One of the more disturbing economic trends in China since the global financial crisis has been persistent elevated credit growth. The economic stimulus pushed through starting in late 2008 helped the country avert a recession. In terms of speed and scale, the stimulus was well-designed. The deficiencies lay in the way the stimulus was financed. Rather than directly using the central government’s balance sheet, the stimulus was instead financed largely through a big increase in credit to the economy.
The troubling consequence of this decision is credit has yet to return to normal levels. The total amount of bank credit relative to GDP and the total amount of credit to GDP are shown in the graph below. The dramatic increase in 2009 has not been followed by a gradual deleveraging. This is particularly true for total credit, which has grown far more rapidly than bank credit. Much of this is due to the growth of shadow banking, driven by the growth of entrusted loans, trust loans, and corporate bonds.
The literature on safe levels of credit to GDP is mixed. Countries have suffered financial crisis with lower levels of credit to GDP than China and there are countries with higher levels that have stable financial systems. China has a high level of credit to GDP compared to countries at its level of per capita income. This usually is a warning sign that there may be excessive levels of borrowing. There are, however, some mitigating factors at work. Countries with bank-dominated financial systems and underdeveloped equity markets will tend to have high levels of credit relative to GDP. Additionally, countries with high savings rates and large broad money supplies have elevated credit to GDP ratios, but typically are not dependent on foreign borrowing and have large deposit bases. Looking back over China’s development, the credit to GDP ratio has persistently been high, close to 100 percent as early as 1993.
Given that it’s difficult to make a firm judgment looking only at the stock of credit, perhaps there is more to be gained by examining the flow of credit. Here the regulations from the Basel III process help shed some light on measuring excess growth. In a 2010 report, the Basel Committee confirms that periods of excess credit growth are often associated to the buildup of systemic financial risk. The Basel Committee suggests authorities create countercyclical capital buffers during financial boom periods. To identify such booms, there must be a measurement to determine excess credit growth. After sorting through a variety of different indicators, the Basel Committee recommends using credit to GDP growth in excess of the historical trend. The historical trend is measured by looking at quarterly credit to GDP over a long period, using the Hodrick-Prescott (HP) filter, rather than a simple average, to account for structural breaks in time series.
The graph below shows growth of total credit relative to GDP, the trend line given by the HP filters, and the gap between the actual amount and the trend. A long period of below-trend credit growth during 2005 and 2008 has shifted to above-trend growth in 2009. This is relatively unsurprising given what we know about credit growth in the post-crisis period, which has grown by almost 50 percent of GDP.
What’s more interesting is the divergence between bank credit growth and total credit growth relative to previous trends. Computing the same deviation from the trend using only bank credit rather than total credit reveals that almost all the excess credit growth has come through non-loan credit extension. From the beginning of 2009 through the first half of 2013 the excess growth of bank credit was on average 2 percent, worrying but not yet at a critical level. In comparison, the excess growth of total credit was more than three times higher, 7 percent on average per quarter.
While advising authorities to adjust metrics according to national conditions, the Basel Committee identifies a rule of thumb for excess credit growth. When the credit-to-gdp ratio is 2 percentage points or more above trend, banks should initiate a buildup of the countercyclical capital buffer. When credit-to-gdp ratio is in excess of 10 percentage points of the trend, the capital buffer should reach its maximum level. The suggested size of the capital buffer is 2.5 percent of risk-weighted assets, but could be more depending on circumstances. Having a capital buffer has two benefits: during financial booms it should slow the pace of lending and during times of financial instability it should help maintain the flow of credit to the real economy.
This analysis implies that China should be actively creating such a capital buffer in anticipation of future financial risks, especially given that the growth rate over the past year has been well in excess of 10 percent. In addition to taking measures to reduce the growth of credit in China, regulators also need to prepare for the financial risks that have already materialized during the post-crisis credit boom. China has committed to adopting the countercyclical capital buffer as part of its adherence to Basel III guidelines. Chinese banks are currently above their regulatory capital requirements, but they will need to add even more to create a sufficient countercyclical capital buffer.