Can You Predict Housing Prices in a Month by Looking at the Consumer Sentiment Index? [BOK Focus]

The Housing Price Outlook Consumer Sentiment Index (CSI), published monthly by the Bank of Korea, has been found to have a high correlation with nationwide apartment prices one month later. This month, the Housing Price Outlook CSI reached its highest level in 2 years and 8 months, raising expectations that nationwide apartment prices will likely rise next month as well.

On the 29th, the Bank of Korea analyzed data from January 2013 to June 2024 and found that the Housing Price Outlook CSI showed a high correlation coefficient of 0.869 with the nationwide apartment sales price growth rate one month after the survey date. The correlation coefficient two months later was also high at 0.802.

The correlation coefficient is an indicator that shows the relationship between two values, ranging between -1 and 1. The closer the absolute value is to 1, the stronger the relationship between the two values. Values close to 0 indicate no relationship. Generally, a correlation coefficient above 0.5 signifies a strong correlation.

A Bank of Korea official explained, "When the Housing Price Outlook CSI rises compared to the previous month, the likelihood of actual apartment sales prices increasing 1 to 2 months later also grows. Essentially, the results from the consumer sentiment index can be seen as reflecting housing prices 1 to 2 months ahead."

The Housing Price Outlook CSI announced by the Bank of Korea this month was 115, up 7 points from the previous month. This is the highest level in 2 years and 8 months since recording 116 in November 2021. With the Housing Price Outlook CSI rising, the possibility of nationwide housing prices increasing 1 to 2 months later has also increased.

The Housing Price Outlook CSI is one of the individual indices that make up the CSI and asks respondents about their outlook on housing prices one year from now compared to the present. A value above 100 means that the number of households expecting housing prices to rise exceeds those expecting prices to fall, while a value below 100 indicates the opposite.

The Housing Price Outlook CSI also showed a high correlation with housing prices during the COVID-19 period. From January 2013 to August 2020, the correlation coefficient with housing prices was 0.77, still indicating a strong relationship. A Bank of Korea official noted, "At that time, it was a period of high volatility with fluctuations in oil prices after the end of COVID-19. Generally, correlation coefficients tend to increase during periods of significant domestic and international volatility."

The recent increase in correlation is attributed to greater volatility in consumer responses. The Housing Price Outlook CSI recorded its all-time highest and lowest values between 2020 and 2023. The highest ever was 132 in December 2020, and the lowest was 61 in November 2022. As housing prices rose and fell sharply during that period, consumer response volatility increased, leading to a stronger correlation between the consumer sentiment index and housing prices than in the past.

[Image source=Yonhap News]

원본보기 아이콘

The Housing Price Outlook CSI was found to have a strong correlation with housing prices from the present up to two months ahead, rather than with prices one year later. The correlation coefficient with nationwide apartment sales prices six months after the survey was 0.43, showing a significant drop in correlation. The correlation coefficient one year later was 0.069, indicating almost no correlation.

A Bank of Korea official said, "Although the Housing Price Outlook CSI asks about housing prices one year ahead, in practice, it tends to reflect housing prices from the present up to two months ahead more. When current housing prices are rising, respondents tend to answer that prices will rise."

© The Asia Business Daily(www.asiae.co.kr). All rights reserved.