Pricing Capital Right Won’t Get Capital Allocation Right in China

As has been tradition, the International Monetary Fund (IMF) released in early January its annual overview of the Chinese economy, known as the Article IV. But this year, the accompanying “Select Issues” report focused more on China’s interest rate reforms.

While there’s no doubt that China’s interest rate is far from market based, the IMF’s conclusion that simply liberalizing interest rates will ameliorate China’s capital allocation problem needs something of a rethink.

It is true, as the IMF report points out, that the bank lending rate is still heavily influenced by the People’s Bank of China. But central bank intervention is not the main barrier to reform. In reality, interest rate liberalization will be less effective in improving capital allocation and perhaps even backfire if problems with the two largest borrowers, state-owned enterprises (SOEs) and property firms, aren’t resolved first. Together they account for more than 70% of all corporate borrowing.

When it comes to SOEs, banks have a structural incentive to continue lending to them, even at below “market” rates. That’s because as I have explained previously, relative to bond investors, banks are likely to get their money back when an SOE defaults. This works because of the corporatist relationship between local governments and local banks. In order to preserve its relationship with the bank, when an SOE in its jurisdiction defaults, the local government will find ways to compensate banks.

As a result, banks can still make a profit on low-interest loans to an SOE, which means what the banks perceive as fairly priced lending is different from what the market price may be. In other words, so long as local governments act as implicit guarantees on SOE defaults, banks are more or less insensitive to the price of risk. So even if interest rates were higher as a result of liberalization, banks are still unlikely to lend to SOEs at those higher rates.

Knowing the banks’ preference for lending to entities such as SOEs and local government financing vehicles, Beijing simply sets a quantitative limit on how much lending can go toward those entities. This is capital allocation by fiat. It may be less efficient, but it is probably more effective at this point. If banks become freer to make lending decisions, one of the IMF’s prescriptions, it’s likely that more lending will go to SOEs, thereby worsening the capital allocation problem.

When it comes to property, interest rate liberalization would likely have the unintended consequence of exacerbating property bubbles. As a result of rising risk in the property sector, the Chinese government has clamped down on how much banks can lend to property developers. This of course means that borrowing has become more expensive for property firms, leading them to seek financing overseas. It is not uncommon for property developers to pay nearly 10% interest on their loans. To put this in perspective, currently less than 8% of total bank lending charges an interest rate above 9%.

What this suggests is that property developers are basically willing and able to pay high interest rates to get financing. So if banks are then allowed to charge higher interest rates, property lending will increase significantly from the current level. Given that Beijing has already identified property lending as one of the biggest financial risks, liberalizing interest rates at this time will actually make more capital flow to property, an undesirable outcome.

The bottom line is that interest rate reform needs to move in sync with solving the SOE and property lending problems, which at their heart are complex political economy challenges. Reforms in both areas have been difficult and have progressed slowly over the past few years. But without meaningful changes in these areas, interest rate liberalization will not be able to achieve the intended result of better capital allocation.

Houze Song is a research fellow at MacroPolo. You can find his work on the economy, local finance, and other topics here.