --> 04:58.285
but we don't, do not yet have good evidence.

04:58.305 --> 05:00.765
And that's why I went to sort of historical Holland,

05:00.765 --> 05:02.725
because you have this historical context

05:03.185 --> 05:04.285
of very low interest rates.

05:04.825 --> 05:08.325
Uh, and I also have, I was able to obtain sort of data on

05:08.325 --> 05:10.485
where people then end up storing their money,

05:10.495 --> 05:12.445
which could allow me to test that empirically.

05:12.865 --> 05:16.605
So what I did is I, collected a lot of archival data,

05:16.945 --> 05:20.525
partially held by the Amsterdam City archives,

05:20.865 --> 05:23.645
and doing also a lot of data collection myself.

05:23.825 --> 05:27.165
So I think the Netherlands is sort of unique in the sense

05:27.235 --> 05:31.125
that, among sort of a small set of countries that a lot

05:31.125 --> 05:32.765
of good archival data have survived.

05:32.765 --> 05:35.125
So I used sort of mostly two sources.

05:35.425 --> 05:38.645
The first one are all housing transactions

05:38.715 --> 05:41.645
that happened in Amsterdam during the 17th and 18th century.

05:41.745 --> 05:43.285
So these are imagine sort

05:43.285 --> 05:45.285
of these old handwritten records which

05:45.445 --> 05:46.645
volunteers transcribed.

05:47.065 --> 05:48.925
And I was able to obtain the data from the

05:48.925 --> 05:50.125
Amsterdam CD archives.

05:50.585 --> 05:54.765
Now, the second database, which I collected myself,

05:55.075 --> 05:56.885
were estate tax records. So

05:56.885 --> 05:58.485
These were records when people died,

05:58.485 --> 05:59.765
they had to pay estate tax.

06:00.425 --> 06:05.125
And when you didn't inherit, towards a child, you had

06:05.125 --> 06:06.285
to pay significant tax.

06:06.305 --> 06:08.645
And that recorded every individual assets.

06:09.145 --> 06:11.365
So that allowed me to ob observe for everyone

06:11.425 --> 06:14.125
who died in this period who left an estate that was staed

06:14.125 --> 06:15.845
where they invested their money in.

06:16.505 --> 06:18.845
And that allowed me to look, do they invest in real estate

06:18.845 --> 06:21.005
or do they invest sort of in government bonds?

06:21.825 --> 06:24.805
Uh, and the final thing I did is I looked at auction sales

06:24.805 --> 06:27.645
of properties because it's very hard today to find out.

06:28.345 --> 06:31.005
Now we can find out the interest rate on a government bond

06:31.005 --> 06:32.845
that's typically pretty easy to obtain.

06:32.845 --> 06:34.365
What is the return on the house?

06:34.395 --> 06:35.805
Well, it's much higher, harder

06:35.805 --> 06:37.005
because you need to know the rent

06:37.425 --> 06:38.565
and you need to know the price.

06:39.265 --> 06:40.805
And so I went to auction records

06:41.495 --> 06:43.405
where they sold investment property,

06:43.405 --> 06:44.405
which gave me the price,

06:44.865 --> 06:46.565
the sales price at the rental price.

06:46.585 --> 06:48.725
So I had an estimate of the yield, the,

06:48.725 --> 06:49.885
the rent divided by the price.

06:49.895 --> 06:52.325
First of all, establishing is

06:52.785 --> 06:56.165
how does the yield on housing evolve over time

06:56.185 --> 06:58.285
and how does this compare to government bonds?

06:58.985 --> 07:01.085
And what we see is that there's sort of,

07:01.165 --> 07:02.285
I look at the 18th century,

07:02.485 --> 07:05.645
I look specifically at the period from 1700 to 70 15,

07:06.265 --> 07:09.765
and what you see is that from 1700 to 1714,

07:10.665 --> 07:12.645
rental yields are relatively high,

07:12.825 --> 07:14.765
and bond yields are also relatively high.

07:14.765 --> 07:17.485
But the gap between the two is sort of pretty substantial,

07:17.495 --> 07:19.885
about 2.5 percentage points.

07:20.745 --> 07:22.925
Now, what you see is between 1713

07:22.985 --> 07:24.365
and 70 14 is

07:24.365 --> 07:26.685
that actually the interest rate on government goes down,

07:26.705 --> 07:29.125
but only a little bit, only about by half a percent,

07:29.755 --> 07:33.445
whereas the rate of return on these,

07:33.915 --> 07:38.325
real estates goes down significantly by more than 2%.

07:38.665 --> 07:40.365
And so what you see is that the re the,

07:40.365 --> 07:42.525
the return on real estate measured by the yield,

07:42.545 --> 07:45.005
so the rent divided by the price compresses significantly,

07:45.705 --> 07:48.365
and then after 70 14 it goes down.

07:48.405 --> 07:49.845
So what is going on? There's enormous

07:49.955 --> 07:51.005
boom in the housing market.

07:51.055 --> 07:52.845
House prices go up substantially,

07:53.075 --> 07:55.405
whereas interest rates go down only a little bit.

07:55.585 --> 07:57.525
And now the question is what is driving that?

07:58.895 --> 08:01.515
Now what I did is I linked that to the issuance

08:01.515 --> 08:03.995
of government bonds because government bonds in 18th century

08:03.995 --> 08:06.595
Holland were only issued in times of war.

08:07.255 --> 08:09.035
And what was happening, there were two wars.

08:09.135 --> 08:13.595
One between 1,717 14 at that period when the, the spread

08:13.595 --> 08:16.395
between BO bonds and real estate was high

08:16.415 --> 08:18.955
and one between 70 40 and 70 48.

08:19.475 --> 08:20.715
Interesting enough, there were two wars

08:20.715 --> 08:21.955
and they were both related to the fact

08:21.955 --> 08:24.155
that the foreign emperor had died either

08:24.695 --> 08:27.595
the Spanish succession war, so it was the emperor of Spain

08:28.015 --> 08:29.875
or it was the Austrian, queen.

08:29.875 --> 08:31.155
And there was sort of a debate about

08:31.215 --> 08:32.835
who could success these persons.

08:33.375 --> 08:34.995
And as a result, the Netherlands,

08:34.995 --> 08:37.435
because it was sort of concerned about the balance of power

08:37.455 --> 08:39.875
and its straight interest, had to fight war abrupt,

08:39.875 --> 08:42.835
which was very costly and led them to issue a lot of bonds.

08:43.735 --> 08:46.035
Now, what happened in the peace period in between is

08:46.035 --> 08:48.035
that there was no bond issuance at all.

08:48.695 --> 08:50.315
So what I'm arguing is that because,

08:50.655 --> 08:53.395
and that's sort of what I'm showing in the data, is that

08:53.395 --> 08:54.835
because there were no bond issuance,

08:54.930 --> 08:56.645
obviously interest rates were going down,

08:57.705 --> 09:00.125
but interest rates were going down only a little bit.

09:00.545 --> 09:03.165
Why? Well, the case I make in the paper is that

09:03.165 --> 09:06.005
because these interest rates were going down, investors

09:06.145 --> 09:08.605
who were living of income, they now got less, right?

09:08.605 --> 09:11.885
If they put one euro or one gilder at the time in

09:11.885 --> 09:15.165
that investment, they would now not get 3%.

09:15.165 --> 09:16.445
They would get two point a half percent.

09:16.465 --> 09:18.325
And if you live of income of free instead

09:18.325 --> 09:19.525
of two point a half, well that's

09:19.525 --> 09:21.445
25% reduction in your income.

09:21.785 --> 09:24.125
So what could you do? Well, you could buy your property

09:24.785 --> 09:26.565
and property at much higher yields.

09:27.295 --> 09:28.515
And that's what I show in the data.

09:28.575 --> 09:32.635
So what I see is that the very rich, the ones that own a lot

09:32.635 --> 09:34.875
of government bonds, that they shift

09:35.855 --> 09:37.395
and they sell these government bonds

09:37.415 --> 09:38.995
and they start buying real estate.

09:39.995 --> 09:41.655
And as a result of that, well, if a lot

09:41.655 --> 09:42.855
of people wanna own real estate,

09:42.855 --> 09:45.255
particularly the very wealthy prices start going up.

09:45.795 --> 09:46.935
And so I document that

09:46.935 --> 09:50.055
because of these shifts in issuance of government bonds,

09:50.055 --> 09:52.895
investors started shifting their wealth from bonds

09:52.925 --> 09:54.015
towards real estate.

09:55.815 --> 09:57.155
Now what does that imply?

09:58.645 --> 10:00.495
Well, this was something that was only the case

10:00.515 --> 10:02.695
for the very wealthy because those had enough money

10:02.755 --> 10:03.935
to live of income, right?

10:03.935 --> 10:05.375
This is something you only do when you're

10:05.375 --> 10:06.455
retired or when you're very rich.

10:07.155 --> 10:09.975
So what I then that I compare the portfolios of those

10:10.135 --> 10:12.935
that were very rich to those that were not so rich.

10:13.675 --> 10:16.375
And what you see is that that shift is only present

10:16.545 --> 10:18.375
among those that were living of income.

10:18.435 --> 10:21.055
So those that had very high bond holdings to begin with.

10:21.835 --> 10:24.575
And the second thing I do in the analysis is that I show

10:24.795 --> 10:26.415
who ends up buying real estate

10:26.415 --> 10:29.775
because if my hypothesis true that it's really those living

10:29.795 --> 10:32.655
of income investors that are driving up house prices,

10:32.845 --> 10:34.375
they should be the ones buying

10:35.115 --> 10:36.735
and sort of the lower income

10:36.735 --> 10:38.295
homeowners should be the ones selling.

10:38.595 --> 10:39.935
And that's what I see in the data.

10:40.475 --> 10:43.535
So individuals that own a lot of their wealth in bonds,

10:44.125 --> 10:45.615
they become much more likely

10:45.755 --> 10:48.735
to buy real estate exactly in this period when interest

10:48.735 --> 10:52.455
rates are going down and when bonds are not being issued.

10:52.595 --> 10:54.815
And so it really seems to be that those living

10:54.835 --> 10:57.935
of income investors were driving this massive boom in house

10:57.935 --> 11:00.775
prices and the corresponding reduction in the returns

11:00.795 --> 11:02.775
to real estate, at least the expected returns

11:02.775 --> 11:04.735
because house prices are obviously are going up.

11:04.755 --> 11:07.295
So if you own a house, this was great news for you.

11:07.665 --> 11:09.775
There are a few con, few important takeaways.

11:09.795 --> 11:12.495
So I, I think the first thing is that sort

11:12.495 --> 11:14.855
of in the standard finance literature, it was sort

11:14.855 --> 11:17.535
of assumed and obviously there have been some literature

11:17.535 --> 11:19.575
pushing back a little bit of that, that sort

11:19.575 --> 11:21.895
of when interest rates are known that people's preferences

11:22.115 --> 11:25.935
for a risky asset like real estate versus a safe asset like

11:25.935 --> 11:27.295
a government bond would not change.

11:28.285 --> 11:31.545
But what I showed in my paper is that, that this is not true

11:31.605 --> 11:34.505
and that actually when interest rates go down,

11:34.505 --> 11:37.425
there is more money being investment in the in real estate.

11:37.645 --> 11:38.705
And why is that the case?

11:38.775 --> 11:41.585
Well, it is the case because the very rich we can live

11:41.585 --> 11:44.105
of their capital income start buying

11:44.105 --> 11:45.305
these real estate assets.

11:46.325 --> 11:48.745
And I think that is important also for the current context

11:48.745 --> 11:50.265
because there is obviously a lot

11:50.265 --> 11:52.465
of debate about housing markets about say,

11:52.515 --> 11:54.805
young ho home ownership rates declining.

11:55.305 --> 11:58.125
And one sort of potential reason why this could be the case

11:58.585 --> 12:01.045
is because in investors will wanna live of income

12:01.065 --> 12:04.165
and will have substantial capital now buy these properties.

12:04.185 --> 12:06.405
So that's first time home buyers cannot buy.

12:06.865 --> 12:08.325
And that's also what I found evidence

12:08.325 --> 12:09.445
for the historical data

12:09.445 --> 12:12.165
because you look at sort of the distribution

12:12.165 --> 12:15.885
of housing wealth in 18th century Holland Society, it sort

12:15.885 --> 12:18.805
of increased significantly during this reach for boom.

12:18.805 --> 12:22.365
And importantly, even though the boom eventually went bust,

12:22.365 --> 12:24.165
that inequality did not go away.

12:24.585 --> 12:25.805
And some of these investors want,

12:25.805 --> 12:27.445
they own the properties they would not sell.

12:28.265 --> 12:29.925
And I think that's something that is important

12:30.045 --> 12:31.365
to take into account today.

12:31.905 --> 12:34.005
Why? Because we've seen central bank policies

12:34.005 --> 12:36.405
with extremely low interest rates for long,

12:36.505 --> 12:38.725
and my study suggests that that could lead

12:38.725 --> 12:40.805
to a higher concentration of housing wealth

12:40.805 --> 12:41.885
and lower home ownership.

12:41.945 --> 12:44.965
And in many societies, home ownership is incredibly valued.

12:45.665 --> 12:48.245
And so that I think is really the key takeaway from my work

12:48.385 --> 12:50.805
is that sort of these low interest rates could actually also

12:50.805 --> 12:53.645
have very big implication who ends up owning housing

12:53.705 --> 12:55.045
and who ends up renting housing?

12:55.745 --> 12:58.325
Uh, and sort of there's been some follow up work and, and,

12:58.325 --> 13:00.525
and there also has been now some evidence in Australia

13:00.745 --> 13:03.925
and also in Germany that this was at least correlation also

13:03.925 --> 13:05.485
going on in those countries that

13:05.585 --> 13:07.325
as interest rates were going down,

13:07.755 --> 13:10.125
very wealthy people were investing more on real estate.

13:10.185 --> 13:13.165
So this does not seem just an 18th century story.

00:00 - I started actually interested in being, in this topic, in the, in the 2000 tens when interest rates were reaching extremely low levels, partic we saw unconventional monetary policy, so central banks were buying up securities, interest rates were effectively close to zero.

00:18 - and there was sort of a question coming up, whether the fact that you effectively make no interest if you put your money in a government bond or if you put it on a savings account, was sort of leading investors, to shift their money towards risky assets. and what you saw already in some sort of, some existing literature is that people looked at equity markets and there's some evidence that sort of investors,

00:42 - particularly sort of the type of investor that lives of income. So you can think of someone who is approaching retirement and wants to, effectively consume dividends. So that means you have enough money, you get a dividend payment every few months or, and you can use that money to live from. So they don't wanna sell their assets, they effectively wanna live off the dividends. And that's sort of what has been, have been established.

01:06 - but I was sort of wondering because typically actually the most common investment how people do this is real estate because real estate is much more predictable. If you own a house, you can rent it out and you can cash the rent every month, and that's much more predictable than knowing whether a company will pay out you a dividend. but that's very difficult because housing markets are extremely complicated

01:28 - and there's all kinds of factors. and I was working, already with sort of long-term historical housing markets data, and there was effectively one episode, which is the Netherlands in the 18th century, or to be precise, I should say Holland, which was one of the big profits of the Netherlands, which, which sort of saw extremely low interest rates, at least relative to the time during the 18th century of about 2.5% after taxes.

01:55 - And, and just to, to tell that sort of the last, the next time that happened after the 18th century was after 2010. So it sort of took two, three centuries for interest rates to come to very low levels. And my, my, my interest was sort of to look, well, is there any evidence for, reaching for yield in that particular context? Because that's when interest rates were extremely low and I had all kinds of data that sort of could allow me

02:17 - to look whether that was going on. So what happened, established? So one of the first things I mentioned is there have been studies looking at sort of, living of income in the equity market. So, so, so, so, so, so trying to think, you know, do we see people that are older or follow this kind of rule? Because if you ask people, many people say, I just want a live of income when I'm old. I want to have saved enough

02:38 - that I can simply live of the dividends. And people have established that that is the case very much today in history. and as a result of that, there, there were quite some papers establishing that already, but it's very difficult to figure out what the effect is on prices. and the reason is why is that? Because prices move for all kinds of reasons. And the second thing is there was very little in real

03:02 - estate, and there, there had been sort of some sense, I think, Isabel Schnabel, a board member of the ECB was sort of suggesting that sort of this low interest rates and volume that the ECB had created and this, and, and these purchases they were buying, could sort of have lead to, to higher house prices. But there was absolutely no literature on that at all. and so that was something that I, that I wanted

03:27 - to contribute to specifically the institutional setting in the papers is mostly what has been published is looking at the United States, of course. And so that's where there has been a lot, a lot of work, and particularly also related to the fact that sort of, in the United States, today the United States are of the leading economy, and what that means is that United States government debt is also extremely high in demand.

03:51 - and debt might actually push down the interest rates or US government debt even lower than what they already were, right? So this is not only a story of the ECB, which I refer to, but also the United States, which is the biggest economy. and that sort of special status of United States government, that was something that was also present in Holland. Of course, by then markets were not that international, but Holland debt was sort of perceived as sort of the best

04:14 - and safest in the world at the time, and it was sort of the main saving asset to invest in. And partially as a result of that, interest rates were also of course extremely low. But then when interest rates go down even more, you might get into this sort of reaching for yield type of behavior that, that I'm interested. And what I mean by that is sort of the idea that when interest rates are very low, that actually investors start putting, allocating more

04:38 - of their money to risky assets. And there's been one experiment, for example, in the review of financial studies where they looked at participants in a lab, are they going to allocate more of their wealth towards risky assets? So think of equities, think of, real estate when interest rates are low, and at least in the lab, people seem to be doing that, but we don't, do not yet have good evidence. And that's why I went to sort of historical Holland,

05:00 - because you have this historical context of very low interest rates. and I also have, I was able to obtain sort of data on where people then end up storing their money, which could allow me to test that empirically. So what I did is I, collected a lot of archival data, partially held by the Amsterdam City archives, and doing also a lot of data collection myself. So I think the Netherlands is sort of unique in the sense that, among sort of a small set of countries that a lot

05:31 - of good archival data have survived. So I used sort of mostly two sources. The first one are all housing transactions that happened in Amsterdam during the 17th and 18th century. So these are imagine sort of these old handwritten records which volunteers transcribed. And I was able to obtain the data from the Amsterdam CD archives. Now, the second database, which I collected myself,

05:55 - were estate tax records. So These were records when people died, they had to pay estate tax. And when you didn't inherit, towards a child, you had to pay significant tax. And that recorded every individual assets. So that allowed me to ob observe for everyone who died in this period who left an estate that was staed where they invested their money in. And that allowed me to look, do they invest in real estate

06:18 - or do they invest sort of in government bonds? and the final thing I did is I looked at auction sales of properties because it's very hard today to find out. Now we can find out the interest rate on a government bond that's typically pretty easy to obtain. What is the return on the house? Well, it's much higher, harder because you need to know the rent and you need to know the price. And so I went to auction records

06:41 - where they sold investment property, which gave me the price, the sales price at the rental price. So I had an estimate of the yield, the, the rent divided by the price. First of all, establishing is how does the yield on housing evolve over time and how does this compare to government bonds? And what we see is that there's sort of, I look at the 18th century,

07:02 - I look specifically at the period from 1700 to 70 15, and what you see is that from 1700 to 1714, rental yields are relatively high, and bond yields are also relatively high. But the gap between the two is sort of pretty substantial, about 2.5 percentage points. Now, what you see is between 1713 and 70 14 is that actually the interest rate on government goes down, but only a little bit, only about by half a percent,

07:29 - whereas the rate of return on these, real estates goes down significantly by more than 2%. And so what you see is that the re the, the return on real estate measured by the yield, so the rent divided by the price compresses significantly, and then after 70 14 it goes down. So what is going on? There's enormous boom in the housing market. House prices go up substantially, whereas interest rates go down only a little bit.

07:55 - And now the question is what is driving that? Now what I did is I linked that to the issuance of government bonds because government bonds in 18th century Holland were only issued in times of war. And what was happening, there were two wars. One between 1,717 14 at that period when the, the spread between BO bonds and real estate was high and one between 70 40 and 70 48. Interesting enough, there were two wars and they were both related to the fact

08:21 - that the foreign emperor had died either the Spanish succession war, so it was the emperor of Spain or it was the Austrian, queen. And there was sort of a debate about who could success these persons. And as a result, the Netherlands, because it was sort of concerned about the balance of power and its straight interest, had to fight war abrupt, which was very costly and led them to issue a lot of bonds. Now, what happened in the peace period in between is

08:46 - that there was no bond issuance at all. So what I'm arguing is that because, and that's sort of what I'm showing in the data, is that because there were no bond issuance, obviously interest rates were going down, but interest rates were going down only a little bit. Why? Well, the case I make in the paper is that because these interest rates were going down, investors who were living of income, they now got less, right? If they put one euro or one gilder at the time in

09:11 - that investment, they would now not get 3%. They would get two point a half percent. And if you live of income of free instead of two point a half, well that's 25% reduction in your income. So what could you do? Well, you could buy your property and property at much higher yields. And that's what I show in the data. So what I see is that the very rich, the ones that own a lot of government bonds, that they shift

09:35 - and they sell these government bonds and they start buying real estate. And as a result of that, well, if a lot of people wanna own real estate, particularly the very wealthy prices start going up. And so I document that because of these shifts in issuance of government bonds, investors started shifting their wealth from bonds towards real estate. Now what does that imply?

09:58 - Well, this was something that was only the case for the very wealthy because those had enough money to live of income, right? This is something you only do when you're retired or when you're very rich. So what I then that I compare the portfolios of those that were very rich to those that were not so rich. And what you see is that that shift is only present among those that were living of income. So those that had very high bond holdings to begin with.

10:21 - And the second thing I do in the analysis is that I show who ends up buying real estate because if my hypothesis true that it's really those living of income investors that are driving up house prices, they should be the ones buying and sort of the lower income homeowners should be the ones selling. And that's what I see in the data. So individuals that own a lot of their wealth in bonds, they become much more likely

10:45 - to buy real estate exactly in this period when interest rates are going down and when bonds are not being issued. And so it really seems to be that those living of income investors were driving this massive boom in house prices and the corresponding reduction in the returns to real estate, at least the expected returns because house prices are obviously are going up. So if you own a house, this was great news for you. There are a few con, few important takeaways. So I, I think the first thing is that sort

11:12 - of in the standard finance literature, it was sort of assumed and obviously there have been some literature pushing back a little bit of that, that sort of when interest rates are known that people's preferences for a risky asset like real estate versus a safe asset like a government bond would not change. But what I showed in my paper is that, that this is not true and that actually when interest rates go down, there is more money being investment in the in real estate. And why is that the case?

11:38 - Well, it is the case because the very rich we can live of their capital income start buying these real estate assets. And I think that is important also for the current context because there is obviously a lot of debate about housing markets about say, young ho home ownership rates declining. And one sort of potential reason why this could be the case is because in investors will wanna live of income and will have substantial capital now buy these properties.

12:04 - So that's first time home buyers cannot buy. And that's also what I found evidence for the historical data because you look at sort of the distribution of housing wealth in 18th century Holland Society, it sort of increased significantly during this reach for boom. And importantly, even though the boom eventually went bust, that inequality did not go away. And some of these investors want, they own the properties they would not sell.

12:28 - And I think that's something that is important to take into account today. Why? Because we've seen central bank policies with extremely low interest rates for long, and my study suggests that that could lead to a higher concentration of housing wealth and lower home ownership. And in many societies, home ownership is incredibly valued. And so that I think is really the key takeaway from my work is that sort of these low interest rates could actually also

12:50 - have very big implication who ends up owning housing and who ends up renting housing? and sort of there's been some follow up work and, and, and there also has been now some evidence in Australia and also in Germany that this was at least correlation also going on in those countries that as interest rates were going down, very wealthy people were investing more on real estate. So this does not seem just an 18th century story.

Faculti AI can make mistakes