Beyond the Dollar: Rethinking Money Measurement

EconBuff Podcast #49 with Fernando Colunga

https://youtu.be/DY6ryYFozZ4?si=rWcJYToNXzm8ztXP

Dr. Fernando Colunga talks with me about using Divisia monetary aggregates to measure how much money is in the economy. Dr. Colunga explains the central problems with current money measurement. We discuss simple sum measures and Dr. Colunga argues that simple sums can be improved upon by weighting the different types of money. We explore liquidity, with Dr. Colunga explaining how Divisia monetary measures apply weights to various categories of money based on their liquidity. Dr. Colunga lays the framework out for how Divisia measures are constructed and describes when simple sum measurements will differ most from Divisia measurements. Finally, we explore how new money measures might impact monetary policy if employed, and how economists are constructing unique Divisia measures for each country.

SE5: E1

FOLLOW ME HERE:

Spotify: https://podcasters.spotify.com/pod/show/econ-buff-podcast 

Amazon Music: https://music.amazon.com/podcasts/0434c041-ec75-443e-8002-68cfcdd66db6/econ-buff-podcast 

Facebook: https://www.facebook.com/econ.buff.1 

Substack: https://econbuffpodcast.substack.com/ 

My website: https://econbuffpodcast.wixsite.com/mysite

Thumbnail photo credit: Bing AI Image Generation

Transcript:

LEE STITZEL: Hello and welcome to the EconBuff Podcast. I'm your host, Lee Stitzel. With me today is Dr. Luis Fernando Colunga. Dr. Colunga is an economist at the Bank of Mexico. He received his Ph.D. in economics from the University of Texas at Dallas. He joined the Bank of Mexico in 2021. In his current capacity, his research focuses on regional economics, macroeconomics, and monetary policy. His most recent paper on Divisia monetary aggregates has been accepted for publication at the highly prestigious Journal of Money, Credit and Banking. He also serves as a professor for UANL in Monterrey, Mexico. Fernando, welcome.

FERNANDO COLUNGA: Oh, thank you so much for having me here. Thank you so much for the invitation to participate on this podcast. This is my first time being on a podcast like this one, so thank you so much for the invitation to participate in this very friendly conversation. Thank you so much.

LEE STITZEL: Well, thank you for doing us the honor of joining us. Our topic today is going to be Divisia monetary aggregates we're obviously going to kind of have to build up and explain what that is but first what I'd like you to do is kind of explain to the audience what the Bank of Mexico is and kind of what it does right?

FERNANDO COLUNGA: Yes. So the Bank of Mexico is the Mexican central bank, like the Federal Reserve in the United States. Here it is Banco de México, or Bank of Mexico. Some people confuse the Mexican central bank with a commercial bank that has a similar name, Banco Nacional de México, but the central bank is Banco de México. The first and most important function of the central bank is to provide money for the people so people can make transactions. That is the first and most important function of the Mexican central bank At the same time, the central bank has the objective of preserving the purchasing power of that currency, of bills and coins. Right. So in order to do that the central bank will always want to keep inflation low and stable Unlike the U.S. case in which the Federal Reserve has two objectives two targets inflation but also employment Here in Mexico we have only one, and that target is inflation. Right. That is basically what we do in the Mexican central bank we care about inflation, and we do all the analysis required to understand what is going on and what we can do to deal with inflation.

LEE STITZEL: Thank you. That's a very useful summary. So the listeners can then think of the Bank of Mexico as analogous to our Federal Reserve in the United States. And then of course there are nuances and different things there. Right. So are what I'd like to focus on is what Divisia monetary aggregates are that's going to be something that actually longtime listeners of the podcast will have heard Dr Matson talk about a little bit but obviously we've never sort of delved into that and that's the goal for the topic today so can you tell us what Divisia and monetary aggregates are and why they might be important?

FERNANDO COLUNGA: Okay. Yes, Lee, of course. I'm very happy to talk about Divisia because as you mentioned my Ph.D. adviser and I just had a paper on Divisia for Mexico accepted, by the way, for Mexico, by the way, but What is Divisia? That is a very good question because not all of us know what Divisia is Even in my case, I didn't know what Divisia was before going to UT Dallas the University of Texas at Dallas where I did my Ph.D. Right. I had the same question what Divisia was. Well, Divisia The literature regarding Divisia goes back to the 1970s and 1980s, with Professor William Barnett right who proposed that monetary assets in the economy should be aggregated by considering that each monetary asset is different Let me explain why. to use a very simple example. If you have apples and bananas you will want to aggregate apples with apples. and bananas with bananas the same analogy applies for the case of monetary assets that they are different let me explain with a very simple example with two monetary assets let's suppose that you have money in cash and you have money in a savings account you can use the money that you have in cash to make a transaction immediately. Right. You can go to the store to the grocery and make a transaction with cash but you may not use the money that you have in a saving account because there may be a term or restriction, so you have to wait to use the money in the savings account so they are different. Right. And they are different in terms of the degree of liquidity money in cash is very liquid but money in a savings account is not as liquid so since they are different we cannot just do a simple sum which is what central banks do just a simple sum money in cash plus money in savings accounts what the Divisia index does is to consider that each monetary asset has a different weight and weights those assets when we aggregate them by considering this different degree of liquidity it's like, for example, if you have a portfolio. Right. You want to invest your money. So you're going to assign a different weight for the shares of Pfizer and a different weight for the shares of Walmart why because they are different they are different in terms of risk? And return. Right. So because they are different you're we're going to assign a different weight if they were equal the weight would be the same but in the case of money like for some other aggregations since they are different we have to consider those differences and that is basically the idea behind Divisia monetary aggregates for monetary assets we're going to aggregate these monetary assets but we are going to weight each monetary asset. So that is basically the idea and as you can see it makes sense. Right. The idea is very straightforward the idea is very straightforward the problem is that for the case of money it has not been applied to money in the same way that weighting is applied to other financial assets, prices of goods and services, or production right for the case of money. So that is where we come in. Even for the Mexican case right for the Mexico's case we did not have that index and we are proposing that index for the Mexican economy. So that is basically the idea I don't know if I was clear or maybe I caused some confusion. Right. But hopefully I may have been clear.

LEE STITZEL: Right. No, you did great. let me ask a couple of things just to make sure. So when you're saying monetary assets the listener should be thinking about this like different money that are in different places what things would be included in those different categories and what things might they be tempted to think of as money that are not exactly money?

FERNANDO COLUNGA: Okay. So for the U.S. case, for example, there are levels of monetary aggregation, and that is pretty much the same for the Mexican economy in M1. In M1 we're going to have the most liquid asset, for example, cash and money in a checking account it's money that you can use immediately and is very liquid. we also have another aggregation like for the U.S. case M2 and in M2 we're going to have the assets in M1 plus some other assets that they are relatively less liquid like, for example, the money that you may have in a saving account the money that you may have in other instruments or monetary assets like, for example, repos, which are repurchase agreements or money market fund shares. Right. So in other assets the same for Mexico you may also have money which is less liquid in public securities. So in Mexico we also consider money the money that is invested in public securities and in order to do that we broaden M2 to get M3. So we have M2 plus public securities and we have M3. More than that, in Mexico we also consider that some of these assets they may be held by residents but also by nonresidents let's say that Lee he has some pesos he's a nonresident right he's a non-Mexican resident. So we also consider the money that other people, nonresidents from other countries they may have right they may have. So we have four levels of aggregation M1 and M2 which they are pretty similar to the case of the United States but we also have M3 in which we include public securities and we also have M4 in which we also consider that non-residents may have some of these assets. So those are basically the assets that we consider the idea is to consider assets that they are relatively liquid. We don't include shares or we do not include some other assets we include assets that they are relatively liquid right.

LEE STITZEL: So what the traditional method is doing is trying to categorize different monetary asset types by their liquidity. And what you're saying is instead of doing that let's provide a weight. So when you say weight what does that mean? My intuition as an economist is that you're talking about a price or a cost of some type. Can you explain that to us?

FERNANDO COLUNGA: Exactly. Let's say, for example, that you have two assets right money in cash and money in a savings account. Maybe you have 70% of your money in cash and 30% of your money in a savings account We only have two monetary assets, and we want to aggregate the money that you have but if we do the simple sum the problem is that we are not considering that this money in two different assets each dollar has to receive a different weight because they are not equally liquid. So basically what we do is to weight by constructing an expenditure share it's just an expenditure share like people who take microeconomics classes right they learn that they are going to divide right they are going to allocate a budget right according to different goods and services. So basically here we do the same we are going to assign an expenditure to each monetary asset and in that expenditure share we are going to have how much of all of our money? We are going to assign for each monetary asset times the price One Price Right which in Divisia we call the user cost or the opportunity cost which is which is like a spread right it is a spread of interest rates that is going to be the price and if we sum prices and quantities for all these monetary assets we're going to have the total expenditure. So the share is going to be it is going to be the fraction. Right. That we assign or we allocate for each monetary assets with respect with respect to total expenditure. So that is basically how we're going to wait? So and you may say. So how are? You weighting how by constructing this expenditure shares are? You considering you are considering the difference of degree of liquidity let me tell you how that spread,? The price for each monetary asset it is going to be the difference between the interest rates that you may get from holding money in that monetary assets versus the interest rate that you may get from holding the money in another monetary assets in simpler wordss that is going to be your opportunity cost let's say and let's put an example let's say that you have money in cash if you have money in cash what is the interest rate that? You may get from holding cash with you zero the nominal interest rate is zero. Right. But at the same time you may have the opportunity to invest your money in a savings account which is going to give you probably 3%. So the difference between the interest rate that you may get from the saving account and the interest rate from holding cash with you that difference which in this case is three minus zero it is going to be three that is the price that is going to be the price for holding cash with you your opportunity cost in this case 3%. So as you can see the higher the price, or spread, the more liquid the asset. In this case, cash has the higher spread. So we obtain that by using the amount of money that you're going to allocate on that monetary assets times the price over the total expenditure you are going to wait through the price to reflect how liquid one asset is relative to another that is pretty much it.

LEE STITZEL: Okay. So what I like about this right is it's a very powerful application of one of the core intuitions in economics. Right. Exactly which is that we should account for the price signal in this. Exactly. So let me ask you this because this is just kind of a question of interest to me at you know and you would know this I think understanding the history here. So we would never just add apples and oranges we would never just say, “I have two apples, I have three oranges, so I have five.” everybody sees that's immediately silly. Right. So what happened in the history of using you know M1 M2 M3 Aggregates were economists just fooled because it seems like everything's in dollars. So we can just add them together and that was just a thing that we didn't think about or what happened that made this you know a contribution that you and your colleagues can make to economics?

FERNANDO COLUNGA: Right. And since as an economist we know that we can do simple sums when we have perfect substitutes. Right. So from a theoretical point of view, aggregating things that are not equal or perfect substitutes is not realistic. So if we do that if we do that we have an measurement error. Right. We have an error when we aggregate why because theoretically from a theoretical point of view? We are aggregating something that they are in essence different. So if we consider if we consider that this monetary assets they are different we believe that we are correcting we are fixing the problem and we are proposing a measure for money a measure for money that is more realistic. Okay. And we may say we're proposing an index that does not suffer from that measurement error. So that is why we believe that? We're making a contribution again for prices it has not the same weight the price of gas than the price for shoes you spend a lot of money in gas. Right. So when you aggregate the prices to get the CPI the Consumer Price Index you are going to assign a different weight based on what on how much of your expenditure or your income is spent on that on that? And on that good right on gas and shoes right or burgers or drink or whatever. So the same the same ideas apply for Divisia but in this case for monetary assets. So we are proposing another measure, an alternative that we believe that doesn't suffer from that problem from that from that assumption that we have made which is not realistic that is pretty much your contribution. Right. That is basically the idea.

LEE STITZEL: So I think it's straightforward to see now that you are making a contribution. Right. The natural question then is why is that important or if? I can put that another way what is wrong with using simple-sum aggregation?

FERNANDO COLUNGA: Exactly. That is a very that's a very good question and in order to answer that question I may focus on the advantages of having a Divisia index and the disadvantages of using the simple sum right now in our paper what we do is look at the past let me tell you a little bit about what has happened in the past in the past some people may say why don't we use money to analyze what happens when money supply increases or decreases with prices? And production what if we increase the money supply and we want to see what happens with prices we may know from other countries and for Mexico as well what happens if we increase the money supply what happens we may think well now people are going to have more money in their pockets they are going to start to consume more and as long as we don't produce more we're going to cause inflation that is one prediction right. However, theory suggests that if you use money as a simple sum you may not have the correct response on inflation why is that because? You're considering money from a from a perspective in which there is a problem there is an measurement error. So in the past Er When people they want to analyze what is the what are? The implications of increasing the money supply they may have found some anomalies some puzzles why is that if? We increase the money supply the inflation rate goes down it should go up no that is an anomaly that is a puzzle that is something that we not that we may not expect. So what we argue is because there is a problem with the with W with the measure for the money supply what if we measure the money supply as we should but by considering that the difference monetary assets that they conform the money the money supply doesn't suffer that index that measure doesn't suffer from that measurement error and what we find is that if we want to analyze what happens when there is an expansion in the money supply we don't find those anomalies that you may have if you use the wrong measure the simple sum that is one of our contributions what if we if we fix the problem on that index on that measure of the money supply the simple sum what if we do that and what we find is that when we do that we find more let's say the responses of some other variables when you have an expansion or a or a or a tightening a money tightening the responses on our analysis of inflation and production they are more consistent with the with theory with the textbook with the textbook predictions. Right. So that is one of the things that we may consider as an advantage of using a the Divisia index to analyze what happens following a monetary expansion on the response of some other variables that is why we believe that in some way? We're contributing to the analysis that.

LEE STITZEL: So one of the things that economists are very sure about is that inflation is a monetary phenomenon. And. So we all agree on that if you if you find an economist that disagrees with that you know there that's a non-orthodox opinion exactly obviously those views have their place in academics but you know The general consensus is that you're going to have inflation come from you know what's going on with money supply. So what you're saying is if we use simple sums we have what you call anomaly so there are problems with the way that we track inflation and money supply against each other and what you're saying is when you use a better measure which we've already we've already established. Right. This fits that basic economic intuition of using prices is when you fix that thing that we are very sure about the relationship between the money supply and inflation that relationship becomes very clear and a lot of those anomalies go?

FERNANDO COLUNGA: Exactly. That is. Exactly. The idea.

LEE STITZEL: Very good. So let me ask when you are looking at the differences between Divisia and the simple sum is it that Divisia you know in simple sum like they are correlated. Right. So if the simple-sum money supply looks like it's going down Divisia is also going down and it's just a matter of like magnitude or is it that Divisia and the simple sum can diverge quite a lot such that maybe the simple sum says money supply is going up but Divisia says money supply is going down can you give us like a sense of the magn of the problem?

FERNANDO COLUNGA: Exactly. That is a very good question, Lee I really like that question because it makes sense it makes sense a lot L question let me give you an example what if you use the simple some for money for M1 in which you have cash and money in checking accounts how there these assets they are pretty similar money in cash? And money in a checking account they pretty similar you can use your debit card and you can use cash from your pocket. So for that case let's suppose that we produce a Divisia index for these monetary assets what would we expect to find pretty much the same because they are the same they are the same. So if you consider an index for as a simple sum if you consider the simple sum and you consider the Divisia index for monetary assets that they are the same you're going to find pretty much the same so I would say that the correlation for the simple sum and the Divisia index which considers assets that are equal they are pretty similar they are going to be pretty much the same they are going to be highly correlated for sure why that because they are equal in that case it doesn't make sense our? The index the problem starts when you start to aggregate somewh assets that they are different like, for example, money in a savings account. So in that case in that case we may start and we see we see that the Divisia index behaves different to the simple sum and you may say why you may? I mean probably most of the time there is a correlation positive correlation but in sometimes you may find that there are differences there are gaps between one index and another one. So we argue that all of these differences they are going to be given or they're going to be driven by the interest rates the new information that we are adding which in this case relates to the to the prices or to the user costs for each monetary asset I would say is so to Su up I would say that if for monetary assets are pretty similar theis the simple sums you're going to give you pretty much the same but as long as we start to aggregate different monetary assets we may start to see some differences those differences they going to be driven by the new information that we are adding to the simple sum which in this case is the information in related to the interest rates so I would say that for the Mexico case we have observed that when a monetary expansion by using the simple sum that expansion was higher when seeing the Divisia index. So the expansion it can be either higher if you consider one on another one usually what we have seen is that it has been higher when we use the dbci index.

LEE STITZEL: So now I'm a central bank and I understand the importance to Divisia I start to implement that as my you know preferred measure, if you will, how is that affecting my processes? And the way that I conduct monetary policy. So you know hopefully listeners are going to understand kind of how we use targets in monetary policy? So go through that how would? We use Divisia versus simple sum versus like targeting an interest rate or something like that?

FERNANDO COLUNGA: Right. And you know right now many people they don't think or they don't follow or they don't trust a lot of the a lot of on the impact of money on the economy why because they say well? The central banks they use the interest rate to achieve their objective they use the interest R. So why should? We care about money why is that some of? The people that they believe that or that they have this that question is because in the past they have observed that when you they use the money as a simple some they find when they are doing their analysis that they find this anomaly. So why should? We care about money if we have observed that the responses of the variables that we want to analyze they are not the ones that we may that we may expect what they may not realize is that we can produce we have produced a measure of the money supply that does not suffer from that measurement error. Empirically, what we have found is that the responses of some variables like inflation which is the most important for our central bank but also in production and employment they make more sense when we use the Divisia index. So what I may say if in the past some people they stopped believing about the effects of money and that is because we may argue that was a problem with our measure of the money supply now that we are solving that problem we may continue, or start again, to believe that money can have some effect on economic variables it has some advantages, for example, with respect on interest rates the Divisia index consider considers not only money balances but also information about several interest rates and the interest rates that we use they are highly related to the interest rate which is the instrument for the central bank. So we consider our Divisia index like a an alternative indicator an alternative indicator that it encapsulates is it is like a like an amalgam that considers more information that it's it's a wider or broader index that considers both interest rates but also money balances in the right way. So we believe that we are not saying oh central banks they should care right now about the Divisia index we are not saying that we're saying well what if we put on the table another alternative what if we do that and what if we do the analysis by considering this index that is the idea and that is why we live that for central bank having? The Divisia index indices may help to grow the understanding to have a better understanding that is that is the idea.

LEE STITZEL: So that's a good foundation for the basics what Divisia is as we've talked about here you've got work that's specific to this tell us a little bit about what your work on Divisia Aggregates what it's doing and why that might be interesting in your position?

FERNANDO COLUNGA: Okay. You know, right now in the Mexican central bank I am working on a regional office in which we do more about regional analysis I am in Nuevo León, on the border with the U.S., one of the states on the border with the U.S. Pretty close to Texas, by the way, so most of my analysis it has to do with regional stuff regional Chucks but I am still doing my I am still doing my research on Divisia I am still working on Divisia because we believe that in the future we can do a lot of stuff with this index, for example, and in the literature there are some works that they have been done for some other countries that they suggest that using this index helps forecast money demand better. So that is one alternative if the central bank wants to know what is going to be? The money demand that index helps better to forecast the money demand and in on your hands we may also consider what are the effects of having an expansion on the money supply by considering this index but take into account that Mexico is a small open economy and we depend a lot of the on the U.S. We depend a lot of the U.S. 80% of our exports go to the U.S. And or exports are 40% of our GDP. So if we use the Divisia index by considering a model in which we are some Mexico's small open economy We Believe that we may amplify the work that we have done in the past which is has been pretty much producing that index for the Mexican economy. So we believe that we can continue extending. More than that, there are some more some Er works on the Divisia index that they have consider some other monetary assets like, for example, money coming from other sources like, for example, credit cards. So we believe that we can still do a lot of on this I have seen I think that some other Diva indexes indices that they consider something related with the money used for the oh for the climate change climate change some bonds so green bonds or something like that so I think that we can continue doing a lot of stuff on the Divisia indices and we can produce better and wider DV indices for Mexico and for some other countries, by the way, so.

LEE STITZEL: So let me see if I understand this. Right. Because you've made this illusion a couple times you are going to have different indices for different countries. Right. So you're talking about an index specific to Mexico is that because right there's a central bank there and so there's their own money supply. So you need the calculation for that individually or are there also different weights because there are different interest rates in those economies?

FERNANDO COLUNGA: Okay. So that is a very good question when we produce the Divisia index the I would say that the most difficult thing the yes the most challenging part is data regarding to money balances is very straight forward because we have that data in any central bank has that data. Right. So we don't have any problem with that you can download that data and that's it the problem with the Divisia indices is to find an interest rate an implicit interest rat that matches that the match the match for the interest rate that matches with the monetary asset for some assets is very straightforward like, for example, for cash what is the nominal interest rate for cash? Zero. Pretty easy? But for some other assets for some other assets it may be more complicated. So that is the challenging part to find an interest rate that is let's say, is the appropriate the appropriate interest rates for each monetary asset so some people may aruse so I think that this should be the interest rate for this monetary asset let's say let's say for saving accounts and some more people may say no I may live that we should use this another interest rate because you know there are many saving accounts in many banks and so it's very complicated so at the end motional part for constructing Anda index is to find the interest rates related to each monetary asset that we are going to add in our case we believe that after making after doing a very let's say a hard work after having worked very hard investigating doing the research we have found the best interest rates we have the best available match for each of our monetary assets. However, we acknowledge the possibility that some other people they may argue well so maybe for this for this monetary asset we should consider another interest rate so yeah we an for that possibility.

LEE STITZEL: So the second thing that you said there is you also want to get more categories of monetary assets into the index do I that right yes so I think one of the things that can sometimes be confusing to non-economists when we start talking about monetary policy is what is money? And of course I Dr and I Ryan and I are very sort of well aligned on this topic you know. So when I'm teaching money you know I'm teaching that money is a service and. So that comes in a lot of different forms but I think you know most people they hear money they're thinking of green pieces of paper and they think cash and so they're not like you mentioned credit cards a minute ago. So what you're saying is you're trying to add more and more categories of things that serve as money within the economy it is that is that a good characterization of what it is that you're trying to do with your index?

FERNANDO COLUNGA: That is where we would love to do that to consider more sources for liquidity on the monetary side, for example, that money that we may use from a credit card the problem is again data I don't think that we are going to have the data for constructing the a historical index an index coming from the 2001 or 2002. So we may want to do that. However, we will see if the data constraints they allow they allow us to do that but we have seen that for some other countries for the U.S., for example, you have already worked on that. So we may consider that as a possibility for the Mexican economy but it's just like an idea no so then you could make it more accurate and therefore you know sort of exactly go back to that issue we're talking about with anomalies that we would have an even more accurate measure. Exactly. That is the idea even the idea comes from what you were saying Le most of the people we believe that money is just cash we have that idea but money is more than cash Money a saving account it's Money public securities it's more than that. So if we constrain or analysis to a narrow measure of the money even when money is. More than that, I think that we may we may omit or we may have yes we are omitting another fraction of the money that is circulating in the economy and results they may be also a subject to the fact that we are limiting the definition of money. So if money is more than cash I think that we may allow for having more an index that consider more assets that only cash even some people I have heard some people that they say if the problem is that you are aggregating different monetary assets why you don't use only M1 cash? And money checking accounts in that case Divisia and the simple sum they are going to give you the same. Yes, that is. Right. But we have to remember that money is more than that. So when we only consider that money is only the money that we have in the Pockets or in the wallets we are not considering that the money circulating in the economy can be more than that. So the effects of money can be broader they can be they can go beyond that we may expect if we only consider cash. Exactly. That is that is the idea.

LEE STITZEL: What I want to turn to a little bit as you mentioned. Right. We talked about there's different indices for different countries and then you talked about the work of William Barnett and Center for financial Stability and they've got these us Aggregates and then now you're working on Aggregates for Mexico are they are differents between those two Aggregates and if so could you discuss them a little bit?

FERNANDO COLUNGA: Okay. So in strictly speaking regarding to the methodology I think that they shouldn't there shouldn't be differences or big differences because we follow the International standards to produce the diva in so regarding to the methodology I don't think that we should have those differences. However, as I mentioned we may have some differences regarding to the official classification of money as I mentioned before in Mexico we have four levels of aggregation for money and to my understanding for the U.S. case we have they have two right M1 and M2 and there can be some orders for the official M1 and M2 so I may say that if there are some differences they are related to the aggregation for M1 and for M2 we shouldn't have differences but for higher levels of aggregation we may start to see some differences essentially the idea of Divisia here and the UK and the U.S. They are the same I mean the methodology we follow the same methodology. So we shouldn't have method methodologically speaking or empirically speaking those differences.

LEE STITZEL: So let's talk about the specific to the Mexican economy as you mentioned a minute ago then. So if I use these Aggregates what's the difference in the story that I see being told about like inflation or output that if I'm using the interest rates or the simple sum that I don't pick up there can you describe that for us?

FERNANDO COLUNGA: Yes for sure so first of all when we want to describe or compare the effects of the simple sum and the tiia index we may say well if we're saying that the simple sum is problematic we shouldn't compare something that is problematic with something that is not problematic. So we don't like to make that comparison if something is wrong why we're going to compare? The results or effects that things that is wrong with something which is. Right. We don't like that comparison but what we like is to make the compar what we do like is to make the comparison between this index the Divisia index for the money supply along with the usual monetary policy indicator which is the shortterm interest rate. So what we find or the story behind the effects of this index and the interest rate tell us that when we use the shortterm interest rate which it can be an indicator of the monetary policy for each for any country we may find some let's say that we have an increase we have an increase on the interest rate if we have an increase on the interest rate we may expect that production goes down and inflation goes down SL that is something that we may expect why is that because if? We increase interest rates if we have a monetary tightening we may reduce your consumption right now it is higher it is more costly to consume especially durable goods like, for example, a car or to invest in a new house that is going to be pretty much more costly. So the idea is that if we increase the interest rates we redu we reduce consumption we reduce investment we reduce aggregate demand and as result we may observe some pressure for inflation to go down that is what we may expect that is textbook. However, empirically we may found that following the monetary tightening which means an increase on the interest rates the interest rates the price level or inflation rate can go up why is that if? We are increasing the interest rate why the inflation rate it is going to go up that is a puzzle? We call that the inflation puzzle in macroeconomics so where we say well what if we use an index that it incorporates not only information about money balances but also about interest rates and. More than that, not only one interest rate but more interest rates as a monetary policy indicator. So in this case following a monetary tightening we should have a reduction on the on the money supply in this case a reduction on the devisa index and what we find is that following a reduction of the money supply or a through this index what we find is that the inflation rate following a decrease in the money supply is a a reduction in the inflation rate and a reduction in the production level or, in other words, if we have an increase in the money supply we should have more inflation that is what we find. So one thing is textbook but empirically when we want to analyze what happens following a monetary shock we find that if we use the usual monetary policy indicator which in this case it can be the interest rates that can be a little problematic in some cases. So when we use the Divisia index their response following a monetary tightening or a monetary expansion they are more consistent with textbook with what the textbook May predict MH. So that is pretty much the study behind what are the effects of this index are a monetary expansion measured with this index on output and inflation, for example.

LEE STITZEL: So let's say I'm pretty convinced but let's say all the listeners they're convinced they say Okay throw away simple sum measures we got this Divisia measure they still might be confused as to why the story? You just told why does it work that way if? I have a better more accurate measure of the money supply. Right. We're we're we're targeting interest rates. Right. What why do those two things not then align in? The way that you're saying the money supply and inflation eventually align?

FERNANDO COLUNGA: Okay. Yes. So that is a very that is a very good question and you know when we analyze the time series for interest rates and the Divisia index we observe a negative relationship right when the interest rate goes up, money balances go down that is that is. So we may expect that if we are having a contraction on the money supply which it can be through higher interest rates or a decrease in the money balances we should have the same results. However, empirically we find that following a restriction, or monetary tightening if we use the interest rate or if we use Divisia the responses are different you may ask why is that why if they are supposed to give? You the same results one gives you one response and the other gives you another why one some responses are consistent with? The textbook and why some are anomalies. So that is a puzzle that is still a puzzle for us but we believe or we may argue that since Divisia is wide is a wider indicator it's more than one interest rate you know what the central bank has an instrument which it can be the interest rates but the Federal Reserve, or any central bank in another country they also make open market operations they sell and buy securities. Okay. So that is another way in which the central banks they also have an influence on the money supply they buy bonds, and if they buy bonds they are going to give you money in exchange for that bond they're going to give you some money. So we believe that by using the Divisia index we are also incorporating that information we are incorporating and. More than that, those operations are going to have an influence on interest rates and we're capturing that information not only of one interest rate but some interest rates and money balances in other monetary assets that one simple interest rate may not have been capturing. More than that, for the U.S. case, for example, what happened after the Great financial Crisis interest rates they remained near zero so does that mean the central bank is not doing anything else they are not trying to influence or to stimulate the economy in another way the interest rate they already zero so they are not doing anything yeah The answer may be yes: they are doing something else they are trying to stimulate the economy. So how is that trying to put more money on? The economy how by buying some assets some monetary assets that? The people we have in that way the Divisia index is going to capture the money balances the increase on the money supply but also the effects that those monetary assets you're going to have on the interest rate. So that is one way why we consider that? The Divisia index is like having the big picture is considered more than one interest rate it considers more of the operations that the central banks do or that they have and the effects that those operations can have not only one interest rate but multiple interest rates. So that is why we believe that? The results they may they may be different because in some way the Divisia index is considering more information it's giving us more information about the monetary policy stance of any central bank.

LEE STITZEL: Yeah, that's a really good answer but I want to you know for clarity here make sure that I that we've got these two things. Right. So part of what you're saying is a simple sum is not accounting for any interest rates using short-term interest rates means we're accounting for a few interest rates and when we use Divisia we've got a whole index with information from multiple interest rates so it's it's much more robust amount of information. And so then the other thing you said there is a and I think this is very illuminates it very well at the zero lower bound. Right. We start to lose information in what's going on with interest rates and if you only use that as a measure then you've lost what's actually happening in monetary policy. Exactly. So is it the case. Right. Because it should be immediately obvious to the listener. Right. The interest rate gets down to the zero lower bound you know that tool becomes ineffective I mean that is one of the central lessons of 2008 and of course if we'd have been watching we would have seen that elsewhere in the world like Japan earlier but we didn't we had to learn our lesson firsthand. Right. So but is that particular lesson only true in a crisis? It strikes me that it is not only true in a crisis that's not only true at the zero lower bound it's just that it's very obvious to see when we get there.

FERNANDO COLUNGA: Exactly. Because central banks are always operating not only during crisis but during a crisis it is more evident that they want to stimulate the economy like, for example, in the U.S. case in a more aggressive way well if the interest rates are already zero what else we can do what else can we do so one alternative is trying to stimulate the economy trying to put more money into the economy by doing these open market operations but during a crisis it is certainly more evident that this happens yes. However, that happens all the time that is the idea.

LEE STITZEL: So we've been going for about an hour now so I want to kind of bring this in for a landing but I'm going to try to go two for one here so I'm going to ask you kind of a two-part question to let you kind of finish this episode off so first is what we just said was we look back and we see historically and then we use better data and then we better explain the path of inflation in particular. Right. But of course you're talking about all kinds of monetary measures. So if we turn that around and then we say now that we have a better measure for all the reasons that you've discussed better economic intuition better fit with models and theory and better empirical track record all those things that we would like how does that then help us when we want to hit a policy target and not just explain what has happened and then the second part of that question would be what then as maybe your research or other people that are working on this what's the direction that research is going to go to sort of help monetary authorities with that policy targeting?

FERNANDO COLUNGA: Okay. So those are very hard questions yes I would say that it's complicated I mean what we are doing is trying to it's it's hard to convince or to persuade some people that we are in the right way because some people from some schools they reject the idea that money has an effect on the economy they don't believe it at least in the short run and what we believe that at least in the short run money can have an impact so some people they refuse to talk about money because they say no we should concentrate on the interest rates that is that is what we use and what we should focus on so it's not easy it is not easy to convince people who believe only in interest rates that they believe only in interest rates as an indicator of the monetary policy to convince them that we can also use an alternative like this so I would I would say that for the future as long as we promote or we inform people about all of the advantages of having this index we may we as long as we try to convince or to persuade people that we are in the right way we may have more acceptance from the broader audience not only from academics but also from commercial banks, by the way, from central banks and as long as we, for example, this podcast can help right to create to create more knowledge increase knowledge about these topics, which are unfamiliar to some people. Even for me, they were unfamiliar but I believe that as long as we have these channels to transmit how important it is to correct the measures that we have for different variables as long as we do that we may try to create more acceptance among the audience. Right. And try to implement this in the analysis maybe not on the decisions maybe they will not say okay let's see what's going on right now with the Divisia index to see if we do something or not but at least to consider that on the analysis as I mentioned it is very important to see what is going to happen if we have a monetary tightening for economic variables like production we may cause a recession we do not know what is going to happen. So if decisions are going to be made based on information based on data we believe that the data or the analysis has to be made based on the correct measurements for all the variables. So that is one way in which we believe we can contribute we can contribute and, for example, through the publication of this paper in the JMCB we believe that more people may start to ask themselves. So what is that? For example, for the Mexican economy or for some other economies. So what is Divisia? And why should we care about Divisia? So even when the literature on Divisia monetary assets goes back more than 40 years, to the 1970s and 1980s I think that we can still do much more to continue trying to push this discussion on the on all the audience students, academics, central banks, commercial banks, and everyone that we care about what is going to happen with the economy.

LEE STITZEL: My guest today has been Dr. Fernando Colunga. Fernando, thanks for joining us on the EconBuff.

FERNANDO COLUNGA: Thank you so much, Lee. Thank you so much for having me here on this podcast. I really appreciate this invitation.

LEE STITZEL: Thank you for listening to this episode of the EconBuff. You can find all previous episodes on YouTube at EconBuff Podcast. You can check out our website at econbuffpodcast.wixsite.com. You can contact us at econbuffpodcast@yahoo.com.