巴黎银行-新兴市场-投资策略-新兴市场策略：从新兴市场资产价格推断信贷周期-20190319-21页_1mb

EM Strategy Summary: Inferring Credit Cycles from EM Asset Prices

Core Content

This document presents a new quantitative approach to infer the phase of economic cycles in emerging markets (EM) using asset prices rather than traditional macroeconomic indicators. The methodology is based on the idea that market participants may react faster to economic changes than GDP data, which often lags by a quarter. By analyzing financial asset performance, the model aims to provide a more timely and accurate assessment of the current economic cycle stage.

Key Points

• Objective: Extract the implied stage of the economic cycle using EM asset prices.

• Methodology:

  • A normalization process was applied to make all sub-indices comparable within the range [-1, +1].
  • The Hodrick-Prescott (HP) filter was used to isolate the trend component from noise and cyclical factors.
  • A sinusoidal function was fitted to the filtered index to identify the cycle stage and its duration.
  • The model is updated monthly and does not rely on regressing variables or historical assumptions.

• Cycle Stages:

  • The cycle is divided into four stages: early, mid, late, and bottom out.
  • Each stage corresponds to a 90-degree interval in the sinusoidal model.

Main Findings

Key Insights

• Brazil is in the early stage of the cycle, showing signs of recovery.
• South Africa, Indonesia, USA, and Turkey are in the late stage of the cycle.
• China has entered a bottom out cycle, expected to last 3-7 months.
• Poland is entering a bottom out cycle, likely to last about 9 months.
• Mexico is in a similar early stage as Brazil but is still exiting a bottom out phase.
• The model has been applied to multiple EM countries, including Brazil, Mexico, South Africa, the USA, China, Indonesia, South Korea, Turkey, and Poland.
• The results are consistent with the BNPP Global Outlook, indicating a leading indicator for economic activity.

Implications

• The model provides a more dynamic and market-driven view of the economic cycle.
• It allows for a better understanding of market sentiment and expectations.
• The results can be used to inform investment strategies and market recommendations.

Technical Details

• Normalization: A 30-month rolling window was used to normalize the data.
• HP Filter: Adjusted to ensure smooth trend analysis without significant delay.
• Sinusoidal Fit: Used to determine the angle and phase of the cycle, with a constraint to prevent abrupt changes in the modelled angle.
• Model Constraints:
  • No more than 10% of a full 360-degree cycle change per observation.
  • The model is adaptive, using a rolling window approach to update periodically.

Conclusion

This EM strategy report demonstrates how market prices can be used to infer the phase of economic cycles, offering a more timely and accurate alternative to traditional macroeconomic indicators. The results provide insights into the current and future economic activity in several EM countries, supporting investment decisions and market analysis.