Macro Vision October 2, 2017 How the TLP can impact monetary policy In this report, we estimate that, when fully implemented, the new long-term interest rate (TLP) will allow a reduction of about 2.2 p.p. in the benchmark Selic interest rate, ceteris paribus. TLP should also contribute to substantially boost the Selic rate s influence on the economy: we estimated that the Selic rate impact on average interest rate can increase by about 50%. In early September, the Brazilian Congress approved Provisional Measure nº 777, creating the Long-Term Interest Rate (TLP) to replace the TJLP as a benchmark in loan contracts offered by development bank BNDES, beginning on January 1st, 2018. Initially, the TLP will start from the same level as the TJLP effective on that date, but will converge during a period of five years to the change in the consumer price index IPCA plus the real yield of the five-year inflation-linked bonds issued by the National Treasury (NTN-B). 1 This report analyzes how earmarked credit impacts the Selic rate and then how the creation of TLP, along with possible changes in the volume of earmarked credit, may impact monetary policy implementation by the Central Bank. In recent years, the share of earmarked credit in the Brazilian economy widened substantially, to 50% nowadays from 32% of total credit in 2008. Segmentation between the two credit markets is evident, as the average interest rate charged on earmarked credit is systematically lower than for non-earmarked credit. 100% Brazil's credit composition Non-earmarked Earmarked 60% 50% Non-earmarked Average Earmarked Interest rates 80% 60% 68% 66% 62% 61% 60% 57% 54% 51% 50% 50% 40% 30% 40% 20% 20% 32% 34% 38% 39% 40% 43% 46% 49% 50% 50% 10% 0% 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 0% Jul-11 Jul-12 Jul-13 Jul-14 Jul-15 Jul-16 Jul-17 As empirically shown below, the interest rate on earmarked credit is significantly less responsive to fluctuations in the Selic rate than the interest rate charged on non-earmarked loans. This relative insensitivity to the Selic rate and large volumes of earmarked credit produce at least two relevant practical effects: 1 See the appendix for a detailed discussion on the transition from TJLP to TLP. Please refer to the last page of this report for important disclosures, analyst and additional information. Itaú Unibanco or its subsidiaries may do or seek to do business with companies covered in this research report. As a result, investors should be aware that the firm may have a conflict of interest that could affect the objectivity of this report. Investors should not consider this report as the single factor in making their investment decision.
Macro Vision October 2, 2017 (1) Monetary policy becomes less potent: Selic moves have smaller impact on aggregate demand, due to the autonomous component that does not respond to the benchmark interest rate set by the Central Bank. (2) Interest rates charged on non-earmarked credit tend to be higher: Under constant financial conditions (measured by the economy s average interest rate), the larger the share of subsidized credit, the higher interest rates charged on non-earmarked credit must be. To make the investigation of these effects in a simple way, we used the following expression: where, i A is the economy s average interest rate; i A = α i E + (1 α) i N i E is the average interest rate charged on earmarked credit; i N is the average interest rate charged on non-earmarked credit; and α is the share of earmarked credit in total credit in the economy. 2 We modeled below the drivers of interest rates charged on earmarked and non-earmarked credit. One realizes that the TJLP and the Selic (through its impact on the Reference Rate - TR) explain much of the change in interest rates charged on earmarked credit between March 2011 and July 2017. As for interest rates charged on non-earmarked credit, the role of the Selic rate is much more relevant, as this coefficient is much higher than for earmarked credit (1.79 vs. 0.15), while the unemployment rate also matters as it captures the impact of the economic cycle on banking spreads. All variables are statistically significant. i E = 0.009 + 0.15 (Selic) + 1.07 (TJLP) + ε E ; R 2 = 0.89 i N = 0.018 + 1.79 (Selic) + 2.17 (Unemp) + ε N ; R 2 = 0.98 where Selic is the Selic rate, Unemp is the unemployment rate and ε E and ε N represent the respective errors of each regression. Both regressions have good explanatory power and can be used to build a model for the economy s average interest rate, as seen below: i A = α [0.009 + 0.15 (Selic) + 1.07 (TJLP)] + (1 α) [0.018 + 1.79 (Selic) + 2,17 (Unemp)] + ε A ; R 2 = 0.97 where ε A = α ε E + (1 α) ε N This expression makes it clear that, for a certain level of financial conditions (i.e., a given level for the average interest rate) and given the cyclical stance, there would be different levels for the Selic rate, depending on the share of earmarked credit (α) and on the TJLP level. 2 Data provided by the Brazilian Central Bank (BCB). As an alternative to the average interest rate, we could have used the new series computing the average cost of credit operations (Credit Conditions Indicator, recently created by the BCB), but this series is short and does not reflect the evolution of financial conditions at the margin. Page 2
Selic Rate Selic Rate Macro Vision October 2, 2017 We use the expression above to simulate how the degree of subsidized credit through the TJLP or credit volume affects the Selic rate that is needed to keep a certain level of financial conditions, given the economic cycle. As expected, the lower is the TJLP and the wider is the share of earmarked credit, the higher is the Selic rate needed to maintain stable financial conditions. These results are seen in the comparative statics exercises below. 3 12.00% 11.25% 10.50% 9.75% 9.00% 8.25% 7.50% 6.75% 6.00% Static relation between Selic rate and TJLP 5.25% 0% 1% 2% 3% 4% 5% 6% 7% 8% 9%10%11%12% TJLP Current Situation (7.00%, 8.25%) 23.25% 20.75% 18.25% 15.75% 13.25% 10.75% 8.25% 5.75% 3.25% Static relation between Selic rate and share of earmarked credit 0.75% 25% 33% 40% 48% 55% 63% 70% Share of earmarked credit (% of total) Current Situation (50%, 8.25%) Results show that, to keep the economy on the same indifference curve of financial conditions, a hike of 1.00 p.p. in the TJLP would allow the Selic to fall 0.55 p.p. Using the historical 4 p.p. spread between the Selic and TJLP and assuming that, once the TLP becomes effective, the average spread between it and the Selic rate will be near zero, the Selic rate 4 could be approximately 2.2 p.p. lower for an equivalent level of financial conditions. Importantly, this would just be the direct impact. The creation of the TLP should also have a price effect i.e. induce a decline in the share of earmarked credit in the economy, given that its cost will be closer to that of nonearmarked credit. Our comparative statics show that a drop of 1.00 p.p. in the share of earmarked credit would allow the Selic rate to be 0.35 p.p. lower, assuming the same financial conditions. We emphasize that, given the concavity of such relationship, larger reductions in earmarked credit would have marginally diminishing impacts on the Selic rate. The model can also be used to illustrate how the footprint of earmarked credit impacts the effectiveness of monetary policy or the Central Bank s ability to affect financial conditions in the economy through the Selic rate (see chart below). Assuming once again that, after the initial transition period, the TLP has an average spread near zero with the Selic rate, there would be an increase of about 50% in the influence of the Selic rate on the average interest rate of the economy. Moreover, the earmarked credit share in the economy will have a less pronounced impact on such effectiveness, as can be seen from the smaller slope of the post-tlp. Both effects can be seen in the chart below. 3 The exercise suggests that a significant reduction in earmarked credit would drive the Selic rate to very low levels, but this partly reflects the current cyclical situation (high unemployment), which served as the basis for the comparative statics analysis. 4 To produce a conservative estimative, we used 4 p.p. as the typical spread between the Selic and TJLP, the lowest point in the historical series of the 36-month moving average of this spread. The assumption of near-zero spread between TLP and Selic is consistent with the historical relationship between the real Selic rate (swap préxdi 360 days discounted by Focus survey s IPCA expectations for the next 12 months) and the 5-year NTN-B rate. Page 3
Selic rate impact on average interest rate (p.p.) Macro Vision October 2, 2017 1.8 1.7 1.6 1.5 1.4 1.3 1.2 1.1 1.0 0.9 0.8 0.7 0.6 0.5 0.4 Monetary policy effectiveness Current effect Effect after TLP Current Situation (50%, 0.97) 25% 30%35%40%45%50% 55%60%65%70%75% Share of earmarked credit (% of total) Furthermore, by using market rates to guide long-term credit, the creation of the TLP will provide a key contribution to the development of long-term financing instruments in Brazil. That will open another channel for the Selic to influence the economy, with implications in terms of financial conditions (easier conditions, ceteris paribus) and the power of monetary policy (more effective). In that sense, a limitation of the exercises above is the emphasis placed on a partial metric of financial conditions: the economy s average interest rate. Finally, the exercises above are based on comparative statistics, meaning that changes in the economic situation (captured in our model by the unemployment rate), as well as simultaneous changes in the share of earmarked credit in total credit (α) and in the TJLP, may change the elasticity figures reported herein. Such effects tend to be secondary, particularly in the short run, so that the exercises above may be regarded as a good proxy for the impacts caused by the TLP and the reduction in earmarked credit on the economy. Fernando M. Gonçalves André Matcin Page 4
Macro Vision October 2, 2017 Appendix: Transition from TJLP to TLP According to the baseline text approved by Congress, TLP will only apply to new contracts and there will be a smooth transition before the new rate fully reflects market parameters. In this context, the TJLP will exist as long as there are outstanding loans granted at that rate, and its value will continue to be set every quarter by the National Monetary Council (CMN). TLP will become effective for loans contracted after January 1 st, 2018 and will be formed fully by the change in the IPCA and a fixed real interest rate. The fixed interest rate will be equivalent to the daily average of the real yield on five-year NTN-Bs in the previous three months. At this rate an adjustment factor will be applied that converges linearly to one, in annual adjustments, within five years. The initial adjustment factor will be determined by TJLP, NTN-B and inflation expectations for the next 12 months in January 2018. The fixed interest rate will be the one in effect on the date that the credit is contracted and will be applied uniformly throughout the entire financing period. In a simplified way (for more details see BCB Resolution No. 4,600 5 ), we have the following implementation steps for the TLP: Consider the following variables on January 1st, 2018: IPCA*: IPCA expectations for the next 12 months, TJLP*: Current TJLP rate NTNB*: Daily average of the real yield on five-year NTN-Bs in the previous three months The initial TLP value, to be defined on January 1st, 2018, will be: TLP = (1+IPCA*) * (1+ α * NTNB*)-1=TJLP*, where α is the adjustment factor to be defined so that TLP equals TJLP. Thus: α=(tjlp*-ipca*)/((1+ipca*) * NTNB*) Subsequent TLP values will be defined monthly: TLP= (1+IPCA realized) * (1+ α * NTNB)-1, where α increases linearly, in annual adjustments, until it reaches 1 in five years. Macro Research Itaú Mario Mesquita Chief Economist Tel: +5511 3708-2696 Click here to visit our digital research library. 5 Only in Portuguese: http://www.bcb.gov.br/htms/normativ/resolucao4600.pdf?r=1 Page 5
Macro Vision October 2, 2017 Relevant Information 1. This report has been prepared and issued by the Macro Research Department of Banco Itaú Unibanco S.A. ( Itaú Unibanco ). This report is not a product of the Equity Research Department of Itaú Unibanco or Itaú Corretora de Valores S.A. and should not be construed as a research report ( relatório de análise ) for the purposes of the article 1 of the CVM Instruction NR. 483, dated July 06, 2010. 2. This report aims at providing macroeconomics information, and does not constitute, and should not be construed as an offer to buy or sell, or a solicitation of an offer to buy or sell any financial instrument, or to participate in any particular trading strategy in any jurisdiction. The information herein is believed to be reliable as of the date on which this report was issued and has been obtained from public sources believed to be reliable. Itaú Unibanco Group does not make any express or implied representation or warranty as to the completeness, reliability or accuracy of such information, nor does this report intend to be a complete statement or summary of the markets or developments referred to herein. Opinions, estimates, and projections expressed herein constitute the current judgment of the analyst responsible for the substance of this report as of the date on which it was issued and are, therefore, subject to change without notice. Itaú Unibanco Group has no obligation to update, modify or amend this report and inform the reader accordingly. 3. The analyst responsible for the production of this report, whose name is highlighted in bold, hereby certifies that the views expressed herein accurately and exclusively reflect his or her personal views and opinions and were prepared independently and autonomously, including from Itaú Unibanco, Itaú Corretora de Valores S.A. and other group companies. 4. This report may not be reproduced or redistributed to any other person, in whole or in part, for any purpose, without the prior written consent of Itaú Unibanco. Additional information on the financial instruments discussed in this report is available upon request. Itaú Unibanco and/or any other group companies is not, and will not be liable for any investment decisions (or otherwise) based on the information provided herein. Additional Note to reports distributed in: (i) U.K. and Europe: The sole purpose of this material is to provide information only, and it does not constitute or should be construed as a proposal or request to enter into any financial instrument or to participate in any specific business strategy. The financial instruments discussed in this material may not be suitable for all investors, and are directed solely at Eligible Counterparties and Professionals as defined by the Financial Conduct Authority. This material does not take into consideration the objectives, financial situation or specific needs of any particular client. Clients must obtain financial, tax, legal, accounting, economic, credit and market advice on an individual basis, based on their personal characteristics and objectives, prior to making any decision based on the information contained herein. By accessing the material, you confirm that you are aware of the laws in your jurisdiction relating to the provision and sale of financial service products. You acknowledge that this material contains proprietary information and you agree to keep this information confidential. Itau BBA International plc (IBBAInt) exempts itself from any liability for any losses, whether direct or indirect, which may arise from the use of this material, from its content and is under no obligation to update the information contained in this document. Additionally, you confirm that you understand the risks related to the financial instruments discussed in this material. Due to international regulations not all financial instruments/services may be available to all clients. You should be aware of and observe any such restrictions when considering a potential investment decision. Past performance and forecast are not a reliable indicator of future results. The information contained herein has been obtained from internal and external sources and is believed to be reliable as of the date in which this material was issued, however IBBAInt does not make any representation or warranty as to the completeness, reliability or accuracy of information obtained by third parties or public sources. Additional information relative to the financial products discussed in this material is available upon request. Itau BBA International plc registered office is 20th floor, 20 Primrose Street, London, United Kingdom, EC2A 2EW and is authorised by the Prudential Regulation Authority and regulated by the Financial Conduct Authority and the Prudential Regulation Authority (FRN 575225) Itau BBA International plc Lisbon Branch is regulated by Banco de Portugal for the conduct of business. Itau BBA International plc has representative offices in France, Germany, Spain which are authorised to conduct limited activities and the business activities conducted are regulated by Banque de France, Bundesanstalt fur Finanzdienstleistungsaufsicht (BaFin), Banco de España respectively. For any queries please contact your relationship manager; (ii) U.S.A: Itau BBA USA Securities, Inc., a FINRA/SIPC member firm, is distributing this report and accepts responsibility for the content of this report. Any US investor receiving this report and wishing to effect any transaction in any security discussed herein should do so with Itau BBA USA Securities, Inc. at 767 Fifth Avenue, 50th Floor, New York, NY 10153; (iii) Asia: This report is distributed in Hong Kong and Japan by Itaú Asia Securities Limited, which is licensed in Hong Kong by the Securities and Futures Commission for Type 1 (dealing in securities) regulated activity. Itaú Asia Securities Limited accepts all regulatory responsibility for the content of this report. In Hong Kong, any investors wishing to purchase or otherwise deal in the securities covered in this report should contact Itaú Asia Securities Limited at 29th Floor, Two IFC, 8 Finance Street Central, Hong Kong; (iv) Middle East: This report is distributed by Itau Middle East Limited. Itau Middle East Limited is regulated by the Dubai Financial Services Authority and is located at Suite 305, Level 3, Al Fattan Currency House, Dubai International Financial Centre, PO Box 482034, Dubai, United Arab Emirates. This material is intended only for Professional Clients (as defined by the DFSA Conduct of Business module) no other persons should act upon it; (v) Brazil: Itaú Corretora de Valores S.A., a subsidiary of Itaú Unibanco S.A authorized by the Central Bank of Brazil and approved by the Securities and Exchange Commission of Brazil, is distributing this report. If necessary, contact the Client Service Center: 4004-3131* (capital and metropolitan areas) or 0800-722-3131 (other locations) during business hours, from 9 a.m. to 8 p.m., Brasilia time. If you wish to re-evaluate the suggested solution, after utilizing such channels, please call Itaú s Corporate Complaints Office: 0800-570-0011 (on business days from 9 a.m. to 6 p.m., Brasilia time) or write to Caixa Postal 67.600, São Paulo-SP, CEP 03162-971.* Cost of a local call. Page 6