Listen On:

Moody's Talks - Inside Economics

Episode
/
March 22, 2022

Data Deep Dive: Gross Domestic Product

Mark, Ryan, and Cris do a deep dive into GDP. What is it? How is it measured and what are it's shortcomings?

Full episode transcript.

Kennedy notably outlined why he thought the gross national product was an insufficient measure of success.[Note 1] He emphasized the negative values it accounted for and the positive ones it ignored:[6]

Even if we act to erase material poverty, there is another greater task, it is to confront the poverty of satisfaction - purpose and dignity - that afflicts us all.

Too much and for too long, we seemed to have surrendered personal excellence and community values in the mere accumulation of material things. Our Gross National Product, now, is over $800 billion dollars a year, but that Gross National Product - if we judge the United States of America by that - that Gross National Product counts air pollution and cigarette advertising, and ambulances to clear our highways of carnage. It counts special locks for our doors and the jails for the people who break them. It counts the destruction of the redwood and the loss of our natural wonder in chaotic sprawl. It counts napalm and counts nuclear warheads and armored cars for the police to fight the riots in our cities. It counts Whitman's rifle and Speck's knife, and the television programs which glorify violence in order to sell toys to our children.

Yet the gross national product does not allow for the health of our children, the quality of their education or the joy of their play. It does not include the beauty of our poetry or the strength of our marriages, the intelligence of our public debate or the integrity of our public officials. It measures neither our wit nor our courage, neither our wisdom nor our learning, neither our compassion nor our devotion to our country, it measures everything in short, except that which makes life worthwhile. And it can tell us everything about America except why we are proud that we are Americans.

If this is true here at home, so it is true elsewhere in world.

 

Follow Mark Zandi @MarkZandi, Ryan Sweet @RealTime_Econ and Cris deRitis on LinkedIn for additional insight. 

Mark Zandi:                      Welcome to Inside Economics. I'm Mark Zandi, Chief Economist of Moody's Analytics, and this is a deep dive podcast. We started this deep dive into different statistics back now a couple months ago with the consumer price index, where we took a really hard look at the DNA of that particular series and data, and try to give the listener some real context around what we're measuring. Today's deep dive will be around GDP, gross domestic product. Of course, to help me do this forensic look at GDP, I'm joined by Cris deRitis. Cris. Hello. Cris is the Deputy Chief Economist. How are you?

Cris deRitis:                       Very well. How are you Mark?

Mark Zandi:                      I'm good. I'm good. Safely ensconced in your office in Westchester, I see.

Cris deRitis:                       Yes. Yes.

Mark Zandi:                      Yeah. Very good.

Cris deRitis:                       Are you back north here?

Mark Zandi:                      I am. I made my way back from Florida last week. Did it in one day, 16 hours.

Cris deRitis:                       Nice.

Mark Zandi:                      Five pit stops, my wife and two dogs, and it would've been flawless, but we got trapped in traffic around DC, which cost us. But, but I could, I might have said this before. I could be a truck driver. I'll just give me my Wawa coffee. Although, it's getting a little expensive to be a truck driver these days.

Cris deRitis:                       Mm-hmm (affirmative).

Mark Zandi:                      But I could do that, but I'm back. Yep. Also Ryan, Ryan Sweet. Ryan is the Director of Realtime Economics, and obviously in deep with all this data, into the bowels of the data, so good person to have talk about GDP. So, Ryan, why don't I turn the conversation over to you? Why don't you give us a summary of GDP, gross domestic product

Ryan Sweet:                      Sounds good. So, GDP, gross domestic product. This data comes from the Bureau of Economic Analysis. It's frequency is quarterly, but taking a step back, the definition of GDP is the value of all goods and services that we produce within a nation's border in a given year. So, it sounds like an easy definition, but there's a lot of key important aspects to that. First, we're only measuring final goods and services, so intermediate goods aren't counted in GDP. It's got to be produced within our country's borders, and it has to be within a given year. So, when we think about GDP, the equation that we kind of hammer home is C plus I plus G plus NX. So, GDP is a function of, or is a sum of consumption, so, what you and I spend on goods, services, non-durables, that's consumer spending. Business investment, and business investment can include non-residential structures investment.

                                             So, think manufacturing plant, a new office building in downtown Westchester. So, it includes non-residential structures. It includes residential structures, and that includes, new homes, broker's commissions, multifamily apartments, and then the final part of investment includes inventories. Then, G is government, and think of the government as a big consumer. They buy hammers. They buy, they also buy planes, boats, trucks, so they're a big, gigantic consumer. Then the final part is NX, and NX is exports minus imports. That's how we get GDP, C plus I plus G plus NX. We can riff on this later, but there's all different ways of how you can measure GDP. You can look at the expenditure approach, the production approach and the income approach. We'll probably tackle each of those a little bit separately, but I'll turn it back over to you and Mark, and see if what I missed.

Mark Zandi:                      Well, you just described the expenditure-

Ryan Sweet:                      Correct.

Mark Zandi:                      Approach, right? So, you added up spending by consumers, by businesses and by government, and added that up to a GDP. The difference between though, consumption spending and output is inventories, right? So, that's the swing in inventories. You didn't mention that, but that's a pretty important swing factor, particularly quarter to quarter, it can make a big difference. That's been the case, certainly during the pandemic. In the last couple quarters inventories have been a key source of growth in the economy, driving a lot of the GDP, the output. So, inventories closes the gap between the expenditures, the spending and the output.

Ryan Sweet:                      Correct.

Mark Zandi:                      Maybe you can take a minute while we're at it, just talk a little bit about the other two ways of adding up to GDP, the income side of the accounts, the GDP accounts and the production side of the accounts. These all things, these all should add up in theory. Obviously, they don't add up in practice, because there's so many moving parts here and data issues that, there are so-called statistical discrepancies, between these different ways of measuring things, but at least in theory, they add up. So, what are the other two approaches to adding up to GDP?

Ryan Sweet:                      Yeah, we were always taught that in Intro to Macroeconomics that GDP equals GDI, which is gross domestic income, so one person's spending is another person's income. But, when I became a professional economist, I realized they don't add up. I've been, I was duped, in my Principles of Macroeconomics class, but GDI is just adding up the income side of the economy.

Mark Zandi:                      You sound like a conspiracy theorist.

Cris deRitis:                       Yeah.

Ryan Sweet:                      I was duped.

Mark Zandi:                      I was just cogitating around that for a second.

Ryan Sweet:                      I teach this to my students and I tell them, I was like, your textbook's going to say that one person's spending equals another person's income, and it doesn't when you're looking at GDP versus GDI.

Mark Zandi:                      Because of the [crosstalk 00:06:18].

Cris deRitis:                       Measurement issues.

Ryan Sweet:                      Measurement issues. There's also methodology. They're just different ways. Like, of course they're not going to be identical, but-

Cris deRitis:                       But theoretically, I mean, it's a construct, right?

Ryan Sweet:                      Yeah. Yes. In theory, yes.

Cris deRitis:                       In theory. Sure.

Mark Zandi:                      Anyway, I interrupted you. So, what goes into gross domestic income, GDI?

Ryan Sweet:                      This is the income side. So, some of the things, not all of them, are going to be wages, corporate profits, dividends, interests, rents. These are all things that will factor into the GDI part of the equation.

Mark Zandi:                      Have you noticed that the corporate profits, which is a component of GDI, so you, as you mentioned, wages and salaries and compensation of labor, plus what businesses earn in terms of corporate profitability, are they, and there's a lot of other smaller components that go up to adding up to overall gross domestic income, that corporate profits is a share of national income is at a record high? Did you notice that?

Ryan Sweet:                      Mm-hmm (affirmative)

Mark Zandi:                      That's just incredible. Given the rise in labor costs that we've seen, given the rise in material costs that we've seen, energy costs and materials related to the pandemic and supply chain disruptions, despite all of that, businesses have been able to obviously raise prices enough to compensate for their higher costs and their share of the economic pie, I think, I don't have this exactly right. But I think it's almost 15% of the national income now goes to businesses, corporations through corporate income, through profits. That is as high as it's ever been in the data. Is that just, did you guys know that or is that news to you? You knew that.

Ryan Sweet:                      I knew it was high. I didn't know the exact number, but I knew it was large and growing,

Mark Zandi:                      It could be 14.7. I don't remember.

Ryan Sweet:                      Okay.

Mark Zandi:                      But yeah. It rounds to 15.

Ryan Sweet:                      Yeah. It's high.

Mark Zandi:                      Typical is more like 12. I'm, again, I'm making this up, because I didn't really look, but 12, 13%, something like that, but that doesn't sound like a big difference, but that's a pretty big difference in terms of share of national income.

Ryan Sweet:                      Yeah.

Mark Zandi:                      Maybe that's why the stock market's hanging in. Have you noticed? The stock market, no matter what's being thrown at it, Russian invasion, and Fed going on the war path, and now talking about raising rates very aggressively, stock market is kind of hanging in there. It's only down what six, seven, 8% from its all-time high, back at the beginning of the year. So, pretty, pretty amazing. I guess that fundamentally, the stock prices are tied to corporate profitability and that remains very, very good. Businesses are able to pass through their cost to consumers.

Cris deRitis:                       Yeah, and that shouldn't change any time soon.

Mark Zandi:                      You don't see that changing any time soon?

Cris deRitis:                       Mm-mm.

Mark Zandi:                      Yeah. [crosstalk 00:09:12]

Cris deRitis:                       Well, it'll grow into these valuations. Right.

Mark Zandi:                      What's that Cris?

Cris deRitis:                       So, we'll grow into these stock market valuations is the idea, right? Profits will continue to grow.

Mark Zandi:                      Yeah. Yeah.

Cris deRitis:                       So what if the PE ratios are extremely high?

Mark Zandi:                      Yeah, I mean.

Cris deRitis:                       Profitability will continue, right. So the profit [crosstalk 00:09:32]. The earnings will continue,

Mark Zandi:                      Yeah. I guess-

Cris deRitis:                       Unless the FED really does go on the war path and then that's another way you can bring valuations down quickly. [crosstalk 00:09:41]

Mark Zandi:                      The mind-bending thing, if the stock market doesn't go down, then the Fed's going to have to step on the brakes. Right? Because it's counting on a tightening of its so-called financial conditions, weaker stock prices to do some of the work for them, to slow growth. But if that doesn't happen, then Fed's going to have to do even more. Right?

Cris deRitis:                       Yeah. I agree.

Mark Zandi:                      Yeah, okay. Okay. Very good.

Cris deRitis:                       For sure.

Mark Zandi:                      Okay. So, that, we got the expenditure side of the accounts. We got the income side of the accounts and then, okay. Now the output side of the accounts. What's that? Ryan?

Ryan Sweet:                      I can let Cris do that one.

Mark Zandi:                      Okay.

Ryan Sweet:                      I'm really hogging the other two.

Mark Zandi:                      You see the way he said that, because that's an easy one to explain. So, we'll let Cris do it.

Ryan Sweet:                      Let's let Cris do the difficult one.

Cris deRitis:                       He kind of already did it. So, it's the production approach, or the value-added approach. So, it's just the gross value of output minus the intermediate, the value of the intermediates.

Mark Zandi:                      Right. Right.

Cris deRitis:                       It's easy to explain, but it's probably the most difficult to actually implement. That's why the BEA favors the expenditure approach. They view that as the most reliable of the three.

Mark Zandi:                      Yeah. To measure it, it's hard, hard to measure it, I guess.

Cris deRitis:                       Yeah. Yeah.

Mark Zandi:                      I guess it's easy to, well easier ... nothing's easy, but easier to measure output of a factory or a mine, or maybe even agriculture, but when you get to the service side of the economy, like who can measure what Ryan actually does? What's his output? I mean, just think about that for a second.

Ryan Sweet:                      It's priceless. It's priceless.

Cris deRitis:                       Of course it's priceless. Right.

Mark Zandi:                      It's lag too. Right? The output side of, the expenditure side of the accounts and the income side they're well, the expenditures comes out most, and I think this is why we really focus on it. It's very timely. Well-

Cris deRitis:                       More timely.

Ryan Sweet:                      More timely. Yeah.

Mark Zandi:                      Yeah. Income side is lagged about a month, because corporate profits, you can't, they can't be a Bureau of Economic Analysis. You can't get that together as quickly. Ten on the output side, that takes time for them to, BEA, to measure all that and get ... so the timeliness also plays a role in what we are focused on, I think.

Cris deRitis:                       Yeah, You're right.

Mark Zandi:                      Yeah.

Cris deRitis:                       Yeah. When we look at this data. Yeah. Okay.

Mark Zandi:                      So, Ryan, you have a tracking, so-called tracking estimate of GDP. We've talked about this on the podcast before. You want to explain that in what you're doing, because that, it forces you to get into, really into the bowels of the GDP accounts to be able to do the tracking. So, do you want to explain the tracking?

Ryan Sweet:                      Yeah. I didn't know how deep we wanted to get into this, but we'll go, this will be like a deep, deep dive.

Mark Zandi:                      This is called a deep dive. I we'll

Ryan Sweet:                      We'll go deeper.

Mark Zandi:                      Okay.

Ryan Sweet:                      All right. So, we have a US high-frequency GDP model, and GDP coming from the BEA is reported quarterly. But throughout the quarter and over the several months, source data, so data that we know that feeds into the BEA's methodology for certain components of GDP are released. So, we can kind of create a bean counting approach. So, we can talk, kind of take in the new data on construction spending or personal, real consumer spending, retail sales, industrial production or things that we know in the, where they feed into GDP. Then each day, each business day, we run the model and it tells us the new data on ... So, for example, we got retail sales recently. What does that do to current quarter GDP? We report that on a daily basis. We have a good idea, roughly a very good idea of what GDP's going to do as the quarter evolves.

Mark Zandi:                      Right. Right now, for the first quarter of 2022 you're tracking, here we are in kind of mid, late March, so almost the end of the quarter. Of course, the data, the monthly data that comes in that you're using for the tracking estimate are lag. So, we won't get full data from March until April or May in many cases.

Ryan Sweet:                      Correct.

Mark Zandi:                      But what's the tracking estimate now for Q1 2022? What's GDP going to do in the first quarter?

Ryan Sweet:                      Just a hair above 1% at an annualized rate. We're below, we're closer to half a point recently, but some of the data has come in a little bit better. Now, the tracking estimate will be volatile, because there's inherently in the monthly statistical data there's a lot of volatility in it, particularly early in the year when you have, there's potential seasonal adjustment issues. Our current high frequency GDP model, that estimate, that one, a little bit above 1% doesn't factor in any impact of Russia's invasion of the Ukraine. So, that data will start to show up, that impact in probably late March, April data. So, maybe some in Q1 will be affected by what's going on in Europe, but that effect on the US economy will be more noticeable in the second quarter, using the source data.

Mark Zandi:                      I have to say, usually GDP is a pretty good barometer of how the economy's doing, and how you kind of think about the economy's doing. Right now, my sense is the economy is booming. I look at the job market. A half million jobs, every single month, but then I look at GDP of just a hair over 1%, that's pretty weak. It just feels is incongruous with the reality of what's going on.

Ryan Sweet:                      But keep it in mind that-

Mark Zandi:                      To some degree, that reflects a very weak conditions coming into the quarter, when Omicron hit back in December, in January. So, you started from a very low level, and so that might be part of what's going on, but do you have that same kind of feeling that this GDP number just doesn't feel right given what's going on in the economy?

Ryan Sweet:                      Yes. Partly because, if you look at what's dragging the economy down in the first quarter, it's one component. It's inventories. So, we had a big inventory build in the final three months of last year, and inventories are going, we're not going to be able to duplicate that. So, inventories, for example, say they went up by $100 billion at an annualized rate. We have to increase inventories in the first quarter of this year, by at least 1000 billion or inventories are going to be a drag on GDP, because just the way the BEA does it is the change, and the change in inventories that matters for GDP. We're setting up for a weak inventory build in the first quarter, and that's going to pull GDP down by quite a lot. So, a lot of the weakness is inventories.

Mark Zandi:                      Well, that goes, something more fundamental, and that it's the expenditure versus production. So, what you're saying is spending is going to be stronger than production, right? So, and again, GDP is production output, and we're spending more, and that's why we ... It feels better I think, sort of, you look at retail sales and jobs and everything all, but it, the actual production is lower, and that goes to the, it's reflected in the reduction in inventory, or less inventory accumulation. Right? So, it's a difference between output, is between the difference between expenditure and production, the swing.

Ryan Sweet:                      Exactly.

Mark Zandi:                      Okay. All right.

Ryan Sweet:                      But one thing you can do is, you can adjust GDP, so you can look at something called real final sales. So, you can take GDP and exclude inventories, and then you can look at real final sales to domestic purchasers. You can look at GDP minus inventories, minus net exports. So, that's kind of a proxy for the domestic economy, and I think when you look at that in the first quarter, that's going to be really booming.

Mark Zandi:                      Ah, that's a good point. Yeah. You don't track that though? It's not part of your tracking estimate?

Ryan Sweet:                      Real final sales to, yeah, I, yes, we do.

Mark Zandi:                      Oh, so what is that, do you think is how-

Ryan Sweet:                      I have to check.

Mark Zandi:                      You got to check. Okay.

Ryan Sweet:                      Yeah. But if you look at real GDP, excluding inventories, will be much stronger than just 1%.

Mark Zandi:                      Got it. So ,real final sales are, is going to feel more like how I feel about the economy. This is going to be strong.

Ryan Sweet:                      Correct.

Mark Zandi:                      Yeah. Okay. Very good. Well, this gets to another issue and this gets to revisions. Cris, do you want to talk about the revisions today? This is, this GDP numbers has a quarterly periodicity, but it's updated every single month. Therefore, every month you get revisions to the data, and then you get some longer-term revisions. You want to describe those revisions, Cris?

Cris deRitis:                       Yeah. High level. I don't know where exactly you're going here, but certainly don't-

Mark Zandi:                      Nowhere in particular. We're talking deep dive, Cris.

Cris deRitis:                       Yeah, I get. Sure, we can go deep, deep as you want, but we're talking. I think what's clear from all these definitions is that they are estimates. Right? We talked about measurement error. We even talked more fundamentally about how you define these different items. So, the economists at the BEA are doing their best to estimate each of these components, based on the data that they have, and it's from a wide variety of sources. I mean, if you really get into it, you recognize the complexity of this problem. You're combining lots of different data, survey data, that you're, there's all sorts of data that's, that is getting updated and revised as time goes on. That then, causes their estimate for output within a given quarter to be revised and updated as well. So, the process is to continuously update with data as it comes in, and before we arrive at a final GDP estimate.

Ryan Sweet:                      Well, that final estimate is years and years and years down the road to-

Cris deRitis:                       Fair enough, friend. Fair enough.

Mark Zandi:                      Because every year, well, they have an annual revision-

Ryan Sweet:                      Correct.

Mark Zandi:                      Where they bring in even more, what they call source data, so underlying data.

Ryan Sweet:                      Every August.

Mark Zandi:                      Every August, and then they have what they call a comprehensive revision, don't they, or am I mixing things? That happens every-

Ryan Sweet:                      I think it's called a comprehensive benchmark revision.

Mark Zandi:                      Comprehensive benchmark revision. Is that's every three years, I believe.

Ryan Sweet:                      I think it's five.

Mark Zandi:                      Is it at five, three or five years?

Ryan Sweet:                      Cris, what do you think? Cris is-

Cris deRitis:                       I thought it was five, too. That was my guess. [crosstalk 00:19:50]

Mark Zandi:                      Okay. We probably should.

Ryan Sweet:                      That's when they can make methodological changes-

Cris deRitis:                       Yeah. Yeah.

Ryan Sweet:                      So, they can make adjustments. I mean, I remember this is going a way back, but intellectual property had never been part of GDP until they added it, years and years ago. So, that's when they can make big changes to their approach to measuring GDP.

Mark Zandi:                      Yeah. I remember the intellectual property. There was, right now a big chunk of business investment is in so-called, well there's, you mentioned structures, equipment, and there's now intellectual property. In there, there a potpourri of things, including R&D, software. I think even movies, the production of movies is in there as well. That kind of a thing. That never, that wasn't there, if you, I think, if you go back certainly 10, 15 years ago, and now that's a very large component of investment, and a very fast-growing component of investment. It's a very important part of investment in spending. [crosstalk 00:20:47] So, yeah, I agree. Big changes.

Ryan Sweet:                      So, this's a good segue to some of the criticisms, right? One and one big criticism is that it GDP doesn't capture all the activity in the economy, right? So, one reason why we have these methodological revisions is to try to improve it over time, right? So, you mentioned that you, some of the intellectual property issues, and hopefully we're getting better at capturing those. Black market activity is always a good point of debate, right? Whether you like it or not, there are activities in the economy going on. There is output or production of drugs or other activities, for example. Should you include those in GDP if you're trying to really get a comprehensive measure of all the activity in the economy that's going on? Probably, yes, but then measuring that is even more difficult. So, lots of lots just of issues here. The other major issue is what unpaid household activity, or household production, right. That's not captured because it doesn't have a market value, right?

Mark Zandi:                      It's certainly output. I mean-

Ryan Sweet:                      Certainly output.

Mark Zandi:                      Give that a shot for a while. That's pretty hard to do.

Ryan Sweet:                      Yeah, absolutely. Yeah, absolutely. I think by some estimates, right, it can, maybe it's high. It might be as high as 50% or-

Mark Zandi:                      Oh, is that right?

Ryan Sweet:                      I think it's substantial. Well, that's my, I'm thinking back to my, my, yeah, exactly.

Cris deRitis:                       Based on my experience last week, it's more than 75%.

Mark Zandi:                      You had a lot of household. You missed the podcast last week, I think because of household-

Ryan Sweet:                      Daddy daycare.

Mark Zandi:                      Daycare. It was-

Ryan Sweet:                      There you go. Unpaid.

Mark Zandi:                      Unpaid, unpaid household output.

Ryan Sweet:                      Yes.

Mark Zandi:                      It's really not getting into the GDP accounts, that's for sure.

Ryan Sweet:                      Yeah. Definitely under count.

Mark Zandi:                      Well I guess there's even broader conceptual issues, like for example, does GDP really reflect welfare? I mean-

Ryan Sweet:                      Right.

Mark Zandi:                      Is that really a good measure of the, how the economy is performing? It's kind of, we're counting the widgets. Yeah. But is that really the best way of measuring how much, how, how well our economy's actually doing for the population? I think that's a pretty deep question and there's-

Ryan Sweet:                      Yeah.

Mark Zandi:                      Some countries have, like France, I think, have tried to with, with Nobel Laureate, Joe Stiglitz, have come up with different measures of, some variations on the theme of GDP to try to get a broader sense of how economies are doing.

Ryan Sweet:                      Yeah. As I recall, again, going to the memory banks, with Simon Kuznets, right, who was the father of GDP or the modern GDP back in the '30s, I think he was the main critic, the first critic of GDPs. Said, I've created this measure, but I don't want to your point, I'm counting the widgets here, but it's, GDP's not everything. You shouldn't use this to determine whether your economy is operating efficiently, or providing all the benefits that it could, because it is missing a lot. It's also ignoring extraction or, now we talked about an environmental degradation, right? That's not captured. How you're actually producing all this output isn't captured here. It's not capturing people's health or the quality of their lives or political liberty. So, there's a lot of, a lot of missing parts here. It's a measure, but it's, it certainly shouldn't be the supreme measure, and unfortunately I think that it often gets that place in people's minds. Right?

Mark Zandi:                      Take it for granted. Oh, that's how we're doing.

Ryan Sweet:                      Right, right.

Cris deRitis:                       Yeah.

Ryan Sweet:                      That's the yardstick.

Cris deRitis:                       A lot of this comes up around naturals disasters. So, anytime you have a big hurricane-

Ryan Sweet:                      Yeah.

Cris deRitis:                       Say, oh, it's going to be a net positive for the economy because basically, you get a lot of destruction of property, but that destruction doesn't show up in GDP, but the rebuilding will. So on net, when you say it's a net positive people like, well that sounds wrong. You know, Hurricane Katrina did a lot of, Super Storm Sandy, that did a lot of damage, and people weren't better off after those storms. I was like, there's a difference between economic welfare and GDP.

Mark Zandi:                      Well that's, that was a methodological change too. Right? I mean, I remember back in the day BBA did the accounting around natural disasters differently, I believe, and they did account. I think so. The insurance losses were actually a deduction from GDP. They tried to account for the loss. I believe, I believe that's because that's, do you know Ryan? Do you know that's the case or not?

Ryan Sweet:                      They have, I don't know about insurance losses, but I know they have an estimate of, around natural disasters, around the lost output.

Mark Zandi:                      Yeah. Yeah.

Ryan Sweet:                      So, the economic cost of-

Mark Zandi:                      Right.

Ryan Sweet:                      The natural disaster.

Mark Zandi:                      We should actually get Adam Kamins on. He's our ex-resident expert on natural disasters, and all the economic accounting that goes around it. But you can see, the listener, you hear this conversation. This gets really, pretty complex, you know exactly how you should measure these things and handle them. That's the other point, the other kind of missing ingredient is the distributional effects of GDP. So, there has been some efforts to construct so-called distributional accounts. So, what is GDP for different parts of the income and wealth distributions, because you get a better sense of who the, who's benefiting from the output or the GDP. I think there's some moves afoot in DC to try to, like the Federal Reserve has actually created so-called distributional financial accounts, so they can take a look at assets and liabilities of households by different parts of the income and wealth distribution.

                                             So, you can see who's getting wealthier and who's not getting as wealthy, and what share is going to the top part of the income distribution, what share's going to the bottom part, that kind of thing. Which is really important in a world where income and wealth inequality has become more pronounced. So, there's been an effort to try to do this for the GDP accounts, but pretty tough to do, but certainly auditable.

Ryan Sweet:                      Yeah. You already highlighted the capital versus labor split, right?

Mark Zandi:                      Yeah.

Ryan Sweet:                      That's one obvious way we can see how things have migrated over time. So, yeah, that's really important.

Mark Zandi:                      That's the other thing to point out that, given all these measurement issues, methodological issues, GDP from, in the United States, it could be very different from GDP in France or GDP in China or anywhere else on the planet, because the statistical agencies, they have different qualities of source data, timeliness of source data. They use the data in different ways, conceptually and methodologically, to come up with their own estimate. So, we all kind of compare GDP here with somewhere else, but you do that at your own peril, right? Particularly, places that have pretty thin statistical kind of infrastructures for collecting information and data,

Cris deRitis:                       You also have to, you have to normalize it.

Ryan Sweet:                      So, you can't compare the level of GDP in the US to the level of GDP in China or the level of GDP in [inaudible 00:27:58]. You're looking at GDP per capita.

Mark Zandi:                      Yeah. It-

Ryan Sweet:                      To make it closer to apples to apples comparison to-

Mark Zandi:                      Apples to apples. Yeah. Yeah

Ryan Sweet:                      I mean, there are some international standards though, right, that broadly define-

Mark Zandi:                      Yeah, yeah.

Ryan Sweet:                      How these should be an OECD type of standard, but interpretation is up to the individual countries, and what data they might have available, how they estimate. So yeah, lots of variation.

Mark Zandi:                      There's a lot issues, like for example, price deflators, that's measuring prices for things, right? Because we get nominal GDP, which is, just in dollars or euros or whatever, pounds, whatever it is. Yeah. But then you try to account for inflation over time to get to kind of the real output, not the output related to price increases or declines. So, therefore, you got to start measuring price movements. We talked about this when we did the deep dive for CPI, but this gets even more complicated for GDP, because now you're measuring prices for investment goods, for example. So, you're trying to measure investment in information processing equipment, computer equipment. If you do that, you can count the nominal dollars that Apple sells or IBM sells or Dell sell, but then you got to try to account for the price changes, which by the way, you have to account for the quality changes, right?

Ryan Sweet:                      Mm-hmm (affirmative).

Mark Zandi:                      So, then now you're into, this is really mind-numbing. You're, because you can get big increases in real output if you have very fast declining prices for investment goods related to fast technological change. In fact, that was what was happening back when the internet was first getting going in the late '90s and early 2000s, we had very, very fast innovation, technological change that drove down prices, measure of prices because of the quality improvement, which drove up real GDP. We saw, oh my gosh, we're booming. This was the boom times, and it was largely because of the way we were measuring quality changes in the price deflators.

                                             I mean very, very, very difficult to do. Very hard to do. It makes it very difficult. Okay. The other thing I wanted to mention is Ryan, you also construct monthly GDP.

Ryan Sweet:                      We do.

Mark Zandi:                      GDP is, at least in the United States and some countries like the UK and I believe Canada, they actually construct GDP on a monthly basis. We don't, the BEA does not do that here, but you've taken it upon yourself to construct an estimate of monthly GDP. Do you want to describe that?

Ryan Sweet:                      We Maverick.

Mark Zandi:                      Yeah. Right. Right.

Ryan Sweet:                      Now, it's not like I'm sitting in my basement doing this. So, a very similar, like or high [inaudible 00:30:41], our high- frequency GDP model, trying to track current quarter GDP. We also have simultaneously, an estimate of real monthly GDP. So, GDP comes out quarterly, but we know the source data that is available at a monthly frequency, so we add them all up and we create an estimate of real monthly GDP in the US.

Mark Zandi:                      The, I think the last data point is for January, right?

Ryan Sweet:                      Correct.

Mark Zandi:                      Okay, and that showed a pretty big decline, if memory serves.

Ryan Sweet:                      Mm-hmm (affirmative) Yeah, we got off to a really-

Mark Zandi:                      [inaudible 00:31:14] earlier conversation. Yeah.

Ryan Sweet:                      Omicron, weather. Yeah. The economy got off to a really slow start to the first quarter/

Mark Zandi:                      Right, and even though it feels like it's picking up here in February, and particularly in March, Omicron is faded, and despite Ukraine, Russia-Ukraine's invasion, it feels like things are strong. It's going to depress that quarterly estimate.

Ryan Sweet:                      Yeah.

Mark Zandi:                      That's the one little over 1% growth that we're getting.

Ryan Sweet:                      Yeah. We dug ourself a hole getting this quarter started. Also, on top of that, we have inventories that are going to be an enormous drag, trades going to be a weight. So, really, the domestic economy can be booming, but you get these offsets in other parts of the GDP accounts that will temper growth in the first quarter. I don't know if we want to go down this rabbit hole, but there's still some signs of residual seasonality in GDP. GDP is seasonally adjusted, meaning, we try to take away, we know that the economy doesn't do really well in the winter. We adjust for that. But there are still, within some of the detailed signs of residual seasonality, meaning that they're still not capturing all the seasonality that is occurring in the economy.

Mark Zandi:                      Oh, that's interesting. Yeah. I knew it was a problem in the employment data, or at least it felt like it was a problem in the employment data, but it's also bled into the GDP data as well, you were saying.

Ryan Sweet:                      It isn't a much bigger issue, five or six years ago. When we were starting to really launch and do the high-frequency GDP model. So, we had to put in adjustments, so so-called dummy variables to account for these residual seasonality. So, each quarter had a dummy variable, whether or not seasonality was a problem. Then they, when they did the comprehensive revision, the benchmark revision, the most recent one, they tried to fix a lot of it, and they did fix most of it, but some is still there.

Mark Zandi:                      Right. Right. Now, on the monthly GDP, why do you suppose the BEA does not construct that? It's just a matter of resource that they don't construct the, an estimate of GDP every month?

Ryan Sweet:                      Because they know, that's my baby. [crosstalk 00:33:21] No one else is going to yeah, look at it.

Mark Zandi:                      Just do any better. Why would they even attempt to try that?

Ryan Sweet:                      I bet that's coming. I haven't heard, but I bet. I mean, Canada's doing it. The UK's doing it.

Mark Zandi:                      Yeah.

Ryan Sweet:                      They could probably do it a lot better than I'm doing it.

Mark Zandi:                      Yeah. I've been confused by that. Okay. Any other, did I miss anything on GDP, any other issues around GDP that we should alert the listener to, when we talk about this?

Ryan Sweet:                      No.

Cris deRitis:                       I mean, you can drill down into GDP. I mean, you have industry-level GDP, which comes from the BEA as well. So, there's lots of-

Mark Zandi:                      Wasn't, that goes back to the production side of the accounts, doesn't it?

Cris deRitis:                       Correct. It does.

Mark Zandi:                      Yeah. Okay. Oh, I guess I should also, we should also mention that they, there is, BEA also provides GDP estimates regionally, right?

Cris deRitis:                       Mm-hmm (affirmative).

Mark Zandi:                      For states and I believe metropolitan areas. You can get GDP now? I believe you can. Am I wrong? I guess something else we should check. Yeah.

Cris deRitis:                       We'll have to check that one. I know for states, you're 100% right on states. I don't know about metro areas.

Mark Zandi:                      That's relatively new too. That wasn't the case, back when I first started as an economist, so that's new as well. Okay. So, I think we've covered it, right? Any other things you want to alert the listener to on the GDP accounts, the so-called National Income and Product Accounts, we shorthand GDP, but it's really the National Income and Product Account from the Bureau of Economic Analysis. Oh, I did. Oh, one other thing I did want to mention, or just get the sense of, given all of the changes in private source information, we get data, companies are getting pretty good at collecting data. We work with ADP, for example, the human resource company, and we get information on, I think 25 or 6 million employees, every single month.

                                             We have a relationship with Equifax credit bureaus. We get a lot of credit file data that provides a lot of information. Every, all the big tech companies are capturing a lot of information and data. Do you have any sense of that getting incorporated into GDP accounts? Is the BEA working to do that? I'm sure that's a pretty, there's a lot of thorny issues around privacy and intellectual property, that kind of thing. But has there been any moves there? I have not been following that. Just curious if you know of any.

Ryan Sweet:                      It doesn't have to be with just with the BEA. It could be some of the source data that feeds into GDP. I know for a fact that some agencies are incorporating alternative measures of consumer spending, particularly on the services side, which is, it's easy to count the number of vehicles that get bought off a lot, but it's difficult to count how many hamburgers were purchased in Westchester. So, I think they are starting to use some alternative data on the services side.

Mark Zandi:                      Right. Cris, anything that-

Cris deRitis:                       Yeah. I would say, just say there's, there is a lot of nuances when it comes to GDP measures, and what the death physicians mean, and especially for people who aren't deep into the topic, right, even the terminology we use. Investment, right? A lot of people think that when they buy stocks in the stock market, right, oh, that's investment, but it's not. So, it's not in the sense of what GDP is trying to capture or measure. So, I just would advise if you are going to use this measure and use it in models or to make predictions, you really need to get into the depth, the bowels here, and really understand what all the different components actually mean.

Ryan Sweet:                      Yeah. Did we mention that financial securities aren't counted in GDP? So, like stocks, bonds, they're not counted.

Cris deRitis:                       I don't know if we did.

Mark Zandi:                      Right. I don't think we did. [crosstalk 00:37:01]

Cris deRitis:                       Well, why would they be counted in GDP?

Ryan Sweet:                      So, I do this exercise with my students. I'm like, where is GDP counted, and I give them different scenarios. I'm like, you buy 100 shares of GM stock. They're like, oh, that's investment.

Mark Zandi:                      Right.

Ryan Sweet:                      Because, they need investment. [crosstalk 00:37:14]

Cris deRitis:                       Oh, oh, oh, oh.

Ryan Sweet:                      I was like, you pay tuition to Westchester University. Where's that investment, mentally. So it's-

Cris deRitis:                       Got it.

Ryan Sweet:                      It's a fun exercise for them to learn.

Cris deRitis:                       Yeah.

Ryan Sweet:                      The way you think where things should show up, isn't necessarily where they are going to show up-

Cris deRitis:                       Where people actually show naturally think that they go, right?

Ryan Sweet:                      You buy a used car or an existing home versus a new car. Right. So, there are all these-

Cris deRitis:                       Those, both are on there.

Mark Zandi:                      That's a great point. People buy six, 7 million existing homes a year, but the only way that shows up in GDP is the output of the realtors that are, right?

Cris deRitis:                       The broker's commission.

Mark Zandi:                      That are actually doing the transaction. Yeah. It's, because you're not building anything. There's no new output. You're just transacting an existing home for another. New homes, the construction of that, that is output. That cuts into GDP, but the existing homes does not. Used cars- [crosstalk 00:38:05]

Cris deRitis:                       Yeah. To avoid double counting.

Ryan Sweet:                      Exactly. Like the first home we bought was, it was built in 1963. That got counted in GDP in 1963. When we bought it in 2010, it shouldn't be counted in GDP again.

Cris deRitis:                       Yeah. Shifting ownership doesn't matter. Right?

Ryan Sweet:                      Exactly.

Cris deRitis:                       It's just, and from a macroeconomic standpoint.

Ryan Sweet:                      Right, right.

Mark Zandi:                      Well great. Maybe last call. Anything else?

Cris deRitis:                       I was thinking to end with this, and maybe you know this Mark, there was a Kennedy quote around GDP where, and I can't remember who, which Kennedy, but something along the lines of GDP measures everything except what really matters in life. You remember this?

Mark Zandi:                      Yeah. I do remember that. Was that a Kennedy quote? I don't remember that.

Cris deRitis:                       I'm pretty sure, but okay. [crosstalk 00:38:52] I thought you would.

Mark Zandi:                      It's like one of those things, computers are everywhere, but in the GDP accounts.

Cris deRitis:                       Right, right.

Ryan Sweet:                      Yeah. That was out there for a while.

Mark Zandi:                      Oh yeah. Who was, who said that?

Cris deRitis:                       Was that Greenspan?

Mark Zandi:                      I don't think it was Greenspan. No, it was an MIT professor. How can I forget his name? Really good econ professor, MIT or Harvard. That went back to-

Ryan Sweet:                      Was it Katz? Right?

Mark Zandi:                      No, no, no, no. You would know. I just-

Ryan Sweet:                      Otter?

Mark Zandi:                      Arthur. No.

Ryan Sweet:                      No.

Mark Zandi:                      Okay, no. Nope. You would know. But anyway, that's an old quip that we could see computers taking over business, every aspect of business, but it didn't seem to show up in GDP.

Ryan Sweet:                      Right.

Mark Zandi:                      But all of a sudden it did.

Ryan Sweet:                      Right.

Mark Zandi:                      We got to some kind of, I guess, critical mass of computers and people, businesses really figured out how to use them effectively and got to some kind of tipping point, and that really then showed up in the data in a very significant way. But, nonetheless, it kind of highlights the difficulty in measuring anything like this. Okay. I think that's a pretty deep dive. I think we went pretty deep, and if listeners have any other additional questions, fire away, or if you want us to do a deep dive on any other particular statistic, let us know. That would be very helpful and we'll do it. So, with that, we'll call this a podcast. Thank you.