S&P Global Ratings believes default rates of pools of receivables in Japanese auto loan ABS transactions we rate correlate to some extent with Japan's real GDP, consumer price index (CPI), unemployment rate, and Nikkei 225 Stock Average.

Our analysis found some correlation between the default rate and the three macroeconomic factors. Real GDP and the CPI showed negative correlation with the 12-month moving average default rate, with correlation coefficients of -0.49 and -0.51, respectively. The unemployment rate, meanwhile, had a positive correlation, with a correlation coefficient of +0.48. Borrowers of loans backing Japanese auto loan ABS are individuals, and the macroeconomic environment surrounding them materially affects their ability to repay loans. In addition, our analysis found a similar correlation for the Nikkei index, which correlates with Japan's real GDP.

Our analysis covers pools of loan receivables underlying auto loan ABS we rate, including fully repaid transactions. We took the default data of these transactions since 2003 as well as macroeconomic factors in the same period and examined their correlation to determine which macro factors affect the default rate. We also analyzed the default rate's sensitivity to changes in macroeconomic factors by measuring the volatility of the factors during the global financial crisis, when the default rate remained relatively volatile.

Correlation With Macroeconomic Factors

We calculated and analyzed correlation coefficients between the default rate of underlying assets in rated auto loan ABS and key macroeconomic factors (see table 1). We also examined correlation coefficients among macroeconomic factors (see table 2).

Table 1

Table 2

Economic trend (real GDP, stock prices)

The correlation coefficient between Japan's real GDP and the 12-month moving average default rate of the underlying loan pools was -0.49.

During the economic recovery after 2008 following the global financial crisis, Japan's real GDP rose while the default rate of underlying loans decreased. However, we saw an opposite trend during the COVID-19 pandemic, with both real GDP and the default rate falling. We attribute this to restrictions on economic activity in the period, including restrictions on people leaving home. This cut household spending and lifted cash balances. In addition, Japan's unemployment rate remained stable by global comparison, thanks to the government's employment-related subsidies. In our view, this also prevented the default rate from rising during the pandemic.

The Nikkei index is another macroeconomic factor that indicates economic trends. We observed a mild correlation between the default rate of the underlying loans and the index. Specifically, the 12-month moving average default rate and the Nikkei index had a correlation coefficient of -0.45. We believe this reflects some correlation between Japan's real GDP (one-year lead) and the Nikkei index (year on year, one-year lead) with a correlation coefficient of +0.46 (see table 2).

Chart 1

Chart 2

Unemployment

We calculated the correlation coefficient between Japan's unemployment rate and the 12-month moving average default rate to be +0.48, which we believe indicates some correlation. Because borrowers of these loans are individuals, we expect a reduction in individual income due to unemployment would significantly impact the default rate of the underlying auto loans.

Japanese auto loan lenders generally focus on a borrowers' ability to repay rather than loan to value (LTV), the ratio of the loan amount to the value of the vehicle being purchased. To confirm that a borrower's income is stable, lenders check not only their annual income but also their employment status, time in a job, and the amount of other loans, if any.

In our view, correlation between the unemployment rate and the default rate of the underlying loans points directly to the relationship between the performance of these loans and underwriting practices in the Japanese auto loan market, which focuses on the ability of borrowers to repay. In addition, Japan has a low unemployment rate by global comparison because of its strict regulations on layoffs. We consider this one of the reasons for stable default performance of the underlying assets.

Chart 3

Inflation

Japan's CPI and the 12-month moving average default rate of the underlying assets had a correlation coefficient of -0.51.

Japan's CPI has not shown much volatility over the long term. However, prices have risen basically three times in the last 20 years. First, during the economic recovery that ended just before the 2008 financial crisis; second, in 2014 when the consumption tax rose; and third, currently, due to rising raw material costs and a weaker yen. Wages have not risen much when prices were on the rise. As a result, real wages fell in each of these three occurrences, weakening the ability of borrowers to repay their loans.

The CPI negatively correlated with the default rate of the underlying loans--that is, the default rate tended to fall as prices rose. Japan's CPI showed a similar movement to real GDP and the Nikkei index, with prices rising moderately in economic recovery phases. As noted above, real GDP and the Nikkei index negatively correlate with the default rate. This explains the same results for the CPI, in our view.

Chart 4

Automobile market

The correlation coefficient between the 12-month moving average volume of car sales and the 12-month moving average default rate of the underlying loans in the rated auto loan ABS was -0.45.

Automobile sales depend on various factors, including supply-side trends such as automakers' sales strategies, raw material costs, domestic and overseas supply chain systems, and demand-side trends such as changes in population and the number of households, household finances, and consumer needs and preferences. At the same time, the auto industry is one of Japan's key industries. Car sales are closely related to economic indicators such as real GDP. Real GDP negatively correlates with the default rate, and we saw a similar trend for car sales.

We saw no correlation between the unit price of exported used cars and the 12-month moving average default rate of underlying loans. If demand for used cars is high and their prices rise, for example due to auctions, this will give borrowers in financial distress an opportunity to sell their vehicles to repay loans. This could underpin performance of the loans. There is no benchmark to track used car prices in Japan. Since some used cars are exported, we compared the default rate of the underlying loans with the unit price of exported used cars (see chart 6). Unit prices of exported used cars were stable until the end of 2020. This may contribute to the stability of underlying loan performance, but we did not find a clear correlation.

Chart 5

Chart 6

Interest rates

The yield on 10-year Japanese government bonds (JGBs) and the 12-month moving average default rate showed a significant correlation, with a correlation coefficient of +0.78.

The high correlation coefficient relates to the movement of JGBs and the default rate after 2006. The 10-year JGB yield declined moderately in line with the default rate. However, the underlying auto loans in the rated ABS transactions have mostly fixed interest rates. Therefore, we believe the interest rate environment does not have a direct impact on the default rate of the underlying loans. However, we believe if a borrower of an auto loan has another loan obligation that is larger than the auto loan, such as housing loan, and if the repayment burden of that loan increases due to an interest rate hike, this may indirectly affect the borrower's ability to repay the auto loan.

Chart 7

Other macroeconomic factors

We also analyzed the correlation between the real effective exchange rate and the default rate of the underlying auto loans. The real effective exchange rate showed a moderately high correlation with the 12-month moving average default rate, with a correlation coefficient of +0.57. The real effective exchange rate incorporates the inflation rate into the effective exchange rate, which is a measure of relative currency strength.

In the case of Japanese auto loan ABS transactions, borrowers of underlying assets are usually Japanese citizens, and all repayments are made in Japanese yen. Therefore, exchange rates do not affect the performance of underlying assets. We believe the correlation coefficient is high due to the interaction of several macroeconomic factors.

Chart 8

Sensitivity Analysis

As a result of our sensitivity analysis, a 10% fall in real GDP increased the annual default rate of the underlying loans by 0.54 of a percentage point (ppt). In addition, a 1 ppt rise in Japan's unemployment rate pushed up the default rate by 0.28 ppt.

Table 3

Looking at each auto loan ABS transaction, we found that the lower the credit quality of the asset pool the higher the volatility of the default rate under stressed conditions. We extracted transactions of medium credit quality, with a cumulative default rate (the ratio of the total amount defaulted since closing to the initial outstanding loan balance) greater than 1% but less than or equal to 3%. We analyzed the sensitivity of the default rate to macroeconomic indicators, incorporating changes in the default rate and changes in macroeconomic factors during the financial crisis.

In this sensitivity analysis, we used changes in the following macroeconomic indicators during the financial crisis as a reference point.

Table 4

In the sensitivity analysis, we calculated the impact of a change in a single macroeconomic factor on the default rate of underlying assets. In reality, several factors may change simultaneously in times of economic slowdown. This would cause a greater impact on the default rate.

Notes

In this report, figures include rating actions by S&P Global Ratings and S&P Global SF Japan Inc. (SPSF). SPSF is a registered credit rating agency under Japan's Financial Instruments and Exchange Act (FIEA) but is not registered as a Nationally Recognized Statistical Rating Organization (NRSRO) under U.S. Laws. Therefore, the credit ratings assigned by SPSF are Registered Credit Ratings under FIEA but are not Credit Ratings issued by an NRSRO under U.S. laws. The data in this study includes both public ratings and confidential ratings.

Related Criteria

• Global Auto ABS Methodology And Assumptions, March 31, 2022

Related Research

• Japan Auto Loan ABS Performance Watch: Default Dip Over, Aug. 2, 2023
• Default, Transition, and Recovery: 2022 Annual Japanese Structured Finance Default And Rating Transition Study, March 27, 2023
• Japan Structured Finance Outlook: Economic Growth To Shield Performance, Jan. 12, 2023