China: The Impact of Reform on Growth

I got a lot of feedback from my January 5 blog entry because of my argument that the implementation of the reforms proposed in the Third Plenum all but guarantees that growth rates in China will slow down. For that reason I thought it might make sense for me to explain a little more carefully why I think this must happen, and why I think that we can almost judge how successfully the reforms are implemented by how quickly growth slows.

The first point to recognize is that when a country’s growth has been driven by wasteful investment, GDP growth will exceed real economic wealth creation, productivity will be overstated, and debt will rise faster than debt servicing capacity. Why? Because in China we record growth in terms of the cost of inputs, not in terms of the value of the outputs, and so if the cost of inputs exceeds the value of outputs, we will overstate the real value of economic activity.

Now of course this happens elsewhere too, but there is an automatic mechanism for writing down this excess. This mechanism is usually the recognition of bad debt. Companies that invest poorly go bankrupt, and the value of their loans is written off. This writing off of bad loans shows up as a correction to the overstatement of growth and productivity.

In a system in which bad debt isn’t written down, the losses are simply hidden and rolled over. Of course after many years they are effectively written down, but this happens indirectly. In order to service the loans there is an explicit or hidden transfer from some other part of the economy to cover the full extent of the losses, so that future growth is reduced by the amount of the transfer.

Over long periods of time, in other words, real economic value and recorded economic value is the same, but over shorter periods of time they can differ enormously. If a country fails to record bad debt, its growth today will be overstated by that amount, but its future growth over the longer term will be understated by the same amount.

I think most of us agree that a significant share of the loans in the Chinese banking system would be considered, from an economic point of view, as bad loans. They were made to support investments the true economic value of whose outputs are less than the cost of the inputs. Because many of these loans are implicitly guaranteed, it may make prefect legal sense for the banks to treat these as performing, but this does not change the fact that the loans are uneconomic.

I would argue that China’s GDP is overstated by the value of these hidden losses, and over time these losses will be worked out. As long as bad loans (as I am defining them here) are increasing, it is pretty safe to assume that the gap between China’s real economic output and its recorded output is also increasing. This has been the problem with China’s growth of the last several years.

Beijing’s response is the economic reforms proposed during the Third Plenum, aimed at unlocking greater productivity potential in the Chinese economy and returning the country to a sustainable growth path. They will do this by improving the capital allocation process, so that capital will be diverted from SOEs, real estate developers, local governments and other inefficient users of capital, to SMEs, the agricultural sector, and more efficient users of capital. They will also eliminate constraints that prevent more productive use of resources, including weak legal enforcement of business claims, better protection of managerial and technological innovation, educational improvements, and so on.

The implementation of these reforms is not certain. There is likely to be tremendous political opposition for all the reasons I have discussed in the past three newsletters. But, even assuming they are forcefully implemented, the higher productivity resulting from the reforms will not lead to higher reported GDP growth. This is one of the great recent myths that, to me, make no sense at all. The higher productivity will not even allow China’s economy to continue growing at current rates. On the contrary, successful implementation of the reforms will cause GDP growth rates to drop sharply.

There are at least four reasons to expect healthier but slower GDP growth over the rest of this decade if the reforms are implemented.

1. Leverage boosts growth and deleverage reduces it. By now nearly everyone understands that China is over-reliant on credit to generate growth. Much new borrowing is needed simply to prevent borrowers from defaulting on existing loans, so that new lending can be divided into two buckets.

One bucket consists of loans made to roll over the debt of borrowers who do not generate sufficient cashflow from the investments that their original loans funded. The loans in this bucket, of course, do not create additional economic activity, but as debt rises, financial distress costs rise with them (most financial distress costs, as is well understood in corporate finance theory, are a consequence of the way rising debt changes the incentive structures of the various stakeholders and so distorts their behavior in non-economic ways). Of course any disruption in lending would cause a surge in defaults.

The second bucket consists of loans that fund new expenditures. These expenditures, of course, generate economic activity, but if they fund consumption, or if they fund investments the value of whose output is less than the cost of the inputs, they incur additional losses that must ultimately be rolled over by loans that belong in the first bucket. Any reduction in loan growth, in other words, is positive in the long term for Chinese wealth creation, but in the short term will either force the recognition of earlier losses or will reduce economic activity.

Beijing has attempted since 2009-10 to rein in credit growth, but each time credit growth has decelerated, GDP growth rates – as we would expect – dropped so sharply that Beijing was quickly forced to relent. Because growth is more dependent than ever on credit, as Beijing finally acts to rein in credit growth decisively, GDP growth will drop sharply.

2. Hidden transfers will be reduced. As I have discussed many times the investment-led model encourages investment by transfers – hidden or explicit – from the household sector to subsidize investment. In the Japanese version of this model, which very broadly is the version China and the Asian Tigers pursued, the main form of these transfers is the undervalued currency, low wage growth (relative to productivity growth) and, most of all, financial repression.

Because these transfers no longer create net value on the investment side (China overinvests in infrastructure and has excess capacity in a broad range of manufacturing sectors), and the extent of the transfers are at the heart of China’s very low consumption level, the proposed reforms will act to reverse the mechanisms that goosed growth by transferring resources from the household sector to subsidize manufacturing, infrastructure building, and real estate development. These mechanisms put downward pressure on household income even as they subsidized manufacturing and investment and led directly both to higher growth rates and to the investment and consumption imbalances from which China suffers and which it plans to reverse.

It should be clear that as Beijing reverses policies that once acted to increase growth, the result must be slower growth. It is hard to estimate the amount by which growth will decline once all the transfers are eliminated, but when one considers that the total amount of transfers to SOE’s during this century may exceed the aggregate profitability of the SOE sector by as much as five to ten times, it is pretty clear that their impact is likely to be substantial.

3. Excess capacity will be resolved. Beijing recognizes that cheap credit and limited accountability have created excess capacity in industry and real estate. Why build so much excess capacity? Local governments have supported this build-up of capacity to boost growth and, with it, revenues and local employment, and because capital was essentially free (its real cost may have even been negative for much of this century) and because most projects are implicitly or explicitly guaranteed by local and central governments, there seemed to be no cost, and plenty of benefit, simply to pile on capacity.

As Beijing acts to wring out excess capacity, we will inevitably see a reversal of the earlier growth impact. If building capacity generates economic activity (and it must have, or else why do it), closing down excess capacity must become a drag on growth.

4. Losses will be recognized. As I discuss above, because many years of overinvestment have left a large amount of unrecognized bad debt on bank balance sheets, China’s GDP growth has been overstated by the amount of the unrecognized losses. Over the next decade as Beijing cleans up its financial system, this bad debt will either be explicitly recognized or, more likely, implicitly written off over the remaining life of the loan. Either way, as the losses are recognized, growth over the next several years will automatically be understated by the amount previously overstated.

These reforms, and others – like attempts to protect the environment – will ensure that even as China’s real economic productivity improves, its GDP growth numbers will drop as the reforms are implemented. For now most commentators argue that by increasing productivity, real reform will ensure a soft landing of GDP growth rates of 7-8 percent during the rest of President Xi Jinping’s administration. A growing minority worries, however, that rapidly rising debt will force China into a hard landing.

GDP growth is an inverse proxy for reform

Although rising debt increases the probability of a hard landing, for now I expect neither outcome. More likely, I believe, is a “long landing”, during which growth rates will drop by roughly one to two percentage points every year for the rest of this decade. Implementing reforms will protect China from a hard landing. It will however force much lower, albeit healthier, growth rates.

In order to understand China’s growth prospects I think we must recognize that while a growth model can deliver healthy growth for many years, this growth can itself transform conditions to the point where the model is no longer able to deliver. At that point the economy must adjust to a new, more appropriate growth model.

The Chinese growth model is a version – in probably its most extreme form – of the investment-led growth model described by Alexander Gershenkron fifty years ago. To simplify tremendously, growth in “backward” economies is supported by policies that subsidize investment while suppressing consumption (usually by constraining household income growth). These “backward” economies are ones in which the level of capital stock is much lower than the country’s social and institutional ability to absorb investment efficiently.

Early on, many years of high investment allowed China to catch up. Once it did, however, continuing to invest in the same way and to the same degree was no longer wealth enhancing. At this point the economy needed institutional and social reforms to continue growing. The political logic of the system, however, forced, as it almost always does, continued high investment growth and, with it, increasing investment misallocation.

With this almost by definition debt began to rise faster than debt servicing capacity. This, clearly, was unsustainable, but of course it can go on for many years. It was as long ago as 2007 that former Premier Wen described the Chinese economy as “unsteady, unbalanced, uncoordinated and unsustainable”, but it proved politically very difficult for Beijing to implement the reforms his advisors suggested, and so the distortions associated with the growth model continued.

Debt surged even as the consumption imbalance deteriorated until late 2011. We have only seen in 2012-13 the beginning of any partial rebalancing, although during this time there has been at best only a deceleration in the growth rate of credit.

And yet the minimal amount of rebalancing that has occurred in the past three years has already lopped three percentage points off China’s GDP growth rate, just as we predicted. China still has a long way to go to rebalance its economy. By my calculations consumption growth must outpace GDP growth by 3-4 percentage points every year for at least a decade just to allow China to raise the household consumption share of GDP to a still-low 50 percent.

The proposed reforms will certainly unleash greater productivity, but they will also eliminate the very mechanisms that had previously turbo-charged economic activity and which showed up in the form of higher reported GDP growth rates. They will cause a sharp deceleration in economic activity even though growth will be more productive than in the past. The fact that growth rates have dropped by almost a third even before the reforms were implemented suggests to me just how much further they must drop.

About Michael Pettis 166 Articles

Affiliation: Peking University

Michael Pettis is a professor at Peking University's Guanghua School of Management, where he specializes in Chinese financial markets. He has also taught, from 2002 to 2004, at Tsinghua University’s School of Economics and Management and, from 1992 to 2001, at Columbia University’s Graduate School of Business.

Pettis has worked on Wall Street in trading, capital markets, and corporate finance since 1987, when he joined the Sovereign Debt trading team at Manufacturers Hanover (now JP Morgan). Most recently, from 1996 to 2001, Pettis worked at Bear Stearns, where he was Managing Director-Principal heading the Latin American Capital Markets and the Liability Management groups.

Visit: China Financial Markets

1 Comment on China: The Impact of Reform on Growth

  1. China’s reform-oriented leadership knows it has to fight corruption, but any anti-corruption campaign risks pushing further capital flight from Party members who fear prosecution. China’s plan for Foreign Trade Zones (FTZs) is a good way to attract foreign banks.

Leave a Reply

Your email address will not be published.