Resumo do Relatório

Rates 10y Dashboard: Are Rates in Turkey Due for a Reprieve?

20/07/2021
Avaliar este relatório: 1 voto. média: 5,00 out of 51 voto. média: 5,00 out of 51 voto. média: 5,00 out of 51 voto. média: 5,00 out of 51 voto. média: 5,00 out of 5 (1 votes, average: 5,00 out of 5)
You need to be a registered member to rate this.
Loading...
Currencies Emerging Markets Fixed Income Global Strategy Interest Rates USA

The US 10y yield has now dropped below 1.2%, a level it last touched in early February. While not all of the sell-off in long bonds since the start of the year has been erased, much of it has been. With US 10y yield dropping 40 bp since early June, 10y rates in EM have rallied but by much less – only 15 bp or so. The best-performing country has been Turkey.

The 10-year rate in Turkey (TRY 10y) is very high at just below 17% but it has rallied around 140 bp from the peak reached in early June. While the outperformance has been quite significant on an absolute basis, it is less so on a beta-adjusted basis, as the chart below shows, especially when factoring in the previous underperformance.

Data Source: Refinitiv; Note: EM = equally weighted index of rates in 19 countries

Nevertheless, the performance of Turkey’s rates has still been impressive:

  • Rates have rallied despite the spike in inflation, which ruled out a rate cut at the central bank’s (CBRT) monetary policy meeting last week. Headline CPI for June came in at 17.53% YoY, much higher than expected and above the May figure of 16.59%. While the volatile food and energy components contributed to the rise, even the core CPI rose 17.48% YoY. Moreover, the rise in inflation was broad-based and across almost all subcategories.
  • Even with a very high policy rate of 19%, the recent rise in inflation leaves the real policy rate barely positive.
  • The year-end inflation expectation based on the latest central bank survey rose to 15.64%, much higher than the central bank’s forecast of 12.2% in its April Inflation Report, although it is likely to be revised higher in the report to be released later this month.

Despite the recent strong performance of Turkey 10y, we see room for it to continue:

  • We expect the CBRT to start easing later this year, starting as early as September but more likely in October. Although it is likely to stay high, inflation should peak in the coming months on the back of base effect, the drop in oil prices due to the expected weakness in demand with global growth adjusting lower and the upcoming increase in supply following the OPEC+ agreement, and the currency stabilizing – albeit at a weak level – since May. Indeed, as the chart below shows, short-term inflation breakevens peaked at the end of March and, historically, they have been a good predictor of upcoming monetary policy actions.
  • The CBRT has maintained a hawkish stance, which it articulated well in the last statement: “Taking into account the high levels of inflation and inflation expectations, the current tight monetary policy stance will be maintained decisively until a significant fall in in the April Inflation Report’s forecast path is achieved.” While we find the CBRT’s hawkish bent appropriate, we still see room for easing, though limited. If the year-end inflation forecast of 15.64% in the survey holds up, then the central bank has room to ease by 150 bp by year-end, which it would likely deliver cautiously in steps of 50 bp, and still keep the real policy rate at around +2%.
  • Although still weak, the macroeconomic backdrop has improved:
    • Growth has bounced up as almost 50% of adults have received at least one dose of the vaccine and the country is on the path to achieving herd immunity well before year-end.
    • Foreign exchange reserves finally bottomed as both gross and net FX reserves rose by around USD 10bn over the past month.
    • The current account balance continues to improve on the back of favorable external demand conditions and the tight monetary policy stance, with the central bank expecting a surplus for the rest of the year on the back of the rise in exports and tourism. The drop in oil prices, if sustained, would be an added bonus.
  • The technical backdrop is also supportive:
    • Foreign ownership of Turkish local debt is below 7%. While it has risen from the lows of 4% last year, it is still close to the historical lows and far below the high of 25% in 2017.
    • Residents’ FX deposits have held steady close to a multi-year high of around USD 190bn for several months, according to Bloomberg. If they start selling some of these positions, which is likely during the ongoing Eid al-Adha holiday week, then it could feed into a virtuous cycle. To propel the selling of FX deposits, the CBRT raised reserve requirements on FX deposits earlier this month.

Data Source: Refinitiv

However, it is best to treat any position in Turkish rates as short-term in nature and idiosyncratic:

  • The volatility of rates in Turkey has been stratospheric since 2018 (see the chart below), which quite clearly indicates the inherent risk in any rates position.
  • The curve is inverted which implies negative carry, especially as one goes further out in the curve. It is, therefore, better to stick with the short end.
  • Finally, monetary policy remains unpredictable because of political interference, which makes it difficult to regard the central bank as an autonomous body. President Erdogan called for a rate cut in July or August, which now appears unlikely. He has fired three central bank governors over the past 2 years. It is therefore difficult to have any conviction that the current governor won’t suffer a similar fate. As a result, all Turkish assets, including rates, should maintain a high risk premium to account for political meddling.

Data Source: Refinitiv; Note: EM = equally weighted index of rates in 19 countries

Best Longs / Best Shorts

  • The lists of underperformers and outperformers based on our models in the attached Rates 10y Dashboard pdf stayed unchanged.
  • The list of underperformers consists of Chile 10y (CLP 10y), Peru 10y (PEN 10y), and Colombia 10y (COP 10y), while the list of outperformers stayed empty.

Best Crosses

  • The list of preferred relative-value trades based on our models changed to Receive CLP 10y vs Pay Poland 10y (PLN 10y) or China 10y (CNY 10y) or Thailand 10y (THB 10y) or Israel 10y (ILS 10y).
  • The 3m expected returns for these pairs are in the range of 7.5-8.5% (not annualized), assuming mean reversion.

 

Gautam Jain

Gautam Jain
Estrategista - Ph.D, CFA
New York, EUA

Aviso legal

DISCLAIMER: Este Relatório de Análise foi elaborado e distribuído pelo Analista, signatário unicamente para uso do destinatário original, de acordo com todas as exigências previstas na Resolução CVM nº 20 de 26 de fevereiro de 2021 e tem como objetivo fornecer informações que possam auxiliar o investidor a tomar sua própria decisão de investimento, não constituindo qualquer tipo de oferta ou solicitação de compra e/ou venda de qualquer produto. As decisões de investimentos e estratégias financeiras devem ser realizadas pelo próprio leitor, os Analistas, ou a OHMRESEARCH não se responsabilizam por elas. Os produtos apresentados neste relatório podem não ser adequados para todos os tipos de investidores. Antes de qualquer decisão de investimentos, os investidores deverão realizar o processo de suitability no agente de distribuição de sua confiança e confirmar se os produtos apresentados são indicados para o seu perfil de investidor. A rentabilidade de produtos financeiros pode apresentar variações e seu preço ou valor pode aumentar ou diminuir num curto espaço de tempo. Os desempenhos anteriores não são necessariamente indicativos de resultados futuros. A rentabilidade divulgada não é líquida de impostos. As informações presentes neste material são baseadas em simulações e os resultados reais poderão ser significativamente diferentes.
O(s) signatário(s) deste relatório declara(m) que as recomendações refletem única e exclusivamente suas análises e opiniões pessoais, que foram produzidas de forma totalmente independente e que a OHMRESEARCH não tem qualquer gerência sobre este conteúdo. As opiniões aqui expressas estão sujeitas a modificações sem aviso prévio em decorrência de alterações nas condições de mercado. O Analista responsável pelo conteúdo deste relatório e pelo cumprimento da Resolução CVM nº 20/21 está indicado acima, sendo que, caso constem a indicação de mais um analista no relatório, o responsável será o primeiro analista credenciado a ser mencionado no relatório. Os analistas cadastrados na OHMRESEARCH estão obrigados ao cumprimento de todas as regras previstas no Código de Conduta da APIMEC para o Analista de Valores Mobiliários e no Manual de Controles Internos para Elaboração e Publicação de Relatórios da OHMRESEARCH. De acordo com o art. 21 da Resolução CVM nº 20/21 caso o Analista esteja em situação que possa afetar a imparcialidade do relatório ou que configure ou possa configurar conflito de interesse, este fato deverá estar explicitado no campo “Conflitos de Interesse” deste relatório.
O conteúdo deste relatório é de propriedade única do Analista signatário e não pode ser copiado, reproduzido ou distribuído, no todo ou em parte, a terceiros, sem prévia e expressa autorização deste Analista. Todas as informações utilizadas neste documento foram redigidas com base em informações públicas, de fontes consideradas fidedignas. Embora tenham sido tomadas todas as medidas razoáveis para assegurar que as informações aqui contidas não são incertas ou equívocas no momento de sua publicação, o Analista não responde pela veracidade das informações do conteúdo.
Para maiores informações, pode-se ler a Resolução CVM nº 20/21 e o Código de Conduta da APIMEC para o Analista de Valores Mobiliários. Este relatório é destinado exclusivamente ao assinante da OHMRESEARCH que o contratou. A sua reprodução ou distribuição não autorizada, sob qualquer forma, no todo ou em parte, implicará em sanções cíveis e criminais cabíveis, incluindo a obrigação de reparação de todas as perdas e danos causados, nos termos da Lei nº 9.610/98 e de outras aplicáveis.