Economic Growth, De-Population, and Macroeconomics with UPenn Econ Professor Jesus Fernández-Villaverde

Jon Hartley and Jesus Fernandez-Villaverde discuss Jesus’s career, schools of economic thought and the role of institutions in economic history, economic growth (including recent declining GDP growth rates and declining fertility), business cycles, drivers of the early 2020s inflation, dynamic stochastic general equilibrium (DSGE) and vector autoregressive (VAR) models, schools of economic thought and the role of institutions in economic history.

Recorded on November 1, 2024.

ABOUT THE SPEAKERS:

Jesus Fernandez-Villaverde is a Professor of Economics at the University of Pennsylvania, director of the Penn Institute for the Study of Markets (PISM), a Research Associate of the National Bureau of Economic Research (NBER_, and Research Affiliate of the Center for Economic and Policy Research (CEPR). He has also been a visiting Professor at Harvard, Yale, Princeton, Oxford, and Cambridge and a visiting scholar at several Federal Reserve Banks, the European Central Bank, and the Bank of Spain. His research agenda focuses on economic history, macroeconomics, and econometrics, with a focus on the computation and estimation of dynamic stochastic general equilibrium (DSGE) models.

Jon Hartley is a Research Assistant at the Hoover Institution and an economics PhD Candidate at Stanford University, where he specializes in finance, labor economics, and macroeconomics. He is also currently a Research Fellow at the Foundation for Research on Equal Opportunity (FREOPP) and a Senior Fellow at the Macdonald-Laurier Institute. Jon is also a member of the Canadian Group of Economists, and serves as chair of the Economic Club of Miami.

Jon has previously worked at Goldman Sachs Asset Management as well as in various policy roles at the World Bank, IMF, Committee on Capital Markets Regulation, US Congress Joint Economic Committee, the Federal Reserve Bank of New York, the Federal Reserve Bank of Chicago, and the Bank of Canada. 

Jon has also been a regular economics contributor for National Review Online, Forbes, and The Huffington Post and has contributed to The Wall Street Journal, The New York Times, USA Today, Globe and Mail, National Post, and Toronto Star among other outlets. Jon has also appeared on
My guest is Jesus Fernández-Villaverde,professor of Economics at the University
of Pennsylvania, a macroeconomist andfrequent visitor to Hoover.

(01:09):
Welcome, Jesus.

>> Jesus Fernández-Villaverde (01:10):
Thank you.

>> Jon Hartley (01:11):
Jesus, you grew up in Madrid, Spain.
How did you first get interestedin economics and macroeconomics?

>> Jesus Fernández-Villaverde (01:17):
Well, so I grew up in the 1980s and
at that time the Spanish economywas not doing very well,
we had high inflation, high unemployment,high government deficit.
The economy was not growing very fast.
So I guess that forsome kids who were more or
less interested in howhis country was going on,

(01:39):
started thinking about economics andmore in particular into macro was natural.
So one day I decided Iwanted to be an economist.
And here I am, like maybe 40 years later.

>> Jon Hartley (01:52):
That's funny.
I mean, you did your PhD in Minnesota and
your advisor was actually Leo Haney,a senior fellow here at Hoover.
Tell us about your path from growingup in Madrid to Minnesota and
getting your PhD there.

>> Jesus Fernández-Villaverde (02:08):
Well, so already when I was an undergraduate,
I started trying to read things a little
bit more advanced than perhapsthe average undergraduate.
And I bump into the veryfirst textbook that
Tom Sarrient brought on macroeconomics.
It's a black book,
I always joke that Tom has writtenthree textbooks in macroeconomics.

(02:32):
And I really like it.
I enjoy it a lot.
It opened like a lot of new views for me.
And he said,professor of economics at Minnesota,
because at that time Tom was at Minnesota.
So I thought, wow,I want to go to Minnesota.
Then this was before the Internet andall these type of things.
So I wasn't quite aware that Tom hadmoved, I think at the time to Chicago, but

(02:53):
nonetheless, I applied to Minnesota,they accepted me, and the rest is history.

>> Jon Hartley (02:58):
That's amazing, I mean,
what a fascinating placeto study macroeconomics.
It's produced somany famous macroeconomists,
it's really the home or one of the homesof the rational expectations revolution.
We think Tom Sargent, Ed Prescott,many others who are part of that and
many Nobel Prizes have come out of that,and a tremendous amount of

(03:22):
influence in terms of howmacroeconomic modeling is done.
Find me a macroeconomic model that'spublished today that doesn't include
some level of micro founded individualagents underline the model.
You do a lot of work ona number of things, growth,

(03:43):
economic history, solving DSG models.
In fact, I think you're the first tosolve DSG models using quantum annealers.

>> Jesus Fernández-Villaverde (03:51):
[LAUGH] >> Jon Hartley
of the Penn Initiative for the Study ofMarkets, which you're also the founder of.
Tell us what that's all about.
Okay, so a few years ago I started to grow a little
bit unhappy with the state ofundergraduate education in economics.

(04:12):
I don't think we are doinga very good job at it.
I think that we go to class and we tellstudents a lot about Lagrangians and
first order conditions andhow to solve optimization problems,
but we don't really give them an economicunderstanding of what is going on.
And I'm not against math, I use a lot ofmath in my research, but I think that math

(04:33):
must be built on the foundation ofunderstanding the economic intuition.
And I also was a little bit worried thatthe students were a little bit separated
from things like the historicalbackground of market economies,
some of the philosophical foundationsof market economies, and that
there was some scope for a center at Pennthat could provide some of those courses.

(04:56):
So I want to talk with my chair andthen with the Dean.
I was able to get a little bit of support.
Not much.
It has been a little bit of a struggle,to be honest.
And we have been able to offera lot of undergraduate courses.
I think this year we are offering six.
Some of them I really thinkare quite cool, like economics and
philosophy andhistory of financial markets.

(05:18):
And we also have a littlebit of visitors program.
So I think it's very nice,
I think it has made a big contribution tothe undergraduate education at Penn and
hopefully to a stream of graduate studentsand postdocs that help us over there.

>> Jon Hartley (05:33):
Well, I'm sure many students go into
the Penn Economics Department andWharton as well.
Appreciate that.
I want to get a bit into your research and
thinking around thingsstarting with growth.
So you have a recent paper titledthe Wealth of Working Nations, and
in it you're estimating,I think GDP per capita across countries.

(05:55):
But you're also taking into accountthe age of the labor force,
which varies prettysignificantly across countries.
I think Japan,most of its population is much older.
So that distorts perhaps the sizeof its working population.

(06:17):
Or perhaps your GDP per capitafigures don't really account for
some of that,the size of its working population.
Can you tell us more about that andyour other thoughts on demographic
trends around aging populations,declining birth rates, and
obviously also declining GDP per capitagrowth rates across the developed world?

>> Jesus Fernández-Villaverde (06:41):
Of course.
So, this research came froma very simple observation.
So, you know, you stop the average personin the street and you ask them, you know,
give me an example of a rich economythat has not been doing very well for
the last 10, 20, or 25 years.
Chances are that this personreads the financial press at all,

(07:02):
they will immediately say Japan.
So, you know, out of curiosity,I started looking at Japan and I realized,
look, there are much fewer Japaneseof working age now than 25 years ago,
just because Japan was the firstrich economy that witnessed a very,
very fast drop in fertility.

(07:22):
And yes, Japanese work a little bitlonger on average than other people,
maybe 68, 69 in comparison with 65,64 in the U.S. but
even when you control forthat longer working age,
there are much fewer Japanese inworking age now than 20 years ago.
So, what I did, and this is research workwith Gustavo Ventura at Arizona State and

(07:44):
Wenjao, one of my former studentsat Tsinghua University, was to say,
let's look at the data.
And what happens if you somehow measureGDP in these units of working age
population where you try to take account,as I was saying before, of this.
And then turns out to be the casethat Japan is doing pretty well.
In fact, over the last 15 years isthe country among all the G7 that has

(08:08):
grown the fastest.
It's just that there are fewerJapanese and in comparison,
other countries that sometimesare argue are doing very well.
But once you take account that there are,for instance in the U.S,
many more Americans now than 25 years ago,well,
the performance of the United Statesis not that great.

(08:29):
Well, it's still good, butit's not as good as some people argue.
And what we are trying to say is thatthe present of Japan is the future of
the rest of the world.
I have in Britain a lotlately on demographic trends,
and the biggest demographic trendis the way I always put it,
winter, the demographic winter is coming.

(08:50):
Every country in the world, even poor and
middle income economiesare going through a very,
very fast fertility decline,which means that there is going to be this
big headwind of decreasing workingage populations all across the globe.
And that will force us to startrethinking about how we actually measure

(09:12):
economic performance.

>> Jon Hartley (09:14):
That's fascinating.
So I guess one question I have for
you when you make thisadjustment to GDP per capita and
change it to GDP per worker,I'm curious you're saying
that Japan has been growingfaster by that measure even
versus the US,i'm curious what about levels though?

(09:37):
If you were to look at GDP,I guess per worker for
getting the growth rates, is GDP perworker still much higher for the us?
Cuz when I think about theseGDP per capita trends,
at least the story that you get fromlooking at GDP per capita across

(09:59):
countries today is that mostcountries in the developed world,
so Europe, Canada,most of Western Europe, Canada and
Japan basically haven't hadany GDP per capita growth or
much GDP per capita growth sincethe global financial crisis.
And in Japan's case, even prior to that.

(10:21):
Whereas the US just keeps growing andit's very unique and
arguably maybe it's because it hasthis tech sector that's very unique.
These tech giants and culture ofentrepreneurism really maybe only exists
in the US, how does that all look whenyou look at these GDP per worker numbers?

(10:42):
In terms of levels.

>> Jesus Fernández-Villaverde (10:42):
Yes, so in terms of levels,
the US is still much richer.
Since we basically argue that once youlook at everything in working age terms or
in worker terms, there is no bigdifferences in economic growth.
It basically means that Canada,Western Europe and
Japan have not been able to close the gapwith respect to the United States.

(11:05):
And that's what I will bring in all thearguments you make about entrepreneurship
and tech and maybe a better businessenvironment, because something that
economists believe is some idea oftechnological diffusion that fine.
You develop some new wonderfultechnology on Silicon valley, but
that 10 years later, 15 years later,you should be able to implement that

(11:27):
technology in Italy orin Spain or in Germany.
And the fact that Italy, Spain or Germanyor Japan, since you asked about it,
do not seem to be able to closethe gap with the United States,
only seem to be more orless keeping up at the same level,
tells you that there is somethingdeeply wrong with those economies.
But that's a slightly different storythan the story of Japan being stagnated.

(11:50):
It's just for whatever the reason,Japan cannot close the gap with the US.
And the second thing that people shouldreally understand is if the US is
going to have in 20, 30 yearsthe current demographics of Japan,
then what we are thinking now as the very,very vital and
dynamic US economy may not look so vitaland so dynamic 20, 30 years down the road.

(12:13):
And we want to keep this in mind.

>> Jon Hartley (12:16):
And I, I appreciate a lot your focus on growth.
And you know, I think, you know, RobertLucas, you know, famously said it best,
you know,once you start thinking about growth,
you can't start thinkingabout anything else.
And it really is so important.
The difference in say GDP per capitabetween the US and say China is so stark.

(12:39):
China's maybe 1/7th the GDP per capitaof the U.S. or something to that degree.
That's just a massive dispersion growthwhereas, or in living standards,
whereas a recession, something even a verybad recession like the Great Recession
might be only say a 5% sort of divotin the GDP per capita growth rate.

(13:00):
I'm curious what your thoughtsare on just this issue of.
I don't think growth and business cyclesare totally independent of each other.
We think about there's these things,
topics of hysteresis wherecountries essentially never
recover from really bad troughs,and they're on new paths.

(13:21):
But I guess if we're in a world wheresay GDP per capita is say flat,
or isn't really growing very much,how do we think about, I guess,
recessions in terms of defining them,how we should respond to them
if the trend starts mattering a lotmore for the cycle, if that makes sense.

(13:44):
There's been, I think, very famous papersBritain called the cycle is the trend or
the trend is the cycle, and so forth.
And it's very apparent I think outside theU.S. it's a big question outside the U.S.
in part because the U.S. has thisvery unique ability to grow out of
recessions very quickly inits post war time series.
But there's been a lot morehysteresis across I think.

>> Jesus Fernández-Villaverde (14:06):
Yes. >> Jon Hartley
I'm curious what you think about that.
Yes, no, I always.
So let me tell you an anecdote.
When I was an undergraduate student.
So this is my sophomoreyear in college and
I'm taking intermediate macro andthe professor walks in and
we were using the textbook backthen was Darwin's Fischer.
I don't know if you used that one,maybe some of the listeners use that one.

(14:29):
I think like page four or page five ofthat edition had that graph where Dorvus
and Fisher draw like a linear trend andthen fluctuations around that trend.
And the professor comes andsays, well, that's false,
because that's not how the world works.
The world basically you haveexpansions and then you have drops and
then you very rarelyrecover further drops.

(14:51):
You kind of lose that 2%,you lose forever.
So he says I want you to tearout that page from the book.
And we are kind of,everyone is saying, well,
silly thing that the professor is saying.
So he actually goes to the first row andthere is this girl and
I still remember her name butI'm not going to say it here in public and

(15:11):
picks the book from the girl and takesthe page and in front of all the class
actually turns out the page andsays I want all of you to do the same.
And of course, you know, all of usvery happily took out the book and
turned the page apart.
So yeah, he made his point,I'm still convinced.
When you look at many countries reallywhat is happening is that you lose 2,

(15:34):
3 points of GDP during a recession andyou don't seem to ever recover them.
And even the US you can argueafter the Great Depression,
the big financial crisis of 2007,2008 probably lost around 2,
3% of GDP and we have not beenable to really recover them.
I'm going back to the pre financialcrisis trend and I think we have more and

(15:56):
more evidence these daysthat the big difference.
Between at least rich countries isthat relatively poorer countries
have more of these drops andthey just never recover from them.
So in that sense,I think the US is very peculiar and
is more the exceptionthat the norm out there.
So a lot of what I have done in myown research is trying to understand

(16:19):
these links between the business cycle andeconomic growth in the long run.

>> Jon Hartley (16:24):
Well, that's fascinating.
And I guess one wonders too,I guess if we end up having declining
population growth rates,whether or not that overpowers,
I guess, the growth inproductivity across countries.
And if that's the case,maybe we would have negative

(16:45):
GDP growth rates wouldbecome a common phenomenon,
which seems almost crazyto think growing up in.
I guess if you grew up in the late 20thcentury, even though growth rates have
slowly declined over time from what theywere at in the middle of the 20th century,

(17:05):
at the height of advancedeconomy productivity booms.
And we've had productivitybooms like in the 90s.
Do you think that negative GDP growthrates could be the norm in the future?

>> Jesus Fernández-Villaverde (17:17):
Yeah, definitely.
So this is the way I put it,very simple for my undergrads.
So I said, look, on average the USlabor productivity has been growing
a little bit less than 2% a year.
Let's say 1.8, 1.7 and working agepopulation grows around 1% in a year.
So pure accounting 1.8 laborproductivity growth plus 0.1.

(17:41):
Sorry, plus 1% growth of population,of the working age population,
that gives you 2.8.
So that means that onaverage you grow 2.8.
When the economy is doing well forwhatever the reason,
maybe a fiscal expansion, you grow 3.8.
When the economy is not doing so well,
maybe the Fed is coolingdown the economy 1.8.
So on average you are around there.

(18:02):
Now let's think for a second thatproductivity still grows 1.8 and
you will see how thisis a heroic assumption.
But now working age populationis falling 1% like in Japan.
So again,just by pure accounting 1.8 minus 1%,
it means that on averageyou are going to grow 0.8.
So as soon as the economy enters intoeven the mildest recession or the Fed,

(18:27):
the monetary authority even increasesthe interest rates just a little bit,
you go into negative growth.
Now things are going to be probablyworse because we know that a lot of
technological innovationis done by young people.
So an economy with a lot of olderpeople keeping a productivity growth

(18:47):
of 1.8 is going to be very unlikely.
So let's suppose that we haveproductivity growth of only 1.2 and
you still have a drop inworking age population of 1%.
It means that your average growth isgoing to be 0.2, which means that in
around 50% of the years you are goingto have negative GDP growth.

(19:08):
And it's a completely different world.
You see why thinking aboutdemographics completely
changes your scenario ofhow the world economy or
like any advanced economy is goingto look like around the year 2050.
So we are really entering intoa complete different environment.
And it's quite striking to me how very,

(19:29):
very few economies haveinteriorized that point.
And the few that have thought about ituse the words artificial intelligence,
because apparently artificial intelligencenow is the solution to everything.
[LAUGH] But even if you thinkabout it carefully, you know,
it's very difficult to see the largesteconomies in the world growing very,

(19:51):
very fast in the year 2050,given the current demographic trends.

>> Jon Hartley (19:56):
It's s amazing in terms of rewriting textbooks,
you mentioned the textbook thatwas written by Sam Fisher and
Rudi Dornbusch, two famous MIT economists.
It's funny how a lot of textbooks mayneed to be rewritten in the future.
Because I guess there's a couple ofdefinitions of recessions, and I think one

(20:20):
of the most famous ones is two consecutivenegative quarters of GDP growth.
And if we're in this newperiod of negative GDP growth
quarters happening all the time,
I think we'll have to revisit whatexactly it means to be in a recession.

>> Jesus Fernández-Villaverde (20:38):
Exactly.

>> Jon Hartley (20:39):
And I guess we'll still have the NBR recession dating committee
that's been started andled by Bob Hall for quite some time,
is now led by Valerie Ramey, andboth Hoover senior fellows here.
But yeah, it's gonna be interesting.
I think groups like that,along with the textbook writers,

(21:00):
macroeconomists like ourselves, are gonnahave to sort of rethink what recess
mean in the context of being in some sortof permanent negative GDP growth periods.
But perhaps we should be definingthings on a GDP per capita basis or
maybe GDP per worker basis and negativeproductivity growth rates in that sense.

(21:21):
I want to talk a little bit about models,just a bit because you spent
a lot of time on models and coming upwith new methods for solving them.
And you've done a lot ofreally great work on that.
And so on sort of this topic,topic of business cycles and DSGE models,
VAR models, curious in your mind,for listeners that may not be

(21:43):
familiar with DSG models versus VAR,dynamic stochastic general
equilibrium models versus VARs orvector autoregression models.
How do you see the strengths andweaknesses of DSGE models vs VARs?
On this podcast we spent a lot of timetalking to former central bankers,
other macroeconomists.
And funny, one recurring theme on thispodcast seems to be that central bankers

(22:08):
don't really seem to finda whole ton of value in DSGEs.
I mean, there's a bit of economicintuition that they have, but
it turns out that VARs generallyproduced much better forecasts.
But a lot of economists, central banks,
are still producingthese DSGE type models.
I mean, what is the best case inyour mind for the DSGE models?

(22:30):
I guess from both an economic theorystandpoint and from just sort of,
I guess, a practical standpoint of say,a policymaker.

>> Jesus Fernández-Villaverde (22:38):
Okay, so [COUGH] first of all,
I have written many papers on DSGEmodels and many papers on VARs.
And as you say, I think they haveadvantages and disadvantages.
To me, the main advantage of a DSGEmodel is that it really ties down or
discourse or narrative.
And what I mean by that, I have also beena little bit involved in economic policy.

(23:01):
I have seen many, many times, andI'm not going to name any concrete person,
someone coming andsaying, X should happen.
And let me tell you a simple regression or
simple VR that tells you that Xwill happen if Y occurs first.
And then two years later they come andthey say, no, no,
Z will happen if A happens.
And you say, look,Z if A happens and Y if X happens.

(23:25):
They are incompatible,they are logically inconsistent.
How do you fit this square,how do you square this circle?
And I think that the DSGE models,and it's just not the DSGE models,
I mean, in general equilibrium models,they really tie down your hands and
they force you to think throughthe whole logic of the system.
Now, having said that, what people forgetis that the reduced form of a DSGE model,

(23:50):
at least when you linearize itis just a vector autoregression.
It's just a vector autoregression that hasa lot of restrictions imposed by theory.
So if you go to the data andyou estimate an unrestricted VR,
of course, you are going to doa little bit better than when you
estimate the restrictedVR from the DSE model.

(24:11):
So what do you gain?
You gain extra flexibility from the VR,you lose some economic interpretability,
and especially you lose the abilityto run a lot of counterfactuals.
So if I were a central banker,what I would like to have is a battery of
models that will help me to thinkabout all these situations.
And in a metaphor that I use all the time,this is like saying,

(24:34):
what is better, red wine or white wine?
No, that's silly, okay?
Sometimes you have a nice steak andyou want red wine, and
sometimes you're having a nice fish andyou want a light white wine.
And I think that central bankers andin general policymakers should
understand that economics offersa wide many of choices, and

(24:56):
that really the ability of a greatpolicymaker is to select the right
wine pairing for the night, andthat's what leads to success.

>> Jon Hartley (25:06):
I might be of the view that red wine strictly dominates
white wine.
>> [CROSSTALK]>> Jesus Fernández-Villaverde: You are not
European, Jon, so I need to forgive[LAUGH] your lack of civilization [LAUGH].
It's too funny.
Yeah, I mean, it's interesting, andI think it's interesting to think

(25:28):
about the DSG literature and how much ofit is really moved to central banks and
finance departments awayfrom macro departments.
I mean, do you have any general thoughtson just sort of macro as a field?
I feel like there's a lot offolks from certainly the applied
micro world who kind of say, well,

(25:49):
macro models are kind of the sciencefiction, this work of science fiction.
And a lot of them lack identification and,say, natural experiments or
examples of exogenous shocks.
And of course,you can sort of take parameters,
estimates from those sorts of studies andplug them into macro models.
But there's a lot of assumptions even indoing that, whether it's external validity

(26:12):
and combining all these,putting all these parameters to one model.
I'm just curious, what are your thoughtson sort of the state of macro in general,
and where do you think it could do better?
Are you a believer thatempirical macro with better
identification is kind of onefuture path to sort of bring

(26:34):
more credibility revolution,maybe, to macro?
Many out there, say, Jon Steinsson andEmi Nakamura and their students and
many others are sort of entering.
I myself am sort of oneof these people as well.
But I'm curious if you have a sort oftake on sort of the future of macro.

>> Jesus Fernández-Villaverde (26:51):
Yeah, so a couple of thoughts over there.
The first is I think that in macro,we have not been forceful enough
over the last ten years to explainwhy we also have identification.
I think that some of the people in theapplied micro, they have argued, you don't
have identification in that sense, andI think that those arguments are weak.

(27:13):
And they are weak because they arethinking about identification in a very
particular context.
Those type of identification arguments,
I don't think they make terribly sensewhen you're thinking about the aggregate.
It's inherently very different to thinkabout what is the consequence of,
let's say,reducing a class size in a high school,

(27:34):
that thinking about what are the effectsof reducing the class sizes in
all high schools of the United States?
And I think that macro models andmacro econometric tools were
designed to answer the later wherethey are interested in the former.
And in the same way that they say, well,
maybe we don't believe many of yourresults, my counterargument is,

(27:56):
well, guess what, I don't truly believea lot of your results either [LAUGH].

>> Jon Hartley (27:59):
[LAUGH] >> Jesus Fernández-Villaverde
I think that a considerable amount of whatis done in applied economics these days,
I don't find it terribly useful.
Now, having said that, there is a secondthought, which is I think that at this
moment in macro, we are still digestingthe revolution of heterogeneous agents.
And we are trying to really understandwhere you want to apply heterogeneous

(28:20):
agents and where you want to apply theenormous amount of micro data that we have
been accumulating lately.
Myself, I have been working a lot onthat over the last 10, 15 years and
it's still not very clear where we wantto be in the balance between a lot of
detail and incorporating a lot ofimportant margins of adjustment versus

(28:41):
having more stylized models thatare a little bit easier to understand.
So along that dimension,I think the field is still evolving and
I myself don't have a 100%clear answer of where really
the line between parsimony anddetail should fall in the long run.
What I know for sure is that we have somuch new data and we have,

(29:05):
in addition to it, so many goodsolution methods that this is going
to transform the professionover the next ten years.
And where we will end, well, who knows?
But that's why it's fun to do research,
[LAUGH] cuz we don't quiteknow what we are doing.
I feel like, you know, a lot of macro, certainly,

(29:26):
you pointed out heterogeneous agentsis one big trend in modeling recently.
I think another recent trendhas been trying to, I guess,
maybe relax some of the assumptions aroundfull information, rational expectations.
There's a lot of folks that work onso-called behavioral macroeconomics.
And so there's a few trends there.

(29:47):
And then even going back farther thanearlier than that, I think there's for
many years, and to some degreethat still goes on, these massive,
massive battles between sort ofthe various different schools of thought.
And many of these sort of crept their wayinto these policy discussions in public.
And you sort of had very famously for

(30:09):
a long time the sort ofso-called Keynesian schools of,
or maybe old Keynesianschools of Harvard and MIT,
I think Paul Samuelson and many others.
Versus Milton Friedman andsort of the neoclassical Chicago school,
later the so-called newclassical folks in Chicago and

(30:33):
Minnesota, where you spent some time anddid your PhD.
I feel like those battles loomed solarge for a long time, and
maybe just at the same time that appliedmicro folks started to gain a lot of
credibility in terms of the much smallerquestions that they started answering.

(30:53):
But I'm curious, where do youthink those old debates lie now?
Given that I feel like a lot of the giantsof those schools have recently passed,
what are we sort of left with?
I've seen some papers now that try andmake empirical cases that sticky wages and
sticky prices are kind ofthe fundamental cause of recessions.

(31:15):
But I think even some of that evidenceis maybe not perfect or somewhat scant.
Obviously, the end of Minnesota sort ofapproach, or at least one part of it,
argues that productivity shortfallsare the causes of recessions.
And I think if you look at certainbanking crises and tech crises,

(31:37):
maybe that wasn't the case there.
But I'm curious, what is your takeon sort of schools of thought?
Do they still matter now in those battles?
Do they matter less?
And how do they influenceyour own modeling?

>> Jesus Fernández-Villaverde (31:51):
Okay, so first of all, I was always a little bit
unhappy about this idea of presentingmacro as the struggle between two schools.
Three schools or I think when I wasan undergrad I even read this book about
the seven schools of macro because itreally gives undergraduates the impression
that this is like pickingyour soccer team,
that you are either Real Madrid fan orBayern Munich fan.

(32:14):
And in fact most people try to look atthe data, try to look at the theory and
try to make a consistent answer.
My own reading of the situationwas that kind of the Minnesota,
Chicago, Rochester wonin terms of methodology.
They insisted that you wanted to be veryexplicit about the micro foundations,

(32:36):
about how people made decisionsabout information flows,
about being sure that generalequilibrium restrictions hold,
while perhaps more of the MIT harbor sidewarned in terms of substance, of saying,
look, financial frictions matter,nominal rigidities probably matter a lot,
markets quite often do not clear oroutput is demand determined.

(33:00):
And I think that by the early 2000s,especially as you were saying,
some of the older generation start passingaway or maybe not yet passing away, but
at least been a little bit older andretiring from active research.
The next generation was a little bitless personally invested in some of the,
you know, more sticky discussions and

(33:20):
much more open to think about inconstructive cumulative terms.
This is still the case today that when yougo to the market and you read your market
papers, it's easy to tell if this isa macro person from MIT or a macro person
from Chicago, but the difference isprobably smallest that it has ever been.
And I think that what we are trying tofigure it out is okay, fine, people don't.

(33:44):
You were mentioning before behavioral,
and I fully agree with the idea thatpeople don't have full information.
You asked me today, for instance, whatis the exchange rate between the US and
the Euro?
I'm going to get itprobably perhaps within 5%.
And look, I'm an economistfrom Spain living in the US.
So if there was someone who should know[LAUGH] the exchange rate between the Euro

(34:06):
and the US dollar, it should be me.
And yet I only have a fussy idea about it.
On the other hand, we want to thinka little bit carefully about how people,
if people are building their beliefs andpeople are coming up with rules of thumb,
we want to be a little bit moreconstructive that just say no,
people is silly andmake systematic mistakes.

(34:27):
My impression was always that a lotof behavioral macro was a little bit
like that.
Or let me assume that people makethis mistake or have this bias and
let's look at the consequences.
I think that much more productive, andthat's something I have been working on
lately, is thinking about, well,let's Imagine that people have very simple
learning systems and are using deep neuralnetworks and say, well humans actually our

(34:50):
brains are just deep neuralnetworks how we learn about that.
Can we come up with simple rulesof thumb given the observations?
And that gives us a constructive wayto work out through these things.
So in that sense, you know,I think that we are making progress and
we are certainly moving forward andI think we are going to see very good
work on this area overthe next ten years or so.

>> Jon Hartley (35:13):
It's interesting cuz I guess behavioral macro is all about how,
I think a lot of behavioral economics for
a long time was just sort of makingthis point that well, people actually
don't maybe follow full informationrational expectations correctly.
Which is hard to kind of arguein the sense that you sort of

(35:34):
have to argue that one that there isn'tsome sort of a systematic model out there.
But if, I guess the sort of newholy grail of I feel like of sort
of behavioral economics orbehavioral macro as well is how do
people systematically make mistakes andwhat biases or

(35:56):
how are people actually thinking Iguess in terms of complex systems or
you're talking about neural nets but
it hasn't been quite enough to justI guess say on the part of Thaler and
others will look this is a violationof these past assumptions.
Well, maybe we just hadthe utility functions wrong.

(36:18):
What are the correct orwhat are the better utility functions?
And how do you square a lot ofmicro evidence with our models?
And I feel like that's stillvery much an ongoing debate.
On the schools side of things,
it's amazing in all our ECON 101textbooks one of the first things
we're taught in macro is that thereare these different schools of thought.

(36:42):
And I do wonder how the textbookswill be written in the future
given that maybe they matter a bit less.
But I do think there's some very realworld implications of some of these shifts
in terms of.
I think you're rightin saying that MIT and
Harvard is one the substance of things.
But I think because of that we onlyin models tend to care about interest

(37:05):
rates or tend to include interest rates.
So things like monetary aggregates ormoney are completely withdrawn or
non-existent.
Whereas a lot of people on Wall street forexample would like to point out that look
at M2Is was rising a lot when we sawinflation in the early 2000 and twenties.
And even some Minnesota type economistswho I know some new classical economists

(37:28):
who I know who are still around,some younger,
some older still sort of bemoan that shiftin the sense that sure, maybe monetary
aggregates can't be used as a tool becausemy demand isn't really predictable, but
at some level it still explains inflation.
Well, when you see big jumps in M2 variousmonetary aggregates, I'm curious if.

>> Jesus Fernández-Villaverde (37:48):
Yeah. >> Jon Hartley
as a minister.
Yeah, actually, this is interesting that you
bring this up [COUGH] because around 2010,2011 there, maybe even a little
bit earlier 2009, there was all thisdiscussion in particular from people
in financial markets about how the bigmonetary expansion at the Fed at the time

(38:08):
quantitative easing was to create a lotof inflation, maybe even hyperinflation.
And I remember I had an MBS studentshouting in my class [LAUGH] at Wharton
that we were going to have hyperinflation.
And my thinking at the moment wasthe very basic fiscal theory of
the price level that basically statesthat you want to think about really

(38:30):
the interaction between the fiscal sideand the monetary side of policy and
that if the fundamental deficits ofthe future, parts of deficits have not
changed, issuing more money, the onlything you are doing is you are switching
one type of debt of the public government,which is long term debt, for
another type of debt, which isthe short term debt that we call money.

(38:55):
And that was not goingto have much an effect.
In fact there is a very famous paper byNeil Wallace in American economic review,
either 1981 or 1982 that basicallyshows an irrelevance theory.
Anyway, I made that argument andI remember.
So I basically said, look, there is notgoing to be inflation, don't worry,
there will be no inflationover the next 10 years.

(39:16):
And I think that was a big victory foreconomic theory and
it was a big victory formacroeconomics and
anyone who had studied macroeconomicsseriously at the time.
And I remember because [LAUGH] I had a fewfriends in the financial markets and
man did I get hate emailslike you don't know anything.
You are typical ex head at the universitythat don't understand anything whatsoever.

(39:40):
I'm moving all my investment intoinflation protected asset classes.
And I was telling them there is not goingto be any inflation, don't worry about it.
And, I guess that even by2019 when we were talking
about this 2020,[LAUGH] they were still saying okay.
Inflation didn't happen, but we are notquite sure why we are unwilling to believe

(40:03):
the fiscal theory of the price level.
And I guess on the other hand, in 2021,I was among the ones who argue, again,
think through the logic ofthe fiscal theory, the price level.
Now, we are going to have inflation.
And again, now I guess that there werepeople who kind of overlearned the lesson.
You know that old joke that we alwaysgenerals always fight the last war.

(40:25):
And people were saying,there was no inflation in 2010.
There is not going to beinflation either now.
And I was saying, no, no, no, no, no,
[LAUGH] because the argumentis completely the opposite.
The argument is now we are changingstructurally the fiscal situation of
governments with everythingwe did around Covid, and
we are also changing the supplycurve of the economy.
And that's very differentfrom what happened in 2010.

(40:47):
Now, we are going to have inflation.
And again, my, the same friends from thefinancial markets were telling me, no, no,
no, no, no, this is not going to happen.
And guess what?
We had inflation.
So in that sense,even if this is not something
that macroeconomistsare really bragging about,
I will say that macroeconomicswere successful.

(41:11):
Okay, let me be very careful.
The macroeconomics that you will learnin the standard first year graduate
school textbooks was very successfulat predicting both the lack of
inflation in the early 2010s andinflation in the early 2020s.
A very different thing is thatthe type of people that write for
the New York Times orthe type of people who write for

(41:33):
the Wall Street Journal are a veryparticular subset of the profession.
And some of them,I have strong reasons to believe,
are not really up to date onthe best economic theories.
And they say silly things becausethey are playing a different game.
They are playing the rhetorical game.
They are not playing the game oftrying to understand the world.

>> Jon Hartley (41:51):
Well, I guess like one response that I may have in I
partially one of the thingsI say is fiscal theory.
One of the things I think I just finda little challenging with fiscal theory is
measuring expectations aboutfuture deficits, right?
And you can do certain things.
You can look at centuries of data and
maybe assume that peoplehave unbiased expectations.
I's a huge assumption, andtry and test the theory that way.

(42:13):
But at some level, it's hard to know,and you can't really follow
CBO forecasts just because of theircurrent policy and current law and
so forth and don't really reflectwhat fiscal expectations are.
And the story thatperhaps John Cochrane or
Eric Leeper would tellis that one in 2008 and

(42:38):
2009, when there wasa big fiscal spending,
there was a promise to pay it back.
Whereas in this case,with the early 2000s fiscal stimulus,
there weren't any promises to pay it back.

(42:58):
And I think that it's certainly,I think, a somewhat compelling story.
But how you actually measure that,as someone who's just an empirical
macro person, I feel like it's a bitof a difficult to measure thing.
Whereas you look at something like M2,for example,
in the 2010s QE, M2 didn't reallygo up a whole lot, and why?

(43:22):
Because when all the central bank wasdoing was buying up treasuries and
swapping them with reserves.
Whereas what I think wasfundamentally different.
And you also had a stimulus that wasvery focused on government spending
with the ARRA orAmerican Recovery Reinvestment Act.
Whereas the early 2000s fiscal stimuluswas really all about transfers,

(43:46):
whether it was stimulus payments,unemployment insurance, PPP and so forth.
And so in that case,
you had this massive uptick in monetaryaggregate measures like M2, and why?
Because the amount that was going,all this fiscal stimulus
was in part going into people's demanddeposits, which are measured in M2.

(44:06):
So you had a massive jump this time,something that you didn't have last time.
So the character offiscal policy was very,
very different in the early2020s versus 2008,
2009, and I feel like that matters.
I'm curious if you haveany thoughts on that,
even though essentially verysimilar QE was embarked upon.

>> Jesus Fernández-Villaverde (44:28):
Okay, so first of all, for
those listeners who know more aboutthis fiscal theory of the price level,
I have joke on a few conferences andinterviews that I'm a moderate fiscal
theory of the price level person incomparison with both Eric and John.
Eric Leeper and John Cochrane,who I will say they are like the purest,
the pure orthodox fiscaltheory of the price level.

(44:50):
I'm more of a moderate that puts a littlebit of fiscal of the fiscal theory
of the price level witha little bit of NEO models.
So in that sense,I'm perhaps closer to Chris Sims.
So in that sense, I also think thatissues related with supply and
with prices stickiness matter.
But I think that the explanationyou just gave me,

(45:14):
I will argue, is a fiscal explanation.
And that really the only differencebetween what I tried to say before and
the way you put it is more on the emphasisor in the pure rhetorical aspect of,
in the sense of discourse of howyou structure the argument that
in the substance.
And basically in 2000, okay,let me put it in the following way.

(45:35):
In 2010, people were not going tothe restaurant because they didn't
want to go to the restaurant.
They were worried, they wanted to save.
And what the Fed is doing is saying wellI'm going to take your 30 years bonds, and
I'm going to give you cash instead.
And people are saying okay,how is this changing my situation?
How is this changing the issue thatI'm worried about the future and

(45:56):
that I don't want to save a lot.
While exactly as you say in2021 people are not going to
the restaurant becausethey are afraid of COVID,
so we send them $500, andwe tell them, go and ask on DoorDash.
And then, yes, people went and startordering a lot of takeout at DoorDash.
So I think we werebasically saying the same.
But the point I was tryingto say is that if you pick,

(46:19):
I would say the top 10 macroeconomistsin academia in 2010 and
in 2020, 21,they probably did a much better job in
forecasting inflation notin like the next month.
But like in the big picture of what wasgoing to happen using a standard tools
of economic theory that what the savvypeople from the financial markets did.

(46:44):
And I think we should be proud of it andbe,
brag about it because peopleare not really getting this point.

>> Jon Hartley (46:52):
Yeah, I guess like it's.
So I think there's still a differenceto be made though with I guess
like both sort of monetarist andeven your sort of old
Keynesian Islam like reasoningwhich is more, more static.
And where saying that look therewere all these transfers today and
that's what was causing inflation.

(47:14):
And now, I mean there were lots of peoplesaying it was all about supply chains and
so forth, and it can be both too, right?
Like a supply curve shifting leftward anda demand or an aggregate demand
curve shifting, shifting right canproduce inflationary conditions together.
But I think,
there is still a big differencein terms of just the mechanisms.

(47:36):
Whereas the IS-LM and Montrose modelsare very sort of static models.

>> Jesus Fernández-Villaverde (47:41):
Yes.

>> Jon Hartley (47:42):
Fiscal theory at the price level is really all about expectations.
And I still think there'sa big difference there.

>> Jesus Fernández-Villaverde (47:48):
Okay, that's fair enough.
But I think that what the fiscaltheory of the price level,
even if you just go back to the veryold school of Wallace and Sargent and
Wallace and unpleasant monetaristic.
Arithmetic says, I think,that both the old Keynesian story and
the monetary story missa fundamental linchpin,
which is the connection between fiscal andmonetary theory, sorry, policies, and

(48:12):
that if you think about just one of themin isolation from the other, chances
are that you are going to make systematicmistakes in how you think about the world.
And what the fiscal theory of the pricelevel, with all its limitations
is telling you really try to think aboutthese two things together and to me,
that's a big advice.

>> Jon Hartley (48:34):
Yeah, I think that's right.
And I think, at some level,
a lot of these models are somewhatobservationally equivalent.
So it is really difficult to reallydistinguish, say, between them.
There's something to be said, I think too,
about what central banks haveactually been doing too,
which is these massive QElargely announcements or

(48:57):
central bank swap lineannouncements that seem to have
this ability to quell financial panic and
provide liquidity to the financialsystem in moments of panic.
Which is a bit of a different thing thansay, the fiscal policy, fiscal transfers,
say, causing inflation throughthese different mechanisms and

(49:21):
all that's something that's due interms of how central banking is done.
I want to just pivot a littlebit to economic history.
You're working onan economic history book.
Tell us a little bit more about that.
You've done a lot of really great work onthings like Fractured Land Hypothesis,
many other great worksof economic history.

(49:42):
I'm curious.
Tell us more about your economic historyinterest in your forthcoming book.

>> Jesus Fernández-Villaverde (49:47):
Okay, so first of all, I guess I was always
a little bit more interested in economichistory than macro when I went to college.
Economic history when I went tocollege was a little bit boring.
It was about textiles in Yorkshire.
So we basically, spent one semester in myfreshman year in college learning about

(50:08):
the spinning Jenny and the mule and I'mprobably getting all these things wrong,
but about textiles in Yorkshireduring the late 18th century.
And that to me was so boring.
So I kind of lost interest ineconomic history at the time.
At the same time,I was always very concerned about,

(50:31):
let me call the big,big picture questions.
In one particular case forme, since I'm from Spain,
is why did Spain miss the train ofindustrialization in the 19th century?
Why then did Spain industrialize inthe second half of the 20th century?
And as we were saying before, why Spain?

(50:54):
Convergence to the level of the USstagnated by the late 1980s.
So I was always veryconcerned about these issues.
And most of the answers that peoplewere given were loose, they were kind
of big ideas, but not really tryingto sit down and thinking about them.

(51:14):
And in the late 1990s, early 2000s,we started having a new generation of
economic history and of course this yearthe Nobel Prize went to Daron Acemoglu,
Simon Johnson and Jim Robinson, whichI'm very happy about precisely because
they brought in those big questionsback into the center of economics.
And we can fight a littlebit at the margin whether or

(51:37):
not the instrument that they use isthe right instrument or not, but
to me,those discussions miss the big point.
And the big point was to say, look,
the really big issue is why did Japanindustrialize and China did not?
Let's try to work through this and
let's try to use the modern toolsof economics to work about it.
So I thought, Gee, this is in some sensewhat I wanted to do from the beginning.

(52:01):
So I went and asked to be allowed toteach a class on economic history.
And I was lucky enough that mydean accepted that proposal and
I have been working since then on a book.
So now last week I was actually atNorthwestern and I bumped into Joel Mokir.

(52:21):
Joel Mokir,the book is under contract with him for
Princeton University Press and he gave mewhat I think you guys call a dress down.
He basically told me whatis happening with the book.
So the book is sufferingfrom two problems.
The first one is I think I've beena little bit over ambitious by now.
I think I have like 800 pages andthat's probably too many and

(52:42):
I'm still not even like one-third down.
The second one is I really,really need to find,
let's say six,nine months in a desert island
at the top of a mountain,in a monastery away.
Just yesterday I woke up in the morningand I said, today is Thursday.

(53:05):
I'm going to work on the book forthe whole day, impossible.
I got like 200 emails andjust handling emails,
answering students,working with co-authors.
So I really need.
So my main goal now in life,
if anyone knows about that monastery,please let me know.
Find a place where I can go hide fornine months and finish the book.

(53:29):
So I will say the book.

>> Jon Hartley (53:30):
There's a place in California called Hoover.

>> Jesus Fernández-Villaverde (53:32):
Yeah, I keep telling them that.
They don't seem to get the hint.
But really go in a place,hide for nine months,
disconnect myself from the Internet.
Internet, send my wife to a wonderful,
touristic trip around the world andfinish the book.
And I really like the drafts that I have,I have shown it to some people, and

(53:52):
the drafts of the chapters I have, I thinkit's a very creative and innovative book.
I'm really bringing, like,
everything we have learned abouteconomic history over the last 20 years.
It's just that it's a globaleconomic history textbook,
it's a lot of effort and you really needto spend a lot of time learning stuff.

(54:13):
So, for instance, last year I wantedto write about Latin America,
like the chapters thinkingabout why Latin America fail.
So I basically had to learn a lotabout Latin America economic history,
I didn't know that much myself.
And that was fun andI spent like maybe nine months.
So now I'm your person, I'm the guy.

(54:33):
If you want to ask questionsabout Argentinian economic
history in the 19th century,I know how to answer them.
But [LAUGH]->> Jon Hartley: Would you describe
yourself as economic institutions?
Like, is what your sortof fundamental cause?
Yes.
So my punchline yes is institutions.
I think some countries, for whatever thereason, can get the institutions right,

(54:54):
and some countries don't.
So let me give you an exampleof my favorite question,
I asked this question in nearly,I would say 50% of my midterms.
And I don't ask that questionmore often because otherwise,
the students will be ableto forecast it perfectly.
Why did the oil industry firstappear in the United States?

(55:17):
The United States doesn't really havea lot of oil in comparison with other
countries.
And yet the United States is even todaythe largest oil producer in the world.
Well, because there is a fundamentaldifference between the US and
nearly every other country in the planet.
If I go back to my backyard, I'm lookingat my backyard now through the window,

(55:38):
and I dig and I find something, it's mine.
If I go to the backyardof my dad in Spain and
I dig and I find something,is the government's, okay?
Is what, for instance,in the British common law system was,
in the English common law systemis called crown's property.
So if in a country where I dig.

(55:59):
And that belongs to the government.
You are not going to have a lot ofdigging in a country where you dig and
that's yours,you are going to have a lot of digging.
So the US develop a very vibrant.
The first oil industry in the world andeven to today is the main
technological leader in oil industryis precisely what we have, why?

(56:21):
Because the US has better institutions.
It's not that the US institutionsare perfect, the area is better.
Now, what I think I differ perhaps not,but
perhaps a little bit from someof the other institutional
guys is that I like to link this a lotwith a rich historical narrative,
trying to understand these institutions.

(56:44):
Yes, they appear, but
they appear because of the confluenceof a lot of different reasons.
It's kind of hard tohave a one-size-fits-all.
The US is in the way the USbecause of a lot of events, and
a lot of them were random anda lot of them were more structural.
And what I'm trying to tell isa history that highlights institutions.

(57:06):
On the other hand,I also trying to be very balanced.
And in the book I try to also give some,
I will say reasonable amount of spaceto alternative theories and say,
look, maybe there is some other typeof officials that may be important.

>> Jon Hartley (57:27):
Like culture, geography.

>> Jesus Fernández-Villaverde (57:28):
Yeah, exactly.
So geography, I give a lot of space.
And the reason for
that is because I actually don't thinkthat geography per se matters that much.
But for instance, you mentioned mywork before on the fractured land.
Geography is key on determine the typeof institutions that you have.
And these things matterin a lot of subtle ways.

(57:50):
And I don't think that economies andeconomic histories experience have really
work out all those subtleties andall those nuances.

>> Jon Hartley (57:57):
Yeah, and I agree,
I think there's a lot of thesenuances to work through.
With institutions as well, Acemoglu,
Robinson and Johnson just won the Nobel.
I think institutions matter a lot forgrowth, but which institutions and why?
It's a huge question.
And I think even their taxonomy ofinclusive versus extractive institutions

(58:21):
isn't perfect, I think.
And I asked Daron Acemoglu, andthis when he came on the podcast,
which is I think, at some levelhow do you explain the success
of what I would call liberal economicinstitutions or massive market reforms?
If you look at India in the early 90s orChile in the late 20th century, or
you look at Eastern Europeafter the fall of the USSR and

(58:43):
a lot of these sort of stories aboutinclusive versus extractive institutions?
I don't think it fully answer that.
But it's been a real honor to have youon and hear about your career and ideas.
You're also a prolific Op-ed writer inSpain as well in your home country.
And I highly recommend that ourlisteners check out both all your

(59:05):
academic writings as well asyour popular writings as well.
Thank you so much for joining us.

>> Jesus Fernández-Villaverde (59:12):
Thank you.
Thank you for having me.

>> Jon Hartley (59:14):
This is the Capitalism and Freedom in the 21st Century podcast,
an official podcast withthe Hoover Economic Policy Working Group,
where we talk about economics,markets, and public policy.
I'm Jon Hartley, your host.
Thanks so much for joining us.
[MUSIC]