 Ahora les quiero contar un poco sobre la trayectoria del profesor Silvestre Eifinger, profesor de economía financiera en la Universidad de Tilburg, Holanda y profesor visitante de economía en la Universidad de Harvard. Durante su trayectoria investigativa ha enfocado sus análisis en la política fiscal y monetaria y a la integración económica y financiera en Europa, siendo profesor visitante en los bancos centrales de Alemania, Japón, Francia e Inglaterra. Termina siendo asesor especial del Fondo Monetario Internacional y de la Comisión Europea. Sus publicaciones se encuentran publicadas en las principales revistas científicas como el Journal of Money, Credit and Banking, el Journal of Banking and Finance, el Journal of Public Economics, Oxford Economic Papers, el Open Economics Review y el European Journal of Political Economy. Bienvenidos, profesor Silvestre. Gracias por tu buena introducción. Gracias por estar aquí. Quiero agradecer al doctor Santiago Castro por su invitación. Hace un año aquí en la Convención, en junio, el S.C. Es un placer volver, no solo a Colombia, sino también, especialmente a Catarina. Estamos muy, mi hija y yo, muy contentos de estar aquí. Gracias mucho. La investigación que haré, o que presentaré, es que veis mi nombre solo ahí, y no es apropiado en el sentido de que el papel fue hecho juntos con Jonathan Malagon. Y la historia de ese papel es la siguiente. En la Convención de la Convención de la Comisión de la Colombia, el año pasado, presenté un anuncio de riesgo y laquisidad de los mercados globales. Y recibió muchas preguntas de las personas en la audiencia. ¿Cuáles serán los puestos, o que serán los puestos para países latinoamericanos? Así que tenía una gran discusión también con Jonathan. Como dije, Jonathan, podemos hacer un buen papel sobre la política de la Convención de la Convención de la Convención de la Convención, por el sistema de reservas federales, por el ECB, por el Bank de Japón, y vamos a probar eso. Vamos a probar los puestos de los grandes bancos centrales, el sistema de reservas federales, el ECB y el Bank de Japón. Y ¿qué significa para los países latinoamericanos? Creo que estas son preguntas muy importantes. Y la buena cosa es, Jonathan es muy entusiasmante, es que fue una buena cooperación, y Jonathan ha hecho muchas de las estimaciones, el profesor no tiene mucho tiempo para eso. Y lo sé, y estoy muy agradecido a Jonathan que podamos trabajar juntos en este buen papel. Ok, entonces, aquí vamos. Y tengo un montón de slides, así que tengo que ser muy, tengo que acelerar ahora. Aquí tenemos, ok, si no, voy a tomar este mic. Aquí tenemos la introducción. Voy a seguir aquí, eso es más fácil para vosotros. La introducción es muy breve aquí, vamos a la crisis, 2008-2009. Es financiero, originó en las economías desarrolladas, especialmente en los Estados Unidos, y, por supuesto, esta fue la crisis, que fue, yo diría, lo más fuerte desde la Great Depression. Si veamos a la posibilidad de monitorar la política aquí, veas los grandes bancos centrales, así que la eurozone, Estados Unidos y también Japón, y veas también que este intercambio fue al domenio negativo. Y lo que podemos ver ahí es que las opciones para los bancos centrales fueron limitadas. Y que tenían que ir de la política de monitoría convencional a la política de monitoría inconvención y no convencional, porque de la cero cero. Ok, entonces, ¿qué pueden hacer? Y aquí veas también otro problema. Y, por supuesto, la política de monitoría no era posible, porque el hecho de que había un defensor fiscal en Estados Unidos de más de 12%, un enorme ratio de ADP-GDP en Japón y más de 200% aquí, y, por supuesto, el defensor fiscal en la eurozone, que tenía que hacer, por supuesto, con el hecho de que la eurozone también fue confrontada con los constracciones de la estabilidad en el riesgo, que no siempre ha sido tomado en serio, yo debo decir, pero, entonces, sólo había una opción que el Bank de Centro se desplazara. Y lo que sucedió, el Bank de Centro se desplazó por la política de monitoría en Japón y se desplazó, y lo que sucedió, el Bank de Centro se desplazó por la política de monitoría. Aquí veas, las economías desarrolladas, como las de los Estados Unidos, la eurozone, la economía y Japón, han ido a una política de monitoría inconvencional, una política de monitoría inconvencional que no es solo inusual, pero también, yo diría, más allá de las mandaciones, en el caso de la ECB, es dentro de la mandación del sistema federal, porque tienen los objetivos de la inflación baja y de incrementar la output y la empleada. Aquí vemos los assetos totales de los Estados Unidos, la eurozone y el Bank de Centro de Japón. Ahora veas que empezaron con 100 y ahora se desplazó, se desplazó. ¿Puedes imaginar? Así que significa que 40% de la GDP está en términos de las secuencias de los bancos centrales de las balas de balance. No es solo inconvencional, es incluso bizarro. Aquí vemos la división, el U.S. fue, por supuesto, el primero para hacer eso, luego el Bank de Japón, pero cuando los japoneses hacen algo así, lo hacen muy bien. Y luego, aquí tenemos la Unión Europea y la Unión Europea empezó tarde, entonces es detrás de la curva. ¡Bienvenido, Jonathan! No me han perdido las buenas palabras que le he votado a usted, pero de todos modos. Así que aquí vemos la división y el tamaño de esta policía desconvencional. Ahora, dos resultados de QE en el U.S. y eurozone en Japón, así que la policía desconvencional ha sido maybe successful, especially the first round, the first round of QE was more successful than QA2 and QA3, and why? Because it's announcement effect, which matters. If the central bank enters the capital market buying bonds and giving liquidity, the money market is small, but the capital market is enormous. So the announcement effect, the expectation channel, the signaling channel, how you want to call it, is very important. That is the reason why the first round of QE was in the U.S., but also in the eurozone, more successful than the later rounds of QE. And of course the larger liquidity had significant effects on emerging financial markets. And Latin America was not an exception to that. Okay, there we go. Now, the question we wanted to investigate is, is there a differential incidence between conventional and unconventional monetary policy of developed economies on Latin American financial markets? That's the big question. Do we see the difference? Now, here we have the incidence on Latin American financial markets measured in four dimensions, so two are related to asset prices, and two other are related to interest rates. Let me take you through this. So here you have this monitor. You have a conventional part, you have some unconventional parts, here you have the financial market. There is one caveat Brazil, Mexico, Colombia, Chile, and Peru. You know that this group of seven larger countries, there were two countries which had not reliable statistical information. Venezuela, you can understand that maybe, and Argentina. It's not very shopping in that sense. And here you see the asset prices, fixed income market and equity market. And here you see the interest rates, monetary policy rates and active interest rates in terms of lending rates, so both aspects. The hypothesis, what were the hypothesis which were interesting? First hypothesis, unconventional monetary policy developed economies have a similar statistically significant incidence on Latin American fixed income markets, bond markets. Second, conventional unconventional monetary policy in developed economies have a similar statistically significant incidence on Latin American equity markets. Third, in Latin American countries where the exchange rate pass through is low, but both conventional and unconventional monetary policy have null or low incidence on monetary policy and credit market rates, lending rates. In Latin American countries where the exchange rate pass through, that is the fourth hypothesis, high transmission mechanism of monetary unblocked conventional, unconventional monetary policy. In developed economies have statistically significant effect on interest rates. Both monetary policy rates and lending rates, in contrast unconventional monetary policy, is to be expected to have no effects on the interest rates. And then fifth, since the US is the main trading and commercial partner of Latin America, the financial spillo's monetary policy of the Fed are expected to be greater than the ECB and Bank of Japan. And then last, free trade agreements between developed and Latin American economies are likely to exhumate the financial spillo's of international monetary policy. Those are the six hypothesis. Now, literally I will skip that a little bit because of the sake of time, because I always arrive in negative time. We have negative interest rate and I always arrive in negative time and I try to minimize it. So here we have the literature, commercial channel, trade channel, exchange rate channel, credit channel, portfolio rebalancing channel, signaling channel, financial marks channel and liquidity channel. You see that these channels are mostly important for the interest rate of the interest rates and these three channels are mostly important for the asset prices here and then there is one, the liquidity channel, which is mostly important for both of them. Well, I will skip this, otherwise you can read it, otherwise I don't want... Of course I'm a professor, but I don't want to have a whole exposit of the literature. Ok, the methodology. Variables, this is very... So we have of course exogenous variables, so explanatory variables, you could call that conventional monetary policy, which is in this case defined as a one-year government bond yield, unconventional monetary policy, monthly asset purchase by central banks. Then we have endogenous variables, so the depending variables, if you like to... So the bond market, fixed income market, equity markets, monetary policy rate, so... And of course the active interest rate, credit market interest rate or lending rate. You see that those are defined by the ten-year government bond yield, stock market index, monetary policy rate and nominal active interest rates. And here are the sources, you can see that. And of course we have tried to have comparable proxies for that. Oh, that's too fast, control variables. Control variables are necessary to make a good estimation of that. Of course, which you always find in those controls are the CDS prices as normal. CDS is more or less a kind of measure of systemic risk. However, be aware that CDS prices nowadays are less informative than in the past because CDS prices are sometimes used by central bank as financial supervisors information about systemic risk. And then you get the lucas critique, or good or slow, that if you use that information, the value of that information, CDS prices, as a measure of systemic risk becomes less. And so be aware that you often have to be cautious. Then this market volatility indicator fix, this is from the Chicago board options that measures the volatility in the market. Of course, the output gap, difference between actual potential GDP, inflation gap, difference between actual inflation and target, inflation rate, oil prices, commercial integration, you could say trade integration if you prefer that. So it is for free trade agreements, dummy. I will come to that, it's not so successful and I will explain you why. But it's necessary as a control variable. And then the effect of the pass-through, also measured by a dummy, all calculations. So here we have also some new item, I had, sorry, that's the shadow rates. And let me explain a little bit about the shadow rates. I don't know if everybody is aware about the shadow rates, but it's a very important issue by Wu and Xia. It goes back to 1995, Fisher Black. He was the first who spoke about shadow rates. The point is that you cannot, you have to zero lower bound. You cannot go in the negative domain with your normal interest rates. But if you have, for example, quantitative easing, the effective, effective interest rate goes in the negative domain. And how do you measure that? You can measure that with all kinds of forward rates. Wu and Xia, I can really recommend, did that, for example, for the Federal Reserve System. So when the effective Federal Reserve Rate was, you know, between zero and 25 basis points, you cannot go below zero. Then how do you measure then the effective interest rate in the negative domain with the QE exercise? So, and here you see this, this is very nice. Here you see that period, which where, you know, the fat was, you know, at the zero lower bound, and then you see here the shadow rate. Here you see the aeonia, that's the money market interest rate in the Eurozone, which is comparable to the federal funds rate. And you see here that, of course, the aeonia went to zero, but the shadow rates really tumbled in the negative domain. So that's also a kind of benchmark for the effect of conventional and unconventional monetary policy. Question is, but that's a very difficult question, are both policies independent, orthogonal, as econometricians will say. That's a very difficult question to answer, like symmetry and asymmetry in the expansionary and restrictive phase. But that's maybe something for the future. Okay, then we go here to the approaches. I will do that also very briefly. Here you have the panel date approach. What is panel date? Very simple. Cross section time series. You maximize the number of observations. It's, with five countries, a limited number of observation in time. It's the best you can do. And by the way, I think it's one of my favorite approaches. Why? Because you know what you're doing. That's something different than if you go to a factor autoregressive model, or what now people like to do at central banks, DGSE models, makes no sense, because you don't think anymore. It's measurement without a theory. Like Comans, Charlie Comans once said. Measurement without theory, because you have to have a kind of a hypothesis. You have to have something you want to test. Hypothesis, in our case, six hypotheses. And the moment you put everything in the computer, and then especially when you measure only the first and second moment exposed with value at risk of a structural var models or DGSE models, then you lose a lot of information. So here, the differential effect of developed economies spillovers over Latin American countries, eight specifications. What you see here, so the gorgeous variables. So here are the variables and linked to exogenous variables. Now let me explain here bonds, equity prices, monetary policy rate, active rate, landing rates, again. And here you find the conventional monetary policy, C. And here you find the unconventional monetary policy. U.S., E.U., Eurozone, and Japan. So that's the structure and here you find all the control variables. So that's the structure of how it's cooked. You always have to explain the recipe book. Okay, the robustness check is also possible because I spoke already about the shadow rates. Very interesting to check the shadow rates, but the problem is there are no shadow rates for Japan. They are not available. They are only available for U.S. E.U. So here the bond rate, shadow rates, equity rates, monetary policy rates, and active rates. So and here the control variables. So that's the structure. That's just a robustness check. It's a robustness check of the test of the asset purchases. Now, this is already a nice picture where you summarize everything with the dots. Here you have direct, significant, and strong effect. That's the real green dot. Direct, significant, but weak effect in terms of coefficient, indirect, and insignificant. Now, here fix income market, equity market, monetary policy rate, lending, interest rate, conventional monetary policy, so the interest rate policies, unconventional monetary policy part one, that's monthly purchase of bonds, government bonds, and sometimes corporate bonds in the case of TCB. And here the testing of unconventional monetary with shadow rates. Now, what is the outcome? You see that conventional has strong effects, of course, on fixed income markets and monetary policy. That's not surprising, but what you see is unconventional monetary policy, both by the monthly asset purchases and measured by the shadow rates, have, in most cases, direct, significant, strong influence on the bond market and the equity market, and in this case, direct, significant, but sometimes a weaker effect on the victim. But they all have a direct and significant effect. It's very important to see that results. Now, first, this is the question, what is the differential incident, conventional? That's the main question. There we go. So first this, what is the outcome of this? Now, forget the equation. There is a statistically significant relation between the conventional monetary policy in the US eurozone on Latin America, long term interest rate, the bond market. The effect of conventional monetary policy in Japan is statistically not significant. And then here you see the coefficients, which are explained, but I won't elaborate because of the sake of time. Otherwise I go in the negative domain and we won't, don't want it. So conventional monetary on equity markets, so relationship between conventional monetary policy in the US, in eurozone, and the Latin American equity price is indirect. This is observed through long term rates of sovereign bonds, government bonds of the region. The effect of conventional monetary policy in Japan is statistically significant. And then here you see the effects if you're interested in that. I suppose that this presentation also available for people. If they want to, if you want the presentation, I can ask you on the top for that, or the Salma. Okay, so then we have here conventional monetary policy on monetary policy rates. There we go. There's significant relationship between conventional monetary policy, US eurozone, and Latin American monetary policy rates, but the effect is low. Incident of conventional monetary in Japan is not significant. That's also understandable because of the trade and financial relations with Japan are limited. Yes, I know, I see it, yeah. That means that we see that there's effect from US, Fed, the Federal Reserve System. We see that there's effect from the eurozone, but Japan is negligible, that you can forget. Okay, then conventional monetary policy to the lending rates, very important. Here is the result. Relationship between conventional monetary policy in US, eurozone, and Latin American. Lending rates is indirect, so it serves through monetary policy rates. Effect of conventional monetary policy again for Japan is not significant. Now we go to unconventional monetary policy. For the fixed income market, here we have it. There is a significant relationship between unconventional monetary policy, US, eurozone, and the long-term rates in Latin America, but the effect is very low. The incidence is not significant for Japan. Now we go to the equity markets. Relationship between unconventional monetary policy, US, eurozone, and Latin American. Equity price, direct, significant. It means, it's very important, that there is asset price inflation. So the asset prices in Latin America are blown up by QE, for example, of the very reserve, and later on by ECB policies. So that's important to make proper interpretation of these asset prices. Here the effects on monetary policy rates. So there you see that unconventional monetary policy and the policy rates not significant, and lending rates is also not significant, which you already would expect. Okay, results, robustness check, exercises, there we go. Now using the shadow rates, and you know the shadow rates are meant to measure the effect, not only of conventional monetary policy, so at the zero lower bound you cannot go in the negative domain. I can go in the negative domain with my time, but the central banks cannot do that. That means that you have to measure the effects of asset market purchases, quantitative easing for the interest rate, what they would have been. Now, you see using shadow rates as a measure of unconventional monetary policy, you find the same results as before. So it's a robust. That's important to understand, a robust. So there's a clear relationship, not only in asset market purchases, but also in shadow rates. Here we find the effect with equity prices, which is clear that the shadow rates in the US are significant, however in the eurozone, not significant. We have to think about it, but maybe the reason is that in the US, the trade and financial relations, that could be the one explanatory factor in that, and the second one, the US Federal Reserve started earlier with the whole process of QE of asset market purchases. That means, of course, that maybe the measurement of the effect for the equity price in Latin America, you should have longer series beyond 2016. Previous estimates now, the relation between unconventional monetary policy US, et cetera, and interest rate is not relevant. And then we come to the conclusions, and I see that I'm still in positive time. So that's remarkable. But then I use my time and go a little bit in the negative domain, if I may. So here we have the conclusions. Conclusions about the first hypothesis. So you see that this fixed income markets are really influenced by unconventional monetary. That means that the globalization in terms of monetary policy, especially of unconventional monetary policy, is really a serious affair. However, fixed income effect is low, but it's significant. Same way, conventional, unconventional monetary policy have, and that's important, strong effect and significant effect on the equity markets. That's important to know how you make the interpretation of these asset prices. There is asset price inflation. So be very careful with that, with the interpretation of these prices. And then, unlike the third hypothesis, who you know that was about exchange rate, pass through that's not significant, that has not an important meaning, you see also that both, in both cases, exchange rate pass through, was not relevant for the endogenous variables. So let's go to the conclusion, the last set of conclusions. Conventional and unconventional monetary policy, and you have a greater effect than the eurozone. Yes, I'm almost ready. And that is explainable by the trade and financial flows, which are, with the US, much more important. However, it's changing. Trade and financial flows with the eurozone are increasing. That means that the effect in the future, in terms of globalization of monetary policy, both conventional and unconventional, will increase for Latin America. And then we have the, sorry. Then we have the hypothesis six about the free trade agreements, not as controls, they were meant as controls, but not so important. And why not so important? The reason is the following. It's like capital market restrictions. People always look to capital market restrictions. I did a lot of work on capital market restrictions. That's, those are dummies, zero, one. That doesn't matter anything, because capital market restrictions may be there, but not effective. It's much better to look to covered and uncovered intersparity. Covered intersparity is a much better measure for capital market restrictions than the restrictions as such, because they don't necessarily have to be effective. The same applies to these free trade agreements. They may be there, but not effective. And what you want to measure is the effective commercial trade integration. That's what you want to measure. So be very aware about interpretation of that. And then the last, and then I'm finished. That's, of course, a very interesting conclusion. Of course, the financial markets in Latin America are influenced by QE in the US and QE in the Eurozone. But there's good news, eh, there's good news. The good news is that the shocks, the vulnerability to external shocks in Latin America has become less, has become less, has diminished. And then the question is how do we explain that? Now, one is maybe the macro protection policies there, but also the monetary policy rules. And I think that has also to do with the fact that Latin American countries have learned that they need to reduce their vulnerability to, I would say, the shocks coming from the US and shocks coming from the Eurozone. And I think that's good news. And I think Latin American countries, central banks, should continue on that path. Thank you very much.