Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=Transcript of the talk, from the video's captions. Auto-generated: speaker names in particular are unreliable.
=

# Dollar Dominance and the Transmission of Monetary Policy

Authors: Michael McLeay (Bank of England) and Silvana Tenreyro (LSE); presented by Tenreyro — 47th Barcelona School of Economics Lecture, IEC, 11 Oct 2024  
Discussant: None  
Video: https://www.youtube.com/watch?v=KsTyx6q3ymY&t=412s


## Talk (00:06:52 – 00:50:31)

[00:06:52] joining us today it's a real honor thank you very much uh to the organizers [Applause] from here thank you very much to the  
[00:07:06] organizers and to sabadel for having me um and thank you Jordan for the introduction and thank you also for I mean to all the presenters today it was really great to see all this fascinating  
[00:07:20] research and it was really inspiring and and uh fascinating for for me to be sitting here and listening to your interesting work work no so well donear to you today I will talk about dollar dominance and the effectiveness of  
[00:07:35] monetary policy um and this is based on joint work I'm doing with Michael mcle from the bank of England the usual disclaimers apply so my question will be  
[00:07:47] very narrow so let me War uh warn you uh there are many dimensions ofar dominance that are very very important here the question we're tackling is a very specific one can monetary policy help  
[00:08:00] stabilize the economy in a world of dollar dominance that is in a world in which the US dollar is the main currency of invoice in international trade transactions and let me give you a  
[00:08:15] background for uh for the question as we learned from the work of Linda Goldberg and Cedric Teel as well as GDA opth the uh US dollar is the main currency of invoicing in international trade  
[00:08:30] transactions even though trade with the United States is less than 10% of global trade more than 50% of international trade transactions are invoiced in US  
[00:08:43] Dollars now this dollar dominance has led to a new paradigm the dominant currency pricing Paradigm that is Shifting policy views in particular um  
[00:08:56] so actually maybe before I go to that uh you can see um the plot here um illustrating the dollar dominance um so here you can see for example Spain less  
[00:09:09] than um uh 5% of total exports um of Spain are to the United States however over 30% of the international trade transactions by Spain are invoice in  
[00:09:23] dollars so this bars um so in light blue you see the share of exports um relative to Total trade a share of exports with the United States  
[00:09:36] and uh in dark blue you see exports invoicing dollars as a share of of total exports okay so for countries like Argentina and Brazil more than uh 90% of  
[00:09:50] their exports are invoicing dollars even though trade with the United States is less than 20% um so going back to what I was saying um this dollar dominance has um  
[00:10:03] has been shifting policy views significantly the key inference that has been drawn from um the use of the dollar is that it makes the um prices of  
[00:10:16] exports and the demand for exports less responsive to exchange rates and as such this reduces the value of exchange rates as automatic stabilizers and this limits the benefits  
[00:10:30] from having an independent monetary policy which is part of the discussion now in some countries like Argentina for example um if um if um he were to follow the road of dollarization how much it  
[00:10:44] will lose uh from that so according to this view of the world well not much because um monetary policy is not that um that affected after all with dollar  
[00:10:57] dominance um so this is this is a um a bit I got from the IMF uh Global report in 2019 The View that's that is shaping up is that  
[00:11:12] exchange rates are um may need to be supported uh by other stabilization tools perhaps Capital controls because as such they don't help stabilize economies there's very little response  
[00:11:25] from exports uh to exchange rates in the Jon of this literature there's little um expenditure switching in response to exchange rate depreciations or  
[00:11:39] appreciations um now why is this um there are two assumptions made or uh in the um dominant currency literature  
[00:11:52] the first one is that exporters have Monopoly power and the second and most critically um is that prices are sticky and in this um literature they're sticky in US  
[00:12:07] Dollars um now the point that we want to make in this paper is that the prevalence of the dollars dollars um is uh is quite striking in in sectors that  
[00:12:22] tend to sell homogeneous Goods or goods and sales uh or goods and services that operate and highly competitive settings um and indeed many developing and  
[00:12:33] Emerging Market producers are price takers or near price takers in these markets they tend to export Commodities or commodity like products with limiting Market power and even when there is some  
[00:12:47] Market power these um prices tend to be flexible and the extreme example is um Commodities which are quoted always in US dollars but um their prices tend to be very  
[00:13:02] flexible um and I should say that even in advanced economies producers often face very elastic demands in in global markets and and that's the case for example for um the United Kingdom for  
[00:13:16] example um so let let me guide you through the logic of the dominant currency literature and why it leads to those policy implications that monetary policy is not effective and and it it  
[00:13:30] calls for other stabilization tools okay so here we have the picture of an exporter um who is a monopolist and faces a downward sloping uh demand  
[00:13:43] curve uh and we have our marginal cost let's assume for now that the exporter sets the optimal price and we will assume that this price is sticky and is sticky in US Dollars  
[00:13:56] okay now let's consider what happen happens when is there is a depreciation of the local currency the peso depreciates um so in this setting um uh let's suppose that there's some wage  
[00:14:11] rigidity or some rigidity in the prices of non-tradeable inputs in that case the cost of inputs expressed in dollars will tend to fall with with a depreciation so what happens to this  
[00:14:25] monopolies the price um in dollar doesn't change the costs are falling but the price hasn't change the quantities um demanded in the export Market don't change and  
[00:14:38] that's it exports don't move and also in this mods there's no entry so you don't have an entry margin potentially you could think that other uh exporters could enter the market take advantage of this extra  
[00:14:52] profitability um but that doesn't happen in in this uh uh type of models so and this this is where all the sort of policy inferences coming from um now  
[00:15:04] let's think about the opposite example or the The Other Extreme in which the producer is a price taker in international markets and let's let's consider what happens if there is a  
[00:15:18] depreciation of the peso again uh the cost curb will fall um inputs are becoming relatively cheaper when expressing in do uh in this case the price also doesn't  
[00:15:33] change but not because there is a priceing friction there's no nominal friction here uh it's just because the producer is is uh facing a perfectly elastic demand um uh it's a is a price  
[00:15:45] taker um but the quantities change okay so there is a movement in quantities uh demanded here there is expenditure switching towards um this uh  
[00:15:57] this product and uh this leads to a very different uh implication for policy if if taken at face value so I think one of the problems or or one of the issues  
[00:16:11] with the literature is that it has interpreted the pass of the lack of pass through from exchange or depreciations into export prices as evidence of price  
[00:16:23] rigidity in export markets rather than contemplating the possibility that um in equilibrium prices are not changing because the demands are elastic or uh cost cures are increasing um and and so  
[00:16:38] that has basically shaped the analysis and the models that people use to explain the um this feature so somewhere in between is the  
[00:16:50] case of um producers who sell um um goods and services with highly elastic demands and flexible prices um so in this case again we have a  
[00:17:04] depreciation prices change very little not because there is a nominal rigidity simply because uh these demands are very elastic and uh we have an upward sloping cost curve so if we increase quantities  
[00:17:18] uh uh the margin of cost will be higher um so in this setting again we see very little response in prices but a big expansion in quantities and so exchange depreciations can have a  
[00:17:32] stabilization or the effect of um uh boosting exports as in the standard canonical um Mandel flaming model or the ideas that Milton fredman had um uh in  
[00:17:45] mind when he was thinking about um uh the expenditure switching um Power of exchange rates um so just just to um post here  
[00:17:59] the message here is that with monopolists and uh sticky prices in dollars an exchange depreciation does not affect the prices of exports in  
[00:18:11] dollars and the quantities demanded and monetary policy has a very small effect in contrast with commodity or commodity like producers and flexible prices prices and exchange or  
[00:18:26] depreciation again doesn't change the price price of exports much when expressing dollars but it can lead to a large increase in export quantities and in this case monetary policy can have a  
[00:18:39] large effect the constraint here is not on demand is on Supply capacity please feel free to ask questions  
[00:18:50] and um so just um a bottom line I mean we've made a lot of pro Pro progress in understanding that um the dollar is used in in  
[00:19:03] international trade transactions but the fact that uh trade is invoiced in dollars doesn't mean that those prices are uh rigid necessarily in dollars in fact what we um learn from the  
[00:19:17] evidence is that the more competitive the market the more likely it is that producers will use a dollar or some other vehicle currency in um in in pricing um and as I said this m changing  
[00:19:34] assumptions here can have consequential um um implications for the role of monetary policy and its Effectiveness and what we're telling countries to do  
[00:19:45] with um with their um policies and instruments so let me um discuss a little bit the empirics uh here um the first um observation I  
[00:20:00] wanted to share is that a lot um homogeneous products as classified by Rous um those sold in organized markets or those that are based on reference uh prices tend to be a large share of  
[00:20:14] exports for emerging and developing economies um so for developing economies there are over 70% um of their trade is in homogeneous  
[00:20:26] goods for um emerging economies that number is uh uh 60% um for advanced economies uh on average uh that share is about  
[00:20:39] 35% uh Spain is here on the lower end there are some advanced economies that have a large share like Norway Canada Australia New Zealand and similarly um you know there are also liers Chile  
[00:20:51] Ghana T to uh export a much larger share of homogeneous Goods than than the average developing or emerging economy uh the second observation I mean  
[00:21:04] I don't have any data but uh I sure you will believe me is that um commodity prices uh are very flexible they change by the second I mean they are changing at high frequency there is some evidence  
[00:21:17] um uh that I can site here papers like B and kleo Nakamura and steinon that they look at some uh of the disaggregated data even early work by goat R go Bon find that uh the duration of  
[00:21:31] differentiated Goods um prices tend to be five to 10 times higher than those of homogeneous Goods um but for anyone who's following uh commodity markets and uh um the the  
[00:21:46] the frequency of adjustment is uh um is um is super high um the third fact or observation is that I mentioned before  
[00:21:59] is that invoicing in dollars or vehicle currencies uh tend to be more prevalent uh for homogeneous competitive good sectors and that's the intuition that mckinon had back in uh in the 70s uh  
[00:22:13] that then was formalized by baketan and Van winko um the idea is that producers with low Market uh Power uh in homogeneous good markets we tend to price in foreign currencies to  
[00:22:27] um keep very competitive and and keep keep track of their competitors in global markets um goldber and te also have some evidence from from Canadian Imports showing that those firms  
[00:22:41] exporting to Canada tend to price in dollars uh when they're selling homogeneous goods and have uh low market shares um this um chart here shows a  
[00:22:55] correlation between the share of homogeneous uh Goods in exports and uh the share of exports invoice in dollars using the cross-sectional evidence since uh 1985 so we see here a very tight  
[00:23:09] connection for every 10 percentage Point increase in the share of homogeneous Goods in exports we see a nearly 8 percentage Point increase in dollar invoicing in in  
[00:23:23] exports um so that's most of what I have to say for today uh what I would do um now is guide you through um the model that we put together which is you know pretty much a standard model in the literature  
[00:23:38] except that we allow for um flexible prices in export markets um in in the case of homogeneous goods and and then I'll show you some um some further empirical evidence okay so so the model um we use  
[00:23:56] here is the standard Newan setting um we focus on a small open economy and uh we assume that financial markets are incomplete this is not strictly  
[00:24:10] necessary so we deviate a little bit from uh um Jordy and uh and um monachelli's um uh canonical model in international finance um my way to think about J's  
[00:24:25] model um um the the G monell as we know it is that it's more geared towards developed economies in which export um exports tend to be more differentiated  
[00:24:36] so um producer currency pricing tends to be more appropriate and this would be more like the version of that model but apply for apply to um developing or Emerging Markets uh where producers tend  
[00:24:51] to to have little market power um so in terms of production and competition we have supported intermediate um intermediates um in in production this is uh important for uh  
[00:25:05] quantitative purposes because we don't want to overdo the effect uh or the size of effect that we are emphasizing here so we will calibrate it in a uh reasonable way um in a realistic  
[00:25:18] way the model we have um again is slightly different from monachelli um g monachelli in that it allows for a more flexible Market Market structure that allows for more intr sectoral  
[00:25:32] International competition um and heterogenea in price thiness uh so in particular the model will uh have low substitutability across  
[00:25:44] different goods and services but uh more competition within a sector or Goods across varieties and that competition can happen across producers um uh local  
[00:25:57] producers and between local and uh um International producers so we will alter the standard way in which the international finance literature tends to uh aggregate or Nest the production function the typical for those related  
[00:26:12] with that um who know that literature typically we think about consumption of different Goods produced in different countries and within those composite Goods um there's  
[00:26:26] differentiation uh of ities within that sector or Goods uh here we basically the outer layer would be sectors or goods and within that there will be competition across variety both  
[00:26:40] domestically and abroad um then for wage and price setting we will assume uh sticky prices um following uh calvo's uh setup and the  
[00:26:54] setting will Nest um producer currency pricing which is again the G monachelli and um the the the canonical mandal flaming and we will have dollar pricing  
[00:27:08] uh combined with flexible prices uh once you have flexible prices really the invoicing currency be becomes becomes irrelevant okay so the economy here is a small open economy as I said and we take  
[00:27:22] the Dynamics of the rest of the world as given so this is um this is our um representative uh household here uh it  
[00:27:33] maximizes uh lifetime utility and um that utility depends on consumption and labor effort and this is what I was explaining before consumption here is an  
[00:27:46] aggregate of uh different Goods or sectors uh G and within each sector G there are different varieties that can  
[00:27:58] can be produced locally or imported from abroad okay so this is more the type of structure using uh International Trade uh than in international finance um um the uh budget constraint will be  
[00:28:17] standard here um the the household um can use um uh it's labor income rebates from profit profits and any new issue  
[00:28:30] debt either International or domestic um to fund consumption and the payment repayment of interest and and other debts okay so here the um assumption as  
[00:28:44] I said is that there's perfect risk sharing domestically but there's only a risk-free bond internationally so there's imperfect res sharing uh internationally um and then this structure leads leads to a standard  
[00:28:59] uncover interest um parity condition which will link exchange rate to domestic monetary policy when there is a a monetary policy tightening uh we will see an appreciation of a  
[00:29:12] currency um production here is a c dlas aggregate of um um labor inputs and and domestic inputs that again can be produced uh  
[00:29:25] domestically or abroad and we assume that or we allow for um decreasing returns to scale at the industry level um we made the assumption that  
[00:29:39] intermediate inputs are aggregated in the same way as consumption Goods in in um in the utility function so um uh so that allows for a simple aggregation of  
[00:29:52] Demands at the Variety level um so labor markets uh we assume that uh labor is a composite of  
[00:30:04] different uh uh workers that uh belong to different unions and that's a differentiation uh here so um our um union members are indexed by age and uh  
[00:30:18] we have a standard demand and wage setting at the union level following a Calo um framework so in in what I will show you now we will be comparing the  
[00:30:32] implications from flexible prices with Calo sticky prices either under the dominant currency setting or the producer currency um pricing  
[00:30:43] setting um so let me uh uh just say the last bit on the model we close it by assuming that monetary policy follows a  
[00:30:55] um a tailor rule with some smooth here and this is the shock that we're giving to um to the model a monetary policy shock View and all markets clear and let me  
[00:31:11] now turn to the simulations um let me just skip the details of the um calibration it's very standard and follows follows a literature almost um I  
[00:31:25] just wanted to highlight a couple of things first um again in the flexible price specification we allow for homogeneous Goods to have a high elicity of substitution which we take from the work  
[00:31:38] of uh Broad and and Weinstein and um for the rest we use and and we assume that those goods are U flexible U the prices of those goods are  
[00:31:51] flexible um for um the rest uh we assume wage and price rigidity um the implicit assumption here is that these wages and non-tradeable goods prices are adjusting  
[00:32:05] um on average once a year um and again uh we don't know exactly what you know the degree of decreasing returns to scale are in these commodity  
[00:32:19] markets or or homogeneous groups markets uh so we play with different uh specifications by the way I keep saying Commodities but here we really have a monopolistic comp competition structure  
[00:32:31] so we allow for um um non perfectly elastic demands it's just that it's a very high elasticity demand and the framework is is very flexible and allows us to play with these numbers and um um  
[00:32:45] and actually do a lot more than what we're doing here which as you will see is um um is is just trying to um uh explain um what uh um what I show you in  
[00:32:58] the early figures uh so this is the headline um graph in in the paper uh so this show um this these lines here show the impulse  
[00:33:13] responses to a monetary policy Shock by the domestic monetary policy Authority and uh this shows the responses and there different specifications or calibrations of the  
[00:33:27] model so let me start with the producer Curren currency pricing which is again the Canon canonical mandal flaming um and the one behind intuition um early intuition on expenditure switching  
[00:33:41] coming from uh exchange rate fluctuations um so in this case um when there is a depreciation the export price mechanically Falls okay and this is again the case for flexible exchange  
[00:33:56] rates that M frame Freedman had in mind and that leads to expenditure switching and increasing export volumes um for that um for that product okay or for  
[00:34:10] that country's exports um the Blue Line shows um the dominant currency pricing um specification and here by assumption  
[00:34:24] there's no change in uh prices in dollars and because um demands are um um are um because of the Assumption on the elasticities of demand this leads to uh  
[00:34:38] no virtually no change in export volumes um in this case Okay and then finally we have our preferred calibration here in  
[00:34:50] this case export prices in dollars don't change at all so this is the same result in the dominant currency literature and in some of the  
[00:35:02] preliminary um or or preliminary work done on on uh pass through um but uh in contrast to um to the result on export volumes we have a big response on export  
[00:35:16] volumes okay so we can restore the price properties of the dominant currency um literature but um allowing for um the quantities quantity responses in in the  
[00:35:30] producer currency Le this is because demand is elastic and the prices are flexible uh yeah and uh obviously  
[00:35:43] implicit here is Al there are also um assumptions on the supply side that allows for um for an expansion the more you constrain that the more the less of  
[00:35:55] a response you will get and um okay so I mean these are the responses for other variables output uh tends to mimic uh mimic the responses of  
[00:36:10] uh export prices sorry export volume sorry um and um not much more to report here is the averages when we look at the quantitative effects of this so these  
[00:36:23] are the three specifications so this is um the m flaming uh there is sticky uh pricing in the local currency this one is sticky pricing in dollars the  
[00:36:36] dominant currency Paradigm and this is our preferred specification so in all cases with um the monetary policy shock leads to a depreciation of the exchange  
[00:36:46] rate um in the case of um local currenc sorry producer currency pricing we have a response in export prices that is  
[00:36:59] material and uh a consequent for increase in uh quantities exported and produced um here we see the case of dollar pricing there's virtually no  
[00:37:12] change in export prices here and very little change in export volumes and finally this is the other case in which we get a large effect on quantities comparable to the the effect  
[00:37:26] on um producer currency pricing uh while still not getting any change in in prices uh in export prices and going back to um uh to  
[00:37:40] Alberto's question that's you know that's part of the story is uh well it's a combination of of two things elastic demand um and flexible prices so prices change very little here and there is an  
[00:37:54] expansion in in um in activity exports and uh what matters here I mean the the basically and and flexible pricing there's the effect of the  
[00:38:07] depreciation but also the effect of uh moving along the marginal cost car because you increase production and so you move up in the marginal C curve and that will will determine how much prices  
[00:38:20] change in equilibrium in under dollar pricing there's no change because the price is fixed in dollars in under producer currency priceing the the change in prices is exactly the same as the change in uh in the exchange rate um  
[00:38:35] in exchange rate change and uh and here is the combination of um of the exchange rate depreciation and um the marginal cost car and the quantity effect will be just the elasticity of demand times  
[00:38:49] these changes in prices um okay so in in the paper we do some other exercises for those who are interested we um look at different  
[00:39:02] degrees of uh decreasing uh returns to scale in production um and then we play a bit with the calibration in terms of allowing more of a combination of producer currency  
[00:39:16] pricing firms and dollar pricing firms I mean it turns out that if you allow for this fact that when you price in dollars you tend to be homogeneous High elasticity and combined with the fact  
[00:39:29] that those pricing in um the producer currency tend to be tend to sell differentiated Goods um when you merge them together you get the same effects on quantities and uh and the same  
[00:39:43] intuition of mandal flaming and and and um and fredman um so um yes you still needs some uh underlying  
[00:39:59] rigidity or nominal friction which in this case is coming from uh the non-tradables or the you know wages um you still need an effect I mean you need the nominal friction for monetary policy to have an effect  
[00:40:14] yeah yes so we have um so depending on the specification when we do dollar pricing  
[00:40:27] we price them in dollars and um and you know we mimic the Assumption I mean that's that's the dominant currency Paradigm and when we do um producer currency pricing we we price them at the  
[00:40:40] producer um uh prices and so we we that should be relevant I mean we use a calibration from the literature but bearing that should  
[00:40:54] matter um and uh in in fact I mean the more dollar pricing there is in the imported goods obviously the more costly um um imported prices imported  
[00:41:08] goods will become with the depreciation so that will affect um the trade balance and so on and so forth so that's another important margin uh for now we just take the calibration in the literature and um  
[00:41:20] um play with that but going back to um Vladimir's um question and combining it with you to get an effect what's important is that in the cost base of these producers there's some part of it  
[00:41:35] that is um uh rigid in in the Lo in the producer currency that's what gives the monetary policy its Effectiveness and so you could have some imported inputs into production which will tend to lower the  
[00:41:49] effectiveness of monetary policy or lower the you know expansionary effects of a depreciation but as long as you have um um non-tradeable or you know sticky wages um you will you will get an  
[00:42:03] effect and then the discussion is quantitative and it will vary across countries and uh and you know that's there's a lot more to do there like  
[00:42:14] um okay I I think I just made clear the the distinction between the um and now it's a 2023 I think uh a uh paper and uh and the calibration in our model so the  
[00:42:28] key difference is that we have um low elas sorry uh low Market power and high elasticities of demand whereas in in the calibrations in the literature there's um High Market power and and uh  
[00:42:43] inelastic uh demands um the difference is that export prices in one case are sticky in our case are flexible both cases lead to very low pass through or limited pass through  
[00:42:57] from exchange rate changes into export prices but they lead to very different um implications for quantities um so just to highlight that pass  
[00:43:13] through into prices cannot help distinguish the two models and uh just observing lack of pass through is not telling us whether the price sticky there's a nominal rigidity or simply  
[00:43:27] there's just high high elasticities of demand and uh um the producers are price takers um or near price takers um so the  
[00:43:38] next exercise we do is to um use um identified monetary policy shocks to look at the effect on export volumes and activities and uh uh so basically  
[00:43:50] replicate what we do in the model um uh in the data or try to uh do the same exercise now um before let me just tell you some other people have done this uh type of exercises before for some  
[00:44:03] economies there's a not a lot outside the US and Europe uh but uh there's a paper uh by champagne and seel for Canada and what they find is a tightening uh monetary policy tends to  
[00:44:17] lead to false to an appreciation of exchange rate and fals in exports and overall activity um for the developed countries for the the UK we have the same patterns  
[00:44:30] Titan inly to exchange rate um appreciations and uh um unfolds uh in exports this an an increase here is an exchange rate  
[00:44:41] appreciation um and uh our sort of conjecture is that in most developed countries this is mostly um driven by the old intuition of producer currency  
[00:44:53] pricing and differentiated Goods uh given the the of homogeneous Goods is uh tends to be smaller um now what we learn from um large  
[00:45:05] devaluations and this is um a tricky exercise because obviously depreciations or devaluations happen for a reason um what we learn is that after large  
[00:45:18] devaluations uh and here I have a few examples Argentina in 2001 um Brazil after the Russian crisis and um and Mexico um in the tequila that tends to lead to Big expansions in  
[00:45:32] exports relative to Trend or relative to um Global exports um as I said this this I mean depreciations don't happen in a vacuum they are um fully um endogenous  
[00:45:47] uh but to the extent that this depreciation or devaluations tend to F happen when the economy is very weak and exports are weak the fact that we see these big increases in exports actually  
[00:46:01] is uh is quite telling so um um so so that's um that's some um sort of hint in the direction of of of  
[00:46:13] the point we're making in this paper um we go a little bit more there in and I say more there in because it's it's already pretty brave to run bars for some advanced economies given that there  
[00:46:26] are changes in um institutional setups tools and uh it's hard to identify monetary policy shocks uh it's even more the case for developing and emerging  
[00:46:38] economies um but again um we took that step um we used a a database that Branda Marquez and brand Marquez and coauthors  
[00:46:49] put together um they uh uh produce a series of monetary policy shocks which they identify as residuals uh from a  
[00:47:00] tailor type rule which um uses forecast inflation forecast output gaps and exchange rates and they extract from that what they call the monetary policy shock and we feed this monetary policy  
[00:47:14] shock into our um local projection uh uh exercises and again there's a lot of assumptions here and there's a lot of  
[00:47:26] noise but um this is what you would get if you did that that ass sunction again this is done for a a b a sort of for a panel of 38 um emerging  
[00:47:40] and developing countries um so what uh what we have here is that again very similar to what we see for Canada and the UK for these countries as well a tightening of uh monetary policy leads  
[00:47:54] to an exchange appreciation and a fall in um in exports and industrial production which is the data we have here so it's it's very similar and we don't find this evidence that  
[00:48:07] exports are um uh unresponsive or activi is unresponsive um so just um to conclude and leave some  
[00:48:19] time for um discussion um this literature the dominant currency uh literature has highlighted the importance of the dollar in international trade transactions and in the finance world and this is an  
[00:48:33] important um question to uh um issue to study um however the policy conclusions are based on two assumptions  
[00:48:44] that are not backed up by the data um and this is that um and again the the fact that um the assumption that prices are sticking in in dollars and there's  
[00:48:57] Monopoly power um what we find or uh argue in this paper is that the dollar is more likely to be used in highly competitive markets where prices tend to be  
[00:49:11] flexible uh so in the paper we uh try to work with a more General framework that allows for greater Global competition and price flexibility in some Goods or sectors uh while uh having Monopoly  
[00:49:25] power and rigidity in others and the model is very flexible and can potentially be calibrated more um precisely for different countries with with better micro evidence um to  
[00:49:38] parameterize it um and as I said I mean we get very little pass through of exchange rates into prices but we don't think that's evidence of um uh a nominal price  
[00:49:52] rigidity quite the opposite in this type of settings and we find that um the evidence from large devaluations three case studies that we have is consistent  
[00:50:04] with um this specification of the model as is the evidence coming from identified uh monetary policy shocks uh so we think that perhaps it's time to  
[00:50:17] revisit um the policy advice especially because most of the clients of the IMF for example tend to be um uh emerging and developing economies with a high share of homogeneous Goods so I'll PA it  

## Q&A (00:50:31 – 01:12:44)

[00:50:31] here and happy to take questions thank you thank you Sana as always uh very  
[00:50:46] clear very simple and uh um so that it helps a lot in the debate in terms let me just throw a a a suggestion you're  
[00:50:59] looking at export sectors or tradable sectors where the the price is in dollar in dollars and the domestic inputs are a  
[00:51:10] listing part in domestic currency let's think about non-tradeable sectors where you're going to see the opposite perhaps they have inputs in dollars and the domestic  
[00:51:25] prices are not quoted dollars or quoted in domestic currency perhaps one way to assess the effect on quantities is to see whether there is a big reallocation from uh tradeables to non-  
[00:51:38] tradeables when you have such a devaluation so that if the economy the share of tradeables and non-tradeable remains very similar perhaps uh there has been little  
[00:51:52] effect on quantities but perhaps what you would observe that the tradeable sectors expand a lot when you have this depreciation but the non tradeables contract and that would give you some hint about the  
[00:52:07] quantities is this a correct view or am I missing something or have you considered yeah I think that's an excellent question and it has to be combined with the capacity constraints in these industries I mean I haven't  
[00:52:21] thought about know this but I think you know you are right that there should we should see somewhere the re some relocation can move exactly um yeah I mean we do know that in large  
[00:52:34] devaluations from the work of Ariel B Etc the the non-tradeable prices plays a big role uh because those are sticking the local currency so that mechanism is here and but it would be great to look  
[00:52:46] at the data and look at the the reallocation between non-tradables and tradeables one thing I that you know is mly related I don't know if uh um is what happened during  
[00:53:00] brexit um because when the again the depreciation of of Sterling during um after the referendum was not an exogenous events I mean markets were pricing the fact that in the future  
[00:53:15] perhaps productivity in the tradeable sector will grow more slowly or will be lower uh than otherwise uh but the depreciation um actually caused a big relocation or or a big expansion in the tradeable sector at the time it was all  
[00:53:30] puzzling I mean we thought or we were saying Economist were were saying that brexit will be very bad for the tradeable sector but when he after the referendum the depreciation happened the productivity was not changed the trade  
[00:53:43] barriers didn't happen so all the effect there was just exchange rate and we saw this expansion and that was a sweet spot for exporters uh after the referendum so it goes in that direction and I you know in the paper I have with and Bradman and  
[00:53:57] others we see some of these um you know sectoral effects we haven't done it for these countries and I suspect for some it would it could be feasible to get get the data and um but yeah it's something  
[00:54:11] that um we might think to yeah we we we probably could do um at least for some of these countries to uh to test the mechanism and in in cases like the three cases that you had the three countries that you had more homogeneous in ter  
[00:54:26] terms of of export like the Cooper in Chile cocoa in Ghana or Petroleum in in other countries those have a very low cost compared with value because you are taking natural resources that are not  
[00:54:41] included in the in the equation no there are some natural resources here that we just can pump out pump out yeah I mean that's a question of constraint I don't I mean and I think you have to go sector by sector I you know because there will  
[00:54:55] be a difference in uh you know in scaling up I mean this is Argentina um Mexico Brazil and it might be different um you know for some countries in  
[00:55:08] Argentina I think there's a lot of um um what do you call the the sorry I have my brain FX um so the extensive Market is very  
[00:55:22] important we don't have it in here but I think it's you know sort of is this can be as a shortcut for that when there is a big depreciation a lot of producers that were not competitive before selling I don't know leather products can enter that market and this is you know in some  
[00:55:37] sense what what can push um exports as well uh but for other countries for Mexico you know it might be oil and and I think it's important to get into the specifics of these economies and try to  
[00:55:52] understand you know what's you know what what are those capacity constraints and uh uh which we haven't done it and I think it would be interesting what is remarkable is that again I mean we see very similar responses between Advanced  
[00:56:05] and emerging developing economy so so it cannot be that um the invoicing um in dollars is preventing you know this adjustment but the quantities might different than uh I  
[00:56:18] guess it relates a bit to um to what Jama was saying because in some in some cases it might be easier to allocate um uh labor or uh inputs from one sector to the other in some others it might might  
[00:56:32] be different so maybe that's also something to explore that um um diversity here or heterogenity I'll ask a question about because in the model you were looking at monetary policy induced devaluation yeah and I'm thinking whether under DCP you  
[00:56:47] would actually not get a contraction in exports and because when you devalue there's domestic boom but exports don't export prices don't change so you should reallocate inputs towards  
[00:57:00] domestic non- tradeables or domestic production right and I think empirically I don't think you will ever find um so let say you have a monetary expansion so there's a domestic Boom at the same time  
[00:57:13] when there's depreciation of the currency exports don't change if you control for the cost yeah but now there's a desire for there's a crowding out of the exports due to domestic pressure and inputs let's say if it's labor  
[00:57:27] an example it should reallocate to non tradeables if I'm if I'm not mistaken here but it seems like empirically you will never probably detect a case where uh monetary expansion leads to contraction of exports I don't know so I mean I I have to think about it um I  
[00:57:42] mean I guess even in in the dominant currency setting there will be an effect on monetary policy through financial markets that will boost the economy through that it's not going to happen through the export Market but  
[00:57:56] you know through the financial market and and you will see that and so in that regard you should see yeah an expansion in domest in the domestic uh sector uh  
[00:58:09] yeah so in the DAT yeah is yeah that's um it makes sense with uh uh I think the way they calibrate it and why don't they don't get a negative and rather a zero but I mean we're talking about uh you  
[00:58:22] know specific calibration uh is that they still have some Calo setting so some some producers are adjusting in in in their calibration when um so so  
[00:58:36] there's sticky prices in dollars but there will be some producers that get get touched by the magic W um but but yeah you're right that otherwise I mean if they were fully stuck with the prices  
[00:58:49] all of them um then uh we should we should see uh um that movement um yeah and Dav yeah well thanks again for for this  
[00:59:02] very inspiring talk I just wanted to hear your your view about how we could interpret these results for policy makers so the countries that you have mentioned typically have engaged in some form of control you know some exchange  
[00:59:17] rate agreements and so on however what you are suggesting here is that since prices are flexible in the foreign currency maybe it's not so crucial to stabilize the exchange rate is this something you should learn from from  
[00:59:30] your analysis or not so all we are saying I mean I don't want to exaggerate the message we're saying is that exchange rate um can act as automatic  
[00:59:43] stabilizers and um and that's very different than the policy conclusion drawn from the dominant currency Paradigm that they cannot okay so then you might say well a  
[00:59:56] country may still need to put Capital controls for other reasons that are not in this paper and you know totally fine with that but you know it's not coming from this mechanism that has been emphasized in the literature and if we  
[01:00:10] want to discuss Capital controls we need to present the mechanism and then quantify it and and then try to gauge how much is needed but it's not you know coming from that um you are right that  
[01:00:24] in practice countries have manipulated or you know um and I come from a country that how many exchange exchange rates uh do we have uh in Argentina I mean a lot so there's a lot  
[01:00:36] of controls and a lot of intervention by the central banks and there are different exchange rates for different um I mean that's obviously an extreme and perhaps not the one to follow but uh and there have been more successful  
[01:00:50] cases of uh exchange rate intervention and there's more need to study um all the message here is that however you think about the problem is not coming from this friction and if you want to  
[01:01:04] quantify how much you want to put in terms of uh mcro proofs or Capital controls uh you need a model that will help you think about it and uh and and put numbers on  
[01:01:17] it thanks so um I just wanted to ask about like the quantitative takeaways from from what you showed us that basically you if I understood correctly you presented as these three paradigms of the theoretical model they had different qualitative predictions for  
[01:01:31] how exports should respond and then you showed us empirical evidence on exports responding and qualitatively agreeing with with uh one of the paradigms more than the others have you also like thought about the magnitudes that maybe if reality is actually a mix of these  
[01:01:46] three like a convex combination of these three different paradigms that could you maybe invert the exercise and do like impulse response matching and say how close we are actually to each of these paradigms is that something doable I  
[01:01:59] mean I yeah I think we we so so we we Tred to do some quantitative exercises in in in the model no and go but but then the question is um you know if if you're and and and this the result we  
[01:02:13] get is that we when we combine producer currency pricing plus differentiated goods and dollar pricing and homogeneous Goods you get the same result as if they were all  
[01:02:26] um the traditional model Mandel flaming or you know the the flexible um calibration that we have um so what you suggest is like okay let's let's have a third category in which the dollar  
[01:02:41] pricing is actually sticky dollars and you know we um we you know we can do that as well and uh more like uh potentially that the model is actually like a flexible version that nests all  
[01:02:53] of the paradigms where the distance from the different paradigms is a is a parameter and that you could you know quantify that parameter with the size of the impulse responses of exports for example yeah  
[01:03:07] um okay I I'll have to think more about that I yeah so just when I compare Canada and the emerging and developing countries it looks very similar in terms of act you  
[01:03:22] know magnitudes as well and that's what we get from the model so if we were to you know introduce this other category um I think it it won't mat as well but I  
[01:03:33] yeah with better data one could could actually try to pin down better these parameters and perhaps there are you know some countries out there that use allar pricing for differentiated goods and they're  
[01:03:46] sticky my intuition would be I know based on the fact that exporters or firms typically like uh depreciations I mean those in the export Market the fact that they like it so much is that there  
[01:03:59] is you know there's a gain there and you know either you get it through the intrinsic Market or firms you know being able now to sell in bigger or in a wider range of markets uh when there is a  
[01:04:12] depreciation but yeah I mean we we haven't explored it much but yeah um thank you thank you San I just had a very simple question for those of us that are not really in this literature  
[01:04:26] so imagine you started by telling us look if I have this pricing in dollars devaluations may not be effective no essential now you show us look there's a bunch of countries if my demand is elastic even if I price in dollars well we're going to get back the conventional  
[01:04:41] result but then you also show us the case of the UK where you also seem to get the conventional result there I think you said that well there the dollar pricing may not be as relevant it's domestic currency pricing presumably because the producer currency  
[01:04:53] yeah producer because the case a large economy so for which type of countries would this Paradigm apply the Paradigm that you're trying to the bank in a sense yeah I I would not call it paradig just  
[01:05:07] a calibration no no no but I'm saying the this basic notion that you're saying well there's this predominant view that the dollar pricing would suggest that devaluations are not that effective as an automatic stabilizer but here you're showing us in Emerging Markets it seems  
[01:05:21] to be countries like the UK it seems to go in the right direction in which type of countries okay so what we see here is that exchanger depreciations tend to increase exports where we're talking  
[01:05:34] about the UK which sells more differentiated products hugely but you know and versus um uh you know emerging and developing  
[01:05:46] economies my sense is that the model the mandal Flaming setup with differentiated goods and producer currency pricing applies more to C countries like the  
[01:05:59] United States or you know more developed economies obviously nor Canada or Australia that sell more homogeneous Goods but yeah countries that tend to sell differentiated goods and in those countries you're back to Milton fredman  
[01:06:12] a depreciation leads to expenditure switching yeah in for emerging and developing countries you have more dollar pricing but as I said I mean that tends to uh apply to um products that  
[01:06:27] are homogeneous and with flexible prices and again you get the same response for exports uh just in in terms of dollar pricing the prices don't move but uh you you get the Boost in coming from  
[01:06:40] monetary policy so so yeah you know answering your question we should see more um Mandel flaming producer currency pricing in developed economy selling  
[01:06:52] differentiated goods and uh um dollar pricing but with more flexibility in emerging developing economies I agree but polic less to that  
[01:07:05] in both of these countries exchange rate are automatic stabilizers yeah yeah yeah in in both you know in all countries that would be a work inclusion so we're basically just sort of arguing against the recent  
[01:07:19] conclusions from this literature that said no they're not actually we think they are and and um oops um yeah hi thank you um I had a question I mean you've I think it's kind of related  
[01:07:34] to what we're saying um but homogeneous Goods there includes Commodities and I'm guessing other homogeneous goods and with quantities especially I'm guessing  
[01:07:46] um yes developing countries are price takers but those would be highly concentrated um markets as well and I'm ing whether that would have an impact on how monetary  
[01:08:01] policy the the depreciation feeds through the rest of the economy if we have very high markups in the commodity um so you're thinking um so  
[01:08:16] they're competitive in um in global markets these producers have say some Commodities but then they they command some monop monopsony power within uh you know Vis the inputs  
[01:08:28] I mean that's possible um and and so then your question is very important how much a depreciation which will benefit um the commodity producers will per percolate to the rest of the economy and you know there are issues of  
[01:08:43] distribution and um so yeah I I cannot say anything about that I mean it will depend on the country pretty much and and the um structure of taxes and  
[01:08:56] transfers in that country how redistributive the system is um in some you know Norway you know will put will redistribute a lot more than some Latin American countries and they have their  
[01:09:10] Sovereign wealth fund where they put in some of this money as well um but um you know um that that's not a good example actually but yeah so I think yeah you  
[01:09:22] need to um yeah it's it would be Case by case how much this goes to the rest of the economy and um yeah monopsony Power will be  
[01:09:35] important and uh and tax structure uh in the economy um so um yeah I I don't have much to say but it's obviously an  
[01:09:46] important question and uh um so our what we are saying is mostly at the macro level you know about the power of these exchange rate as automatic stabilizers but it's a very  
[01:10:00] aggregate level and um don't have much to say about the trickle down yeah um thank you very much I have just a very short question on you you showed you  
[01:10:11] know the the ratio of um uh dollar invoices and started your your your talk with uh the weight of dollar invoices compared uh to the to you know worldwide  
[01:10:25] I was wondering about the trends do you do you is do you know anything about whether this is increasing or decreasing over time uh the share of dollar invoices is this dominance changing over  
[01:10:38] time is it yeah okay I don't know much I think people have been studying that um there's now a literature on vehicle currency um I'm not sure we have long  
[01:10:51] time series on um on um on on on on these shares um there is a there's a paper by Amin Bo and and co-authors where they they do have some  
[01:11:06] is the data is a little bit patchy but um I mean um um so in the regressions that I showed you most of the identification is coming from the pro crosssection you know cross countries because it's it's a bit bit patchy for  
[01:11:20] when they and we use their data for when they can measure um I'm only familiar with um you know with recent work on V concurrency and uh I think for example there has been a an  
[01:11:35] increase in the use of the Euro for example and uh um but yeah I should know this but uh um yeah I don't know what the trends are um something to look at I  
[01:11:47] yeah as I said in in our database because the time series is so patchy we we just use the cross-sectional uh but um yeah I think it's about time right we made so many questions that you're  
[01:12:01] probably thinking well well it looks like your conference was so uh interesting and it brought a lot of uh questions and thinking so that means that we really want to thank you Silvana for being with us and giving such a  
[01:12:16] great talk and thank you everybody for attending the trada and the conference uh the BSC lecture that uh was as usual sponsored by Bank Saad and Sophia said  
[01:12:29] it was the 47 this year okay so good counting so thank you everybody and uh we'll keep doing many things during the year [Music]