This week on the podcast Matt sits down with Michael Brown, Principal US Economist for Visa and member of the Walton College’s Dean’s Executive Advisory Board. They begin the conversation with Michael sharing how he got interested in economics during his undergraduate studies at the University of Arkansas, what he does in his role at Visa in the macroeconomic space and how he synthesizes so much data down to communicate it to their clients. They go on to discuss the methods of forecasting the US economy and also the challenges and trade offs associated with short term versus medium term versus long term predictions. They also touch on the most challenging pieces to forecast. They end the discussion with a focus on supply chain and the macro picture of the economy as well as some medium to long run trends on the consumer side of the economy that Michael sees coming up.
Episode Transcript
Michael Brown 0:00
Whether it is the invasion of Ukraine, whether it is COVID restrictions in China,
every single one of those policy actions, physical actions has global ramifications,
not just for the global economy, but for on the ground, individual business owners
who are trying to manage through that lack of clarity.
Matt Waller 0:24
Excellence, professionalism, innovation, and collegiality. These are the values the
Sam M. Walton College of Business explores in education, business and the lives of
people we meet every day, I'm Matt Waller, Dean of the Walton College, and welcome
to the be epic podcast. I have with me today, Michael Brown, who's principal US Economist
at Visa. And Michael received his undergraduate degree from the Walton College and
his Master's of Arts degree in economics from the Walton College. And he is currently
on my Executive Advisory Board. Michael, thank you so much for joining me today.
Michael Brown 1:06
Great to be here, Matt.
Matt Waller 1:08
So, Michael, you've really been in economics your whole career and you studied economics
as a student. Would you mind just sharing a little bit about how you got interested
in economics?
Michael Brown 1:23
Sure. Yeah, I guess it all started as an undergraduate at the Walton College, you
know, I was taking, I believe the intermediate level courses. And it was really micro
that was fascinating to me. So the application of mathematics to business problems,
was intriguing to me. And I kind of stayed down that path for a while, partnered with
a professor at the time by the name of Cary Deck, who was a very inspirational person
in my career track. And Cary was adamant that the research really helped us understand
business problems in an applied sense. And Cary was an experimental and behavioral
economist. And that was also quite fascinating. I have a minor in Psychology also,
from the University of Arkansas there and blending those two passions, I thought was
a natural intersection to do micro.
Matt Waller 2:25
Michael, of course, I've heard your presentations, and I followed, I followed your
career. And it's really impressive, the knowledge you've gained, of course, you've
been with Visa for almost five years. And prior to that you were vice president and
economist at Wells Fargo for over eight years, and you've had other experiences as
well. But would you mind telling us a little bit about what you do as principal US
economist for Visa?
Michael Brown 2:57
Sure. Well, it's a far cry from that micro undergrad student that I described to you
just a moment ago. You know, the reality is I spend a lot of my time focused on developments
in the macro economy, there is some behavioral aspects of what we do in terms of understanding
shifts in consumer confidence and psychology, and how that may impact consumption
patterns. But more broadly, if I had to divide up kind of my time into a pie chart,
if you will, I'm an economist, I love charts, right? So, you know, you divide the
time up, I'd say about 70% of my time, is focused on interfacing with our clients
and communicating the work that's done here in the office by our incredibly talented
team of analysts, economists and data scientists distilling that down into an easily
digestible format for CEOs CFOs, and senior leaders across our clients, which include
banks, large merchants, you know, there's fintechs are kind of in the fold, and big
tech companies today, as they're starting to wade into the payments ecosystem. So
a lot of what I do is information, distilling, and presentation. Still have the hard
skills, though, you know, just this morning, we were designing a new round of forecast
models. So still heavily involved in econometric design as well, as you know, we do
some academic work here as well to keep us fresh on some of the new and innovative
techniques out there in both economic forecasting, but also sort of pushing the envelope
with broader economic research topics.
Matt Waller 4:38
Well, I could imagine that it's probably hard to find people that fit into your profile,
because a lot of times people that are real strong in econometrics tend to be you
know, focused on math and maybe more introverted. Clearly one of yours strengths is
communication, but also that synthesis skill. Anytime you've got a lot of different
analyses, especially when they may be pointing in different directions, trying to
synthesize it in a way that people can understand and relate to and then make decisions
upon isn't easy. How did you learn to do the synthesis and communication pieces?
Michael Brown 5:25
That just came with, with experience, you know, my first job out of graduate school,
is actually at the Arkansas Federal Assembly. And I served as kind of a policy analyst
and, you know, very research focused from the beginning. But then, you know, you complete
a research project, and then you're often called upon to testify in a committee, or,
you know, in front of, you know, other members of the legislature. And that was when
I learned fairly quickly that, you know, distilling things down to their constituent
components, and then trying to communicate, just the essentials, was sort of core
to it. So that was really an enlightening experience, having to take a often 40, 50
page research document, distill it down into 10, 15 minutes of testimony, just going
through that process was very challenging at first, I'll admit. When I moved to Wells
Fargo, the story changed, right, it was all about, we have a literally like telling
a story of the economy, and its facets and its developments and its nuances and turning
points, and then trying to convey that in not just an informative way. But frankly,
in a way that's going to connect with the audience that you're in front of. So for
example, you know, if you're talking to a hedge fund, you can be a little more technical
and a little more rigorous. If you're talking to CFOs of middle market clients or
midsize companies, that takes a different approach. So that's when the skill set of
adapting communication styles came into it. And now it's more a function of okay,
now we're speaking mostly to CEOs and CFOs, you kind of have a boilerplate that you
can navigate.
Matt Waller 7:11
Well, I know you've done a lot of forecasting of the US economy. And that's so hard
to do. Would you talk a little bit about methods of forecasting the US economy and
also the challenges and trade offs associated with short term versus medium term versus
long term?
Michael Brown 7:32
All phenomenal questions. So the first comment I would say is you need to be as humble
as you possibly can be to be an economic forecast, a particularly in today's environment
of if you think you're going to be right, the majority of the time you are you were
mistaken yourself. And I learned this long ago, right. I mean, when I first started
forecasting, at my time, at Wells Fargo, the most important aspect of doing it is
coming up with a consistent storyline around what the data supports. And you know,
the numbers that we come out of our models were often kind of ad factoring to make
sure that that consistency is there in the message. And in the storyline, a macro
forecast ranges anywhere between five to 32 line items, we're forecasting about 32,
here at Visa acurately. And those line items should all move consistent with that
fabric of the story that you're telling. And sometimes the models aren't great at
doing that as hard as you try as a as an a good econometrician. They're not always
consistent, and particularly in a post COVID environment as we're in and trying to
navigate now. So one be humble, two grounded in the macro theory, if you're not driven
by the known principles and relationships, that, frankly, are our guiding principles.
It's the reason we're economists and not data scientist, is we're using that framework
of macro theory to guide our thinking around that storyline development. And then
finally communicating it in a clear and concise manner that, you know, has its footnotes
attached, as I like to say, but you're still, you know, accentuating it with with
just the,
Matt Waller 9:26
Well, you know, when you talk about all these different items you're forecasting,
what are what are some of them, you're you're forecasting and which ones are the most
challenging?
Michael Brown 9:38
Well, so the top line, I mean, the biggest thing everyone looks at is US GDP growth,
right? So we, we don't even refer to it as GDP when we're talking to clients. We just
say economic growth, you know, again, we're trying to get out of the realm of, you
know, teaching undergrad econ as part of our updates to our executives, certainly
inflation. Here at Visa a big part of our business is obviously consumer spending.
So for example, we look at both real spending, which tells us how much foot traffic,
they go through our merchants. But we also look at nominal spending, which is going
to correspond with revenues. So again, taking, you know, the the inflation piece,
adding it to the real or inflation adjusted variables. All of these are kind of key
inputs into not just our business, but our clients business as well. The great thing
about our business model here at Visa is when our clients do well, we do well, it's
also so you know, it's it's kind of interesting to come in and provide these additional
perspectives. Now, the hard part, I would say, you know, interest rates are up there
with some of the more difficult line items to kind of forecast oil prices. Oftentimes,
I will consult a ouija board rather than a forecast model. I mean, it is really difficult
to kind of forecast it. And in the third bucket, I would say is anything that would
involve political calculations versus economic fundamentals. So for example, you know,
the the hot topic these days is the federal budget and of course, debt ceiling debates,
that is inherently a political decision. And I learned a long time ago, let's stick
to the economics, we'll let the political forecasting to the political scientists
because it gets really tricky, very quickly to try to figure out the direction of
those kinds of variables.
Matt Waller 11:33
Your point about, you know, trying to come up with the story that weaves everything
together, where the forecasts of each of the lines makes sense together. Because you
would expect certain things to react in certain ways, under certain circumstances,
based on theory, but that gets into something really interesting, which is the short
run versus the long run. Because even if your story's right, because of the stochastic
nature of the variables, the short term forecasts could just have some noise in them
that make the make it inconsistent with the story.
Michael Brown 12:19
You're absolutely right, I mean, you know, we actually divide our time up. Usually,
the mornings of our day, are really focused on the short term, macro fluctuations.
And for better or worse, the bulk of our client base is really keyed in and dialed
in to that short run view. That said, your eye isn't on the long run ball, some, you're
gonna miss a lot. And so the afternoons here, are really us jumping into our huddle
rooms, or our little side chat rooms. And kicking around these longer run themes.
We spend a lot of time talking about the implications of Gen Z, on productivity, growth,
and technological adoption, you know, the trends that are going to shift across demographic
groups, but also what a large and growing retirement base means for the potential
GDP growth of the US. And all of these things, if you're doing any kind of medium
to long run planning are essential to kind of inform clients of as well. So it's not
just demographics. I mean, we go back again, to macro principles, right? We're talking
about productivity growth technology, or, you know, labor augmenting technology in
that production function. And then capital flows globally. And all of that feeds into
kind of our medium to long run perspective. But it's all grounded in those simple
production functions that we all learned, you know, as undergrad and grad students.
Matt Waller 13:53
Michael, there's economists in a lot of different industries, I would think you have
a little bit of an advantage, because any economist has access to archival data, you
also have access to payment, and data.
Michael Brown 14:10
We do I mean, you know, in terms of our payment volumes, we roughly see close to 25
cents of every dollar spent here in the United States. And that's about, you know,
somewhere around 21 to 22 cents globally. So it's, it's a fairly large penetration
rate. And it does allow us to get a pretty high degree of visibility into the consumer
side. But you know, we also have small business cards and commercial cards as well.
So we are able to gather intelligence on multiple facets of the US economy through
the lens of of our card payment data. But I would say that there's also challenges
with that, you know, I mean, I'm sure your students are learning about big data there
at the Walton College in excruciating detail. It has a tremendous amount of noise
associated with it. And as an economist, you know, it's a little funny to say working
for a business. But there's a lot of what I call business noise in our big data. So
I'll give you an example. Let's say we have a bank that was issuing a competitor's
product and they convert to issuing Visa. Well, that was economic activity that was
going on before we could see that in our big data set. And we've got to figure out
ways, creative ways, that control for that. And we've got folks migrating in and out
of our ecosystem daily, right. So it would be impossible to trace down every single,
inbound and outbound kind of transaction, when we're talking, you know, millions and
millions a day. So we've come up with clever ways, thanks to our talented data scientists
to kind of scrub the big data and extract what I call the economic signal from from
the big data and make it a useful economic measure. So yes, there are a lot of advantages,
but there's also a tremendous amount of sort of data cleaning, and then it, then there's,
then there's the sort of balance issue that we're always struggling with as particularly
in our profession, as Applied Business economists, because you're right, we have a
lot of insider information. And as a publicly traded company, we have to protect a
lot of that information. I would love nothing more than to start publishing forecasts
based off of what I saw two days ago in our transactions data. But that would end
up front running our forward guidance, our executive team, our financials. So there's
a business ethics angle to this as well, that we have to be very, very cognizant of.
With that big data, we also have a big level of responsibility to protect the confidentiality
of some of that data as well.
Matt Waller 16:56
I know that, under your leadership, Bloomberg News ranked Visa's US economics team
as among the top forecasters of the US economy. How how did you lead your team to
that achievement? What what did you have to do?
Michael Brown 17:18
First, you know, I'll just kind of walk you through how we onboard our our economist
here. And phase one is really the indoctrination of our approach to econometrics.
And basically, systems thinking, for me to think about the economy as a system? And
what are what are the things that could break the system? What are the things that
can make the system run faster? What are the levers we could pull to, to perhaps make
the system run more efficiently, so you start thinking of it as a system. And we basically,
to divert it a little bit, as I like to say, we basically apply that thought process
to our econometric structures and models. So we have, we use machine learning techniques,
such as Bayesian vector auto regressions, and other types of applied techniques as
well, that will take both public and proprietary economic data that we subscribe to
feed that through our systems of equations. And that helps us understand the co-movements
between things. But again, we also have a lot of experience here. I mean, get off
get to a team with close to 90 years of combined experience, you know, forecasting
and understanding the economy. And that's invaluable, having folks that have lived
through different types of recessions have seen data develop in different ways. And
probably the death kill of an economist is have seen the magnitudes of the revisions
that can come with some of the data that we rely on heavily to find turning points
in the economy. And all of that just kind of gels. I think probably the most rewarding
moment of my months, is sitting in a forecast, which we have the first Friday of every
month after the Non Farm Payroll report comes out. And we sit there and everybody
goes around the room. And they they talk about what what they think is right what
they think is wrong, and there is no hierarchy in that room. Everyone's idea matters.
Everyone's thought matters. Everyone's research matters when it comes to that. And
creating that flat structure. In that one meeting that sets the tone for what we're
going to say, for the next 30 to 60 days is is incredibly important as a leader to
make sure that you're creating that inclusive environment, a free flow of ideas, particularly
in a research function like we serve. So just setting up the environment and setting
the team up for success in their endeavors as an applied econometrician as well.
Matt Waller 19:58
So you must have had to build a lot of trust to get that to happen. Because it's easy
to say. But when you get that many people into a room, it's there's a lot of complicated
sociological and political issues that go into it. So how did you build the trust?
Michael Brown 20:18
Well, so one of the sort of foundations of my management style is I really am important,
I really place a lot of emphasis on one on one conversations. So every single week,
whether I'm traveling or not, it may be 15 minutes that week, or sometimes it's an
hour, and I'm sitting down with each member of the team, and understanding what's
working, what's not working, helping them set priorities, and knowing that I have
their back when they can't get to priority number five on that list, is also an important
aspect of kind of managing workflow. And I think probably another dimension to it
is really listening more than you're talking, you know, those one on one conversations,
if I leave the room, feeling like I spoke more than 30% of the time, I think I walk
away saying I probably did something wrong. And I need to kind of reassess and reevaluate
my approach the next time.
Matt Waller 21:20
Well, congratulations on your terrific achievement there. One shifting gears just
a little bit. You know, throughout history, there are these shocks that occur to economies
I, the biggest one, of course, COVID, that's the most memorable, had an incredible
impact in so many different ways. It's hard to even conceive of all these changes.
And you know, you could think and we would be, we have to bring up generative AI.
But I wonder, is generative AI, a shock? That, I mean, maybe we're not seeing it yet,
but I know, I know, so many alumni in various industries that are telling me really
innovative ways that they're using it. But even in the university, we're finding ways
of using it that are quite remarkable, that truly make us more efficient. But what
are your thoughts? Do you think? Do you think this is going to be a shock?
Michael Brown 22:29
Shock might be a little strong of a word. I, as an economist, I go back to first principles.
And I think the power of what you described is really the augmentation of labor. And
if you think about the long run trends here in the US, we're in a country with an
aging population, slow net in migration, slow year on year population growth. The
key story in macro economics today is the high rates of inflation. So how do you bring
that down? Well, in light of not being able to grow your labor force, you bring that
down through enhancing your existing labor force. And so I'm kind of a glass half
full kind of guy on this, because I think the power to enhance the abilities of the
human capital that we have in this country and others around the world, I think it's
going to be an unlock of potentially higher GDP growth than we otherwise would have
had in the absence of this kind of innovation and technology. Now, that said, I don't
think we're ready to just run full steam ahead. I'll just give you the example here
at Visa, it is prohibited on our systems to use chat GPT or other generative AI tools.
And the reason is that the fear of putting in something that is company proprietary
into that ecosystem and having it uploaded into a public database for other future
consumption, maybe to answer other questions out there. That risk of intellectual
property CPG or loss is really high.
Matt Waller 24:11
Yeah, that's a really good point. The World Wide Web was similar. When it was created,
many companies banned it because it was so easy for information to flow outside through
the World Wide Web. So Michael, it's funny prior to the pandemic, no one very few
people knew what Supply Chain Management was and Logistics was a term they didn't
really use much. And during the pandemic, it was on the front page every day and now
people know about it. It's amazing. And of course, as you know, our undergraduate
program in supply chain is now ranked number one in North America by Gartner. And
the Masters program is ranked number two in North America by Gartner which we're very
grateful for. But would you mind speaking little bit to supply chain the macro picture
of the economy?
Michael Brown 25:06
Well, it's interesting, I'll take you back to one of my trips to Southern California,
I was flying from San Diego back here to the San Francisco Bay Area. And this was
early in 2021. So February, March kind of timeframe. And I remember flying over the
Port of Long Beach, looking at all the vessels lined up. And that was the clearest
visual representation of the public economic data. And what I would say is, it was
incredible the amount of tug and pull on US GDP growth that we saw from the collapse
of inventory, the rebuilding of inventory. You know, I rarely talk about an esoteric
concept we call real final sales to private domestic purchasers, essentially, its
core GDP growth, right? How are consumers, businesses and government spending and
investing? But I had to talk about that measure, because the inventory cycle was so
inconsistent with core demand in the economy, they were so out of sync for so long,
that you, it was sending false signals about the strength or weakness of the economy,
depending on the quarter. So that was an incredible time, where one we learned just
how globally interconnected the world is, and how, when we're doing better, and everybody
else is doing better, it uplifts us all. Because, whether it is the invasion of Ukraine,
whether it is COVID restrictions in China, every single one of those policy actions,
physical actions has global ramifications, not just for the global economy, but for
on the ground individual business owners who are trying to manage through that lack
of clarity. So I think the scarring of what we went through with supply chain is going
to be with us for a while, and it's gonna have some very important implications for
business management going forward.
Matt Waller 27:11
Michael, I know you've done extensive research on the consumer side of the economy.
Would you mind talking a little bit about some of the medium to long run trends that
you're seeing on the consumer side?
Michael Brown 27:24
Well, I think there's probably two broad buckets. One is kind of the demographic shifts.
And right now we're in this phase where a boomers are kind of moving into their retirement
years, they're leaving the labor force, frankly, post pandemic a little bit earlier
than we had anticipated. And that starts having some profound impacts on aggregate
consumption. And I'll give you one example. In economics and macro, we talked a little
bit about this concept of the wealth effect. So as one's value of its of their home,
or their stock portfolio rises, there is some knock on effect on aggregate consumption.
And historically, you know, if you look at the economic literature, it's been range
bound. I mean, depending on the studies you look at, it's between five cents and 10
cents, meaning that for every dollar of wealth, that increases, you tend to spend
about five cents or 10 cents. Now some of this is psychological, right, I feel better
about my financial situation. And therefore, I can go out and spend a little bit more.
But what we're seeing is this wealth effect, as we measure it, has now grown to roughly
around 34 cents. So it's a profound shift in the deviation from what we've seen over
the last decade and a half. And the reason that's so important is because you start
putting the pieces together, we accumulated a tremendous amount of housing and financial
wealth over the last five years as an economy. The second key aspect is we are much
more sensitive to changes in stock markets today than we were, say, 20 years ago.
The third piece, as I mentioned, is you look at the share of the population over the
age of 55, who has retired early, it's incredible post pandemic. And so it makes some
sense that this wealth effect and wealth accumulation is having some impact as well.
So that's a big trend that we're looking at for the next five years as the retirements
continue to accelerate in the coming years. And so I think what we're about to see
over the next 10 years is the evolution of working at just a tech company to that
permeating multiple other sectors of the economy. And you're already seeing this in
financial services with automated trading and trying to enhance productivity, and
that's that sector. And I think, you know, we're just now seeing the tip of the iceberg
as this tech stack moves from Silicon Valley, as I like to say out to the rest of
the country, to kind of permeate some of these industries and that enables purchasing
power of consumers in a whole new way. Any new markets that I think we're just now
starting to get a handle?
Matt Waller 30:04
Well, Michael, this has been really interesting. Congratulations on your tremendous
success in your career and for taking time to visit with me about this. I really appreciate
it.
Michael Brown 30:17
It's been fun Matt, always great chatting with you and very, very pleased to always
support the Walton College here as well.
Matt Waller 30:25
On behalf of the Sam M Walton College of Business, I want to thank everyone for spending
time with us for another engaging conversation. You can subscribe by going to your
favorite podcast service and searching. Be epic be E P IC.
Transcribed by https://otter.ai
Matthew A. Waller is dean emeritus of the Sam M. Walton College of Business and professor of supply
chain management. His work as a professor, researcher, and consultant is synergistic,
blending academic research with practical insights from industry experience. This
continuous cycle of learning and application makes his work more effective, relevant,
and impactful.His goals include contributing to academia through high-quality research and publications,
cultivating the next generation of professionals through excellent teaching, and creating
value for the organizations he consults by optimizing their strategy and investments.
