Yet private consumption growth slowed in 2017. First, the direct impact of new loans was largely offset by downside factors already mentioned; second, it was also attenuated by the fact that a portion of the new consumer loans were actually used for purchasing real estate assets. In China, there is often a fuzzy line between the different types of lending. This phenomenon was probably amplified in 2017 to get around the tighter restrictions on mortgage loans. Excluding this portion of loans, based on estimates of the Conference Board[14], new loans dedicated exclusively to consumer goods purchases accounted for 5% to 6% of total private consumption in 2017 – which is still high. In 2018, the contribution of consumer credit began to decline, which had an impact on certain types of household purchases (cars and other durable goods in particular).
...but the dynamics are being reversed
The tightening of prudential regulations has been extended to include consumer credit activities since 2018; as a result, growth in credit card loans outstanding slowed from 37% y/y at year-end 2017 to 16% in mid-2019. In fact, the authorities have been gradually addressing the problem of mounting credit risks tied to the rise in household debt.
So far, the quality of mortgage loan portfolios held by commercial banks is satisfactory, thanks notably to relatively conservative prudential rules. In contrast, the rapid increase in consumer loans has been accompanied by less rigorous risk management and a higher default rate (albeit still low). In 2017, the non-performing loan ratio was 0.3% for mortgage loans and 1.6% for credit card loans. Moreover, households, like corporates, have benefited from the expansion of shadow banking activities over the past decade. Bank and non-bank financial institutions have proposed new savings and financing products in order to bypass existing regulations. As a result, the loose regulatory environment has been encouraging risk taking. P2P lending platforms developed rapidly between 2014 and 2017, with the creation of more than 6600 platforms. Total loans outstanding in this category represented only a small portion of household debt (3% at the peak in early 2018), but P2P practices were often fraudulent with high default rates.
Household debt growth slowed from 24% y/y in nominal terms in Q2 2017 to 17% in Q2 2019. The deceleration has resulted from monetary tightening measures implemented between late 2016 and Q2 2018, a tight property policy maintained over the past three years[15], and a series of measures aiming to reduce P2P financing and other shadow banking activities. The vast majority of P2P platforms were brought under supervision or shut down, and P2P loans outstanding have fallen rapidly since July 2018 to reach RMB 687 billion at mid-2019, accounting for 1.3% of household debt (Chart 16 page 9).
The slowdown in household credit, which has helped contain credit risks in the financial sector, clearly constrained consumption growth in 2018 and H1 2019. Moreover, households should also begin to feel the burden of debt servicing payments.
Household debt: a new threat
Household debt servicing has increased sharply since 2016. In addition to the direct negative impact on consumption, high debt servicing makes households more sensitive to increases in interest rates and credit tightening measures, and also limits their capacity to take on more debt in the medium term.
According to the Conference Board14, household debt servicing charges (home loans + consumer loans) accounted for between 8% and 11% of disposable income in 2017, up from 6% to 8% in 2015 (and 5% to 7% in 2011). The most recent estimates by the IIF place debt servicing at 11% of disposable income in 2018, compared to 8% in 2015[16]. These ratios are considered to be moderately high, and the average financial situation of households is still satisfactory, thanks notably to a solid savings rate, with savings invested in real estate and financial assets (bank deposits and other financial assets account for more than two times household debt outstanding).
Yet the situation between households is very mixed. The share of total debt held by over-indebted households (i.e. debt exceeding 4 years of revenues) already increased from about 25% in 2010 to nearly 50% in 2016, according to IMF estimates[17]. As a share of revenues, the debt burden has worsened much faster for low-income households. Income inequality aggravates the impact of credit tightening on consumption: the poorest households, which have the highest debt burden and the lowest savings, are also the most vulnerable to income shocks: they will cutback consumption more sharply of new credit is reduced. Inversely, the easing of household lending conditions should have an even more limited impact on consumption since debt servicing is already significant, and unequally distributed between the different revenue categories.
Consequently, the favorable impact of credit on household spending tends to taper off over time and as debt builds up. According to a BIS study[18], an increase in household debt has a positive impact on private consumption and GDP growth in the short term (1-year horizon), but the impact becomes negative in the medium term, notably because households must allocate a growing share of their revenues to debt servicing. When the debt ratio exceeds the threshold of 60% of GDP, it has a significantly negative impact on consumption in the medium term. China has not reached this threshold yet, but it is approaching it rapidly (in our central scenario, it will cross the 60% threshold in 2020).
The expected adjustments in the housing market (more balanced expansion, lower house price inflation, measures to combat speculation) could lead to a growth slowdown in mortgage loans to households, which would leave a little more room for consumer credit. Even so, there is very little space for lending to boost consumption.
In the short term, the authorities could ease their household lending policy again to stimulate domestic demand and counter the effects of sluggish exports. However, even in a healthier regulatory environment, the positive impact of higher debt dynamics is likely to be limited in the short term, and it could have major negative effects on private consumption in the medium term (heavier debt servicing and higher credit risks for banks). Moreover, households now have less capacity to take on more debt. Consequently, we should not expect to see a credit-driven consumer boom in the years ahead.