Into the Mystic: WSJ’s "most accurate forecaster" of 2023 on where the economy is headed

Download Transcript

Spencer Levy 
Over the last five years or so – a period marked by uncertainty and change – economic forecasting has been a humbling endeavor. It's a quantitative discipline that relies heavily on empirical data, of course. But it's also a qualitative enterprise subject to observation and experience. On this episode, we welcome two masters of this ever-evolving art and science. 

Belinda Roman 
It's listening to what people have to say and what you're seeing with your hopefully educated eye that helps you then look at the numbers and say, are these numbers really reflective of what's going on? 

Spencer Levy 
That's Belinda Roman, an associate professor of economics at Saint Mary's University in San Antonio, Texas, where she also has a consulting firm that provides regional, national and global impact analysis and forecasting. The Wall Street Journal named Belinda its most accurate economic forecaster of the past year, hailing her as, and I quote, “the college professor who got a weird year for the economy right”. 

Dennis Schoenmacher
Before we really dive into the models, you need to understand what type of data is available and what are the definitions of the data that you have available. 

Spencer Levy
That's Dennis Schoenmacher, CBRE’s Global Co-Head of Forecasting and Analytics based in London. Dennis is a leader of CBRE’s Econometric Advisors, the company's forecasting and analytics platform whose work translates macro-economic trends into client facing insights. Coming up: the X's and O's of economic forecasting, insights into the tools of the trade, the data sets, the unique perspectives, and what it all means for commercial real estate today. I'm Spencer Levy, and that's right now on The Weekly Take.

Spencer Levy
Welcome to The Weekly Take. And we're going to have a great episode today, starting with Belinda Roman, the number one forecaster in the United States, according to The Wall Street Journal in 2023. Belinda, thank you for joining the show. 

Belinda Roman
Thank you. Thank you for having me. 


Spencer Levy
And then we have our friend and colleague Dennis Schoenmacher. Dennis, thanks for joining the show. 

Dennis Schoenmacher 
Pleasure to be here Spencer. 

Spencer Levy 
Thank you Dennis. Let's start very big picture, starting with you, Belinda. First of all, congratulations on your award from The Wall Street Journal, being the number one forecaster in the United States. And I'm sure your phone is ringing off the hook, including from us. But just tell us, if there's any one tool or type of tools you think that distinguished you from the rest, from Wall Street and everybody else, what might they be, Belinda?

Belinda Roman
I actually think it all comes down to the qualitative work, because the quantitative work, I think you can get done relatively easily. You know, there's add-ins, there's platforms, there's all kinds of things that you can use in your spreadsheets or your R code or whatever you're using to do the technical work. But when it comes down to looking at the numbers and comparing it to reality, it's the qualitative part. It's listening to what people have to say and what you're seeing with your hopefully educated eye. That helps you then look at the numbers and say, are these numbers really reflective of what's going on, of what I'm seeing and reading and hearing? And so that's what I think I spend a lot more time on, is doing that qualitative work. 

Spencer Levy
Give us a little bit more color on that qualitative work. Are you walking around speaking to shopkeepers? Are you speaking to business leaders? Are you speaking to school teachers? Who are you speaking to and what are they saying that's influencing your opinion? 

Belinda Roman 
Yeah, it's a little bit of all of that because I do go around. I spent the weekend driving around looking to see how many foreclosures, seeing what signs are up, seeing what shops are opening, the names of the shops, talking with people. How are you doing? What's the footfall like? I do a lot of that, and I look for the same material in other parts of the U.S. as I'm reading, just to get a sense of what's happening on the ground as well, so that I can then look at my forecast and say, are my numbers reflecting what I'm hearing in general from the people that are actually doing the economy, rather than just reading about it? 

Spencer Levy
Dennis, tell us about the tools that you use, that you rely upon when you're trying to do your best econometric forecast. 

Dennis Schoenmacher 
Yeah, maybe taking a little of a step back first, because EA – Econometric Advisors – has a very rich history dating back decades. It was founded by two professors in the field of real estate economics, and we really take that academic approach to our work. We analyze all the macroeconomic factors together and then really leverage that into our property models. So we forecast rents, vacancy rates, completions and investment performance. These forecasts are conducted at various levels of geography, from stock market to national levels. All in all, in the U.S., we forecast nearly 4000 submarkets, sector combinations. While globally we cover close to 700 markets, and to enhance our short-term models, we use local knowledge to add facts to our models. And this has really been a game changer for us. For example, what we see in that local knowledge is that broker that, for example, knows that a building that's coming on the market is fully let, and this type of information really enriches our models. 

Spencer Levy
Well, two things there. First of all, you did mention the founding of your group by two professors. I will name them. Ray Torto and Bill Wheaton. And I will also say with some regret and just maybe a little bit of love; Ray was one of my mentors in this business, and he just passed away. And so I think Ray would be very proud that we're having this discussion, because he was one of the intellectual founders and just a terrific guy. Now we're going to get to the hard part. The last three years have been the most difficult forecasting years, maybe in the history of econometric forecasting. And I think the reason we all know is Covid, and we can use an analogy. I sometimes say it was like a meteor hitting the Earth. Or you could imagine the globe being a big ball of jello and somebody poked it. But that made forecasting difficult. What's your point of view, Dennis? 

Dennis Schoenmacher
I completely agree with you on that, Spencer. It's been extremely difficult given all the uncertainty we've seen over the last 36 months, and especially after Covid, where we saw such a sharp shock to the GDP, which of course is one of the variables that our models rely heavily on. And when there are shocks that are unexplained in your long term models, that definitely gives you a difficulty, doing out-of-sample sample forecasts. But what we have done over the last 36 months is really constantly looking at our different economic scenarios. And we don't only do that in the U.S. We recently also are applying those in Europe as well. But we really do that to gain a deeper understanding of how real estate markets will be impacted under different economic conditions. And these scenarios have proven to be invaluable in our discussions with clients as the markets change quite quickly. If we recall last year, October, when the ten year hit 5%, end of the year it was closer to 4%. These dramatic shifts are just very difficult to forecast, and the different scenarios have helped us navigate that landscape with our clients. 

Spencer Levy
So, Belinda, do you use this same scenario-type approach? Your base case, your worst case, your better case?

Belinda Roman
I do. What I do is I have different historical series that I go through and I say, well, let's go back to just before 2008, in the financial crisis, or go back to 2000 and kind of set it up that way. So I'm looking and I'm watching how different types of crises impact the economy so that as I get closer to Covid and I look at it and think about, well, what actually happened here? The financial crisis was a different impact than the Covid crisis because that was a mandated closure. And then we reopened. As you know, there were a lot of states that pushed back about it, and we came back very quickly. And so it was that bounce that I was interested in, as opposed to some of these longer term recessionary events. So I do several different models that way to see what I can discern from the particular shock. The outside, the factor that's causing these issues. 



Spencer Levy 
Dennis, has Covid changed the types of tools, maybe not the categories of tools, but the weight you put on certain tools because of some of the things that actually occurred? 

Dennis Schoenmacher
Yeah, I think that's a great question. And I agree with you that Covid had a significant impact on what we really do in terms of modeling. And what we do is we use, like, a dummy model construction for the times that we see Covid to really ensure that we correct for these type of impacts. And we still, in some instances, use dummy effects going forward to ensure we capture these type of situations in our model. But in general, I still remain that we don't incorporate too much geopolitics. We do a lot of research on them, and one I want to call out here, which is from the Peterson Institute for International Economics, they’ve done very valuable research of the impacts of increased tariffs on inflation. And there's a massive shock to inflation if the U.S. would impose tariffs on China. But, we give it more as an afterthought to our clients. And we don't include it into our modeling, per se. Definitely not in our base case scenarios. 

Belinda Roman 
I think that there are some markets that are more sensitive to geopolitical issues. And as a matter of fact, I was just down on the border for an economic summit with Mexico, because Mexico is our largest trading partner. The flip-flop between them and China, and they were talking about nearshoring, and they were talking about the biggest question on the table is, there's two elections, a president in Mexico, a president of the United States. How is that going to change policy and trade through the borders? And that includes data processing, oil and gas, and the possibility of a different way to cross between the Pacific and the Atlantic instead of the Panama Canal. So there was a lot going on there. And I thought, well, that makes the case for including some of this geopolitical activity that's going on, because we need to understand how the smaller pieces, when they get touched by this, how it bubbles up to the macro, and it may take longer than our models projections out. It may take a little bit longer, but it's still there. So that's why I always think about it, especially with oil and gas in some of the places that are really sensitive to it. 

Dennis Schoenmacher
I think you're making a fair point. And if I recall my last trip to the U.S where I presented at the ULI in New York, a thing that resonated with me from that particular event was we're going from global collaboration to global competition, and this will definitely have a lot of repercussions for how supply chains over the world will respond to this. And nearshoring/offshoring is one of the largest topics that we have spent a huge amount of time. Our industrial economist, Nicholas Rita, has spent a lot of time trying to quantify what the potential impact of this is. For example, on the industrial sector. So I definitely feel that that is going to play a role. And again, I think when we're looking at all the geopolitics that are happening right now, 2024 is also the year of a huge amount of elections across the globe, as well, that we need to take into account for what might happen. And I think we should expect the unexpected in a lot of cases, as well. 

Spencer Levy 
So, Belinda, let's bring it a little bit more local for just a moment. How do you see labor as being perhaps the greatest input into your thinking, or not? 

Belinda Roman
Labor is the baseline for the regional and local economic–when you do the equivalent of GDP for the macro, you're looking at regional gross domestic product, RGDP. And so you lose labor, and you lose labor and their wages and their salaries and the growth over time and how that's growing and how each industrial sector is changing. So it makes entirely 100% sense to me that we would be focusing on what's going on in the labor markets, and how are those labor markets related with each other, in particular across borders? Because changes in China are going to affect changes in Mexico are going to affect changes in the United States. So watching that chain of sequences I think is very important. And the other side of labor, of course, is the derived demand. So it's businesses needing to produce for the consumer. So those relationships are extremely important. 

Dennis Schoenmacher
Yeah, I think I would agree with Belinda on this. Employment is such an important driver of our local real estate models, disregarding for which sector we are forecasting. And again, I think what a lot of our focus is on, like, what will be the employment growth and the population growth for all these local municipalities going forward? We've done a lot of work on migration in our map of the months, in our chart of the weeks, where we spend a lot of time really digesting that U.S. census track data. Where are people going? From where to where? How much immigration are we seeing? What type of immigration are we seeing? And I just have seen that in some of the research that we've done, there might not have been enough immigration to the U.S., especially to fill some of those lower skilled jobs. 

Spencer Levy
When I take a look at our forecasts, and we've just talked about the big stuff: the geopolitics, the immigration, cross border, labor, we still got it wrong last year. And when I say We, this is the royal We, this is everybody except Belinda, who is on the show for a reason because she didn't get it wrong. But I think we got it wrong for one macro reason. And that is the level of government intervention, not just fiscally or monetarily, but also from a regulatory standpoint. Belinda, any way to have a crystal ball on government intervention and how it might impact your forecasting?

Belinda Roman
Yeah, that one's the hard one. And that was the hard one in my forecasting, that political volatility. What might be coming and how do we factor that in? And I have yet to find a proxy, something that I can say okay this represents Washington. This represents Austin. This represents, you know, the local governments, I've yet to find it. And that's where I think the qualitative work comes in. Sometimes I'll go through and I'll do some content analysis and sentiment analysis in some of the language to see if there's anything that's popping up there that I should be aware of. But I think that's where we really have to kind of think about what's been said and how to interpret what's being said. Like, everybody listens to Powell and saying, okay, what's he saying? What does he really mean? And I think that's where we use those skills. 

Spencer Levy
Dennis, when I took a look at our forecast last year at this time, we had predicted that interest rates would be about 100 basis points lower at both the long and the short end of the curve. We also predicted that in the out years that GDP would be about 100 basis points higher in the United States than what we see today, where we're coming at a clip along at a 2% kind of rate. We were clipping along around a 3% rate. That's an enormous change in our forecast. What changed to change that much in our outlook for the next several years? 



Dennis Schoenmacher
Well, I think if you look back at our forecast, Spencer, and this is also the time to be humble as a forecaster, we called for a recession that should have already been going on as we speak right now. On the basis of that, we really expected interest rates to come down and hence in the later phase of our forecast period to see higher GDP growth. However, I think what we have underestimated is what you just mentioned, that fiscal stimulus to the U.S. economy. If we look from a macro level, really to the globe, the U.S. has not really responded as a textbook economy to higher interest rates, whilst Europe has done that. In Europe, we've seen higher interest rate, lower GDP growth, and inflation is coming down quite quickly. In the U.S., We haven't seen that, and that's all to do with stimulus. A recent IMF study highlighted that most of GDP growth in 2023 was due to fiscal stimulus in the U.S., and I think recently there was a study done that highlighted that fiscal stimulus post-Covid was higher than what we’d seen after the GFC. And this is just to put in context how of a unique playing field that we are to make our forecast at the moment. That's one of the reasons why we have changed quite a bit and we're now in the camp of higher for longer interest rates and whilst we were anticipating more cuts, I think currently we're looking at two cuts throughout this year, which is a major departure from the few we had 6 to 9 months ago. 

Belinda Roman 
I think Dennis is pointing at something. You know, the stimulus is ongoing, isn't it? Because we had the Covid, then they rolled that back. But then they started another one like the inflation or the CHIPS act or some other act where they're constantly throwing significant amounts of fiscal stimulus at different elements of the economy while the fed is trying to pull everything back. Fiscal and monetary policy are not in sync with each other. They're ones trying to offset the other. And I think that's part of the problem we're having with the economy, where we're not getting that slowdown that we want because there's always another series of we need more spending here, here and here to prop things up or to continue our political agenda. So I think that's an excellent point. 

Spencer Levy 
When you combine the IRA, the Inflation Reduction Act, with the CHIPS act and some of the other things, I think the estimate is close to $2 trillion of stimulus. Is that about right, Belinda? 

Belinda Roman
That sounds about right. And I mean, it's a significant amount of money that has to go filter through the system, and we're still seeing it. There are a number of employment, to come back, Spencer, to your point, there's a number of employment programs that are still giving 200, 300 million for the local economies to reemploy, retrain, do all sorts of things so that churn in the labor markets makes more sense when you start to put these other pieces in there that maybe you don't capture in your numerical models. 

Spencer Levy
So let's turn to real estate for a moment, Dennis. When you look at these 700 markets around the world, how similar or different when you look at a market in the United States, are they from markets overseas? How do you forecast them the same or differently? 

Dennis Schoenmacher
First of all, before we really dive into the models, you need to understand what type of data is available and what are the definitions of the data that you have available. In the U.S., I think we're quite spoiled in terms that we have very good data availability, data on the building level, that we at CBRE Econometric Advisors leverage to build our historical time series. In Europe and Asia Pacific, we have that phenomena of that we have prime rents and prime yields. And those definitions are already different in the way that it's not the average market, but the top 5% roughly of the market. Secondly, it's not built up from the building level, which again, highlights a humongous opportunity for CBRE as the largest real estate company in the world to be the first one developing that. So that's really driving the differences between how you can forecast. In Europe, our forecasts are, I think, a little bit more difficult in the way that the data is not as easy to work with. We have some long-term forecasts, but most of those time series are since the GFC, only for the real large gateway cities, we have those long time series available. One of those other interesting parts is that I think locally what we see is that supply and demand are really important in our models. When I, for example, look at the office market, I think the office market is a very important market to look at, we see in the headlines in the U.S. that it's all really bad. However, when we look at our data and we, for example, look at the top 5% as we do in Europe, you see quite similar trends across the city. But with that, I think that across the world, like, European cities are a little bit more restricted in terms of developments. The supply side is a little bit more reactive in the U.S. But all in all, I really think, given all the repricing that we've seen, that it's a very interesting vantage for investing and real estate at this point in time. 

Spencer Levy
And what are some of the key characteristics you're seeing in those 700 cities that are growing the fastest? 

Dennis Schoenmacher
I really want to touch base here on one thing, which is the impact of AI. I think, very interestingly, a lot of people lately want to talk about the long-term impact. And I think it's really going to change the way we work. But let me look at what Excel did as an example to how it changed the way we work. As we know in the past, Excel, you had many bookkeepers before that. So many bookkeepers had to change, basically the way they work. So you get more, like, financial managers and analysts. And these type of roles were created to really work with Excel. I expect a similar trend with AI. You need to be able to interact with AI to understand what you're asking the AI for you to do. And I think that will have some implications on where in the U.S. labor which is attached to AI will be. First of all, we see nearly half of AI workers are in the six largest tech markets in the U.S. But I think there is a great opportunity for other cities. And these are opportunities for cities that have very large research universities. Two that I can mention because I've been there are Columbus, Ohio and Madison, Wisconsin. And I just am personally really excited with how AI can change the way we work. 

Spencer Levy 
So, Belinda, given what Dennis just said about AI, its potential to change the way we work, the potential to change where people are going to invest and put up data centers… Does that change your point of view on how you are going to model? 

Belinda Roman 
No, and I'm going to tell you why. Because I think the one piece that might be missing from that story is the decision to place something in Wisconsin or to place it in Ohio is also a political discussion, because those were incentives that were given to put your factories and your whatever we're going to do, put it here, put it there. So those are also political decisions that are being made. So we have to take into account the political economy of these industries, which speaks to that qualitative part that I've been pointing out. And then the third thing is that now that we know energy… energy's the big problem. Energy and water. So wherever there's energy and water or the possibility of using energy in water, that's where we're going to see the growth. So if you look at the resource abundance in different parts of the U.S. or even in the world, you'll probably be able to identify where industry is going next. The problem with some of these big metropolitan areas is they're expensive, and they'll be that interest to move somewhere else, because technology can be quite nimble and flexible. It's a moving target, I think, and AI is making it all the more of a moving target. So we have to be cognizant of that and able to move with it if we can. 

Spencer Levy
Let's go back for just a moment to real estate. Dennis, where are you pointing our clients today, and in which asset types? 

Dennis Schoenmacher 
Over the last ten years we've seen such a low interest rate environment. So I personally think that the market beta is gone. So what we talk about with our clients is, it's all about location and sector and building specific. So I personally think stock picking is back in this type of environment. And then it's all about your market selection and sector selection and within the multifamily space, I personally feel that there's a lot of good solutions to find there, especially in markets where supply is relatively muted whilst demand is still pretty strong. We're also in a time where renting is much cheaper than buying a house, so real estate fundamentals in the multifamily space really still look very favorable. Within industrial, I would say that it's still like the darling of many investors. I think the most appetite for market participants at the moment to come back in the market is in the industrial space due to its fundamentals, data centers, but also self-storage and some of these other more alternatives that have become part of the investable universe are on investors radar at the moment. But again, I think location, sector and stock picking is going to be the theme of this upcoming cycle. 

Spencer Levy
Dennis, I'm going to challenge you because we're friends here, okay? And I'm just going to cut right to the chase. One of the things that I really like about commercial real estate is that our data isn't very good, as compared to other industries. As compared to stock traders, as compared to bond traders, and dare I say it, horse racing has some unbelievable data. The reason why we don't have great data is because each asset is different. So, Dennis, what would you say is the state of data in commercial real estate? And where are we going? 

Dennis Schoenmacher 
I like you putting me on the spot here, because I have quite an opinion about where we are. And you can see that more and more alternative data is being harvested across the world to make better informed decision making. And going back to what I mentioned about AI. We have machine learning, we have big data. They are all popping off to help real estate investors and occupiers to make more informed decisions. Right now, in my belief, we're at the start of the journey. However, if you look across the world, the degree of information asymmetry is so large. In the U.S., we're spoiled. In Europe, there is good data, but it's very fragmented. In Asia, it's even more difficult. However, alternative data will become more and more important. And at one point I feel real estate data will become more freely available to people to help make them better investment decisions. But that process will take another 5 to 10 years in my opinion. But I'm truly a believer that it will happen at one time so that information inefficiency is there right now. But I would say that it is changing rapidly also with AI coming in. 
Belinda Roman 
There are many, so many rich datasets out there. Some of them, unfortunately, are increasingly going behind paywalls. And I think that causes a little bit of a problem. But there are other ways to get to the data. And, I mean, as Dennis said, there's lots of alternative data sets that we can start to mine using technology. But I have come up against this now in the international trade data. Some of it's behind a paywall now. So it's been outsourced. Governments are outsourcing what is traditionally their work. That stuff is becoming commercial rather than open access. That's going to be the issue. So we're just going to have to be very creative in how we get to it. 

Spencer Levy 
Well, some of the data, first of all, this is a shout out to the FRED, which is always my first source for everything. The Saint Louis Fed. Most people don't realize this, but the Saint Louis Fed has unbelievable data that can answer many of your macro questions for free, touch of a button. But there are a lot of other things that are harder. But there are also, putting aside the paywall thing, they're just changing. 

Belinda Roman 
Exactly, exactly. I think those apps that kind of track the footfall and where you're going and point of sales and that sort of thing. I did a study for the San Antonio Stock Show and Rodeo. It is actually very fascinating. But what we did was we figured out where everybody is, and there was, like, eight counties in central Texas and 50 different hotels and all the different things that everybody was attending as well as the rodeo. And unless you venture down those roads of thinking about it being a bit more creative about what you're going to study, then you start to see the real impact of all these economic activities that are going on. 

Spencer Levy 
Dennis, have you ever been to a rodeo? 

Dennis Schoenmacher
No, I have not been to a rodeo. But let me give you one example on the data and where I feel there is a need for better solutions. What's one of the spoilers of the real estate cycle? It's always been supply, if prices are high or people start to develop. However, completions and new supply is a fundamental part of our modeling. However, the data is quite clunky. So what we've done recently, which is actually a quite exciting project led by our head of data science Franz, is we're looking at this alternative data, satellite imagery, like what is the stage of development of a particular project that we find in our database. Previously, we had to go through all of these projects manually, but we're trying to see if there are ways to automate this. And you know what's quite interesting? When we look at some of these manual projects, they're quite far off. Sometimes we see that ground is not even broken, and it's indicated that it's quite close to open. In others, we see nothing has happened yet, but the facility is already in work. So again, by having better data and using alternative sources to enrich our models is big on our agenda. 

Spencer Levy
But you know where the tools are really primitive and they need to get a whole lot better? Bull riding. Because I did go to a rodeo in Austin, Texas, and the main event is always the bull riding. And it used to be just folks wearing cowboy hats and boots. Now they're wearing armor. They're literally wearing armor on their heads, on their bodies. And notwithstanding that, four of the six participants were carried out of the arena. So that's how dangerous it is. And, well, I'm not sure our econometric tools can help with bull riding, but if they could, we'd like to help those riders. 

Belinda Roman 
Well, wait a minute. Was that a real bull or the… 

Spencer Levy
It was. I mean, it was terrifying, it was terrifying. 

Dennis Schoenmacher
I was in Dallas three weeks ago, and I met up with Richard Barkham, who’s our global chief economist. He should have gone and taken me to one of these shows. That would have been amazing. 

Spencer Levy 
Dennis, let's give you the first shot at final thoughts on where we're going the next five years in terms of data forecasting. How do we get better? And how do you see changes coming afoot? 

Dennis Schoenmacher 
Yeah, I think the last 36 months have taught us to be very humble as forecasters. And again, I couldn't have done this without the team that we have at Econometric Advisors, without all the data science, data engineers and all the forecasters that we have for all the different property types. Where I think we're going is what I mentioned in the machine learning AI topic. Alternative data is going to enrich our models. That's where I believe the industry will be going. However, we will still embrace the academic literature and the legacy of Ray and Bill in how our shop has been set up. And in terms of forecasting, we hope to cover many more of those alternatives. Belinda, you mentioned data centers. We want to have forward looking views on data centers, on self storage, on student housing, on other parts of that alternative space. And what else is really, really important for us? It's the geospatial element, because location, location, location I think is really important. As you mentioned, Spencer, buildings are so heterogeneous. So getting more granular level data, I think is going to be key for us to service our clients in the future. 

Spencer Levy  
Belinda, same question for you. First of all, do you think you're going to repeat as the number one forecaster for the Journal? And second of all, what do you see as the next 3 to 5 years of data and forecasting? 

Belinda Roman
Yeah, repeat I don't know. I have to say it's very nerve wracking now because, like, okay, now I have to be sure I'm right. Whereas before I didn't have to worry so much. And data forecasting, I’m with Dennis. Alternative data, I'm thinking more about the bigger picture. I think in general economists in particular, we kind of get very set in our ways and start looking down this one narrow road. And we probably need to be a little bit more open minded about what constitutes data. I think anecdotal information is useful and we shouldn't discount it. We’ve got to be more creative, less staid in our thinking. And that's always a challenge because sometimes it can be a very staid discipline. 

Spencer Levy 
Let me ask one more question. Since we have you, what is your forecast for interest rates and GDP for the next year and a half, Dennis? 

Dennis Schoenmacher 
Interest rates in the United States… We're still calling for two cuts this year, and GDP is gradually coming down, but we're still expecting it to hover around 2% for this year and 1.5% in 2025. 

Belinda Roman
I would agree. I think it's two interest rate cuts this year. We've got six, seven months to go. GDP doesn't seem to be coming down as quickly as I want it to come down. And part of that, I think, is because, as we said earlier, it's all that fiscal stimulus. And the question is how much more fiscal stimulus is going to roll out, especially in a political season? So I think our GDP numbers might end up being a little bit higher than we had originally thought. The Fed can't squeeze out some of that growth because it's not a Fed question. It's not monetary policy. It's fiscal policy. And so until fiscal policy gets wrangled back into its usual state of stalemate, you're still going to see a little bit more growth than we’d originally thought. So I think I have us around, I don't know, three maybe a little bit lower. The amount of Fed has that economy now. I don't know if you've seen that, economy now, and I watched that regularly. Because it's that high frequency data. But they start off very high and they keep coming down a little bit lower and lower. So I watch that quite a lot. 

Spencer Levy 
Great. Well there's a little bit of difference between Dennis and Belinda, and we'll see who is right at the end of the year. But you were both right showing up on the show today and both did a terrific job. So on behalf of The Weekly Take, I first want to thank Belinda Roman, associate professor of economics at Saint Mary's University in San Antonio. Great job, Belinda, thanks for coming out. 

Belinda Roman 
Thank you very much for inviting me. 

Spencer Levy 
And then my friend and colleague, Dennis Schoenmacher, executive Director and Principal economist, CBRE Econometric Advisors, based in London. Thanks for coming out, Dennis. 

Dennis Schoenmacher 
Good to see you as always, Spencer, and pleasure to be on this podcast with you, Belinda. 

Spencer Levy 
Thanks again to our guests and to you for tuning in. For more, please check out our website at CBRE.com/TheWeeklyTake. Next week, we head to San Francisco for a look at that market and how the nascent revolution of AI is making an impact on real estate. And looking further ahead, we'll also bring you a fascinating look at the future of cities, which, in fact, was the subject of a recent CBRE report. Meanwhile, we hope you'll subscribe, rate and review us wherever you listen, and also give us a follow on LinkedIn to stay on top of more developments around our show. Thanks again for joining us. I'm Spencer Levy. Be smart. Be safe. Be well.