Description

Summary

China Evergrande is the most indebted and most vulnerable of the major Chinese property developers. Hawkeye901 wrote about the the stock in 2017, but there are a few reasons why a short makes even more sense today. After speculation in the property market reached a crescendo in the first half of 2018, there are signs that lower tier property markets are starting to crack. The government pulling back the cash subsidies that led to the speculative euphoria that developed since they were introduced in 2015. Discounts are now rising, sell-through rates are dropping and the value sales deposits on Evergrande's balance sheet have started to shrink. Evergrande is the most indebted of the major Chinese developers and is highly dependent on foreign currency borrowing. If the property market or the Renminbi shows signs of weakening, Evergrande should be one of the major casualties.

Industry backdrop

Sales volumes of residential property have grown immensely from the liberalization of the property market in 1997. China’s floor space under construction has risen by a multiple of 10x since the late 1990s. Prior to this date, housing was provided by the state. In 1997, workers were given the option to buy long-term leaseholds from work units at highly subsidized prices. The reform created significant household equity in a snap, and a building wealth effect as property prices rose in line with rapidly growing incomes after China’s entry into WTO, which unleashed an export boom.

Urbanization has not been a major driver of China’s property boom. The yearly change in China’s urban population has stayed around 20 million every year since 1996, so it doesn’t account for the dramatic rise in construction. The major driver behind the early days of China’s construction boom was more driven by a wealth effect and rapidly rising incomes in fast-growing economy. The newly urbanized population have never been the typical buyer of commodity housing. Migrant workers tend to stay in company-provided dorms and lack the Hukou needed to enjoy government services in urban areas. They also haven’t had the means to buy property. With salaries of around CNY 5,000 / month a migrant worker would not be able to buy the typical apartment now costing around CNY 1.5 million, as it would imply a price / income ratios of 25x – much above the global norm of 4-6x.

The second phase of the development of China’s property market started in 2008, when the Communist party responded to weakness in manufacturing centers across the country by a CNY 4 trillion stimulus program, pulling forward the construction of infrastructure and doubling lending in a single year. The broadest measure of credit has now risen about 100 percentage points to GDP since the start of the 2008 stimulus program.

The credit boom that was unleashed led to an increase in housing construction and speculation as rising prices led to a wealth effect. Despite rising prices, excess inventories started to build in lower-tier cities as construction activity exceeded underlying demand. Instead of letting the housing market adjust on by itself, the leadership tackled the problem by so called supply-side reform, the first step being to reduce inventories to manageable levels.

The way that the government managed to achieve lower inventory levels was by encouraging local governments to provide generous cash subsidies to the shantytown residents that were displaced whenever new housing was being built on land already occupied by someone else. The funding came from China Development Bank, who in turn borrowed via the discount window from the PBOC. Prior to 2014, displaced residents would receive new housing units but no extra cash. From 2014 onwards the percentage of displaced residents who received cash compensation for shantytown redevelopment projects rose from 9% to 49% in 2016. Researcher Zhang Dawei of Centaline Property has said that:

“Over the past several years housing markets in third and fourth-tier cities had come to heavily depend upon monetary compensation provided for shantytown redevelopment.”

There is zero doubt in my mind that these cash subsidies – a type of helicopter money – was a significant driver behind the property boom that lower tier markets have experienced since 2014.

A certain Mrs Li was quoted in a Bloomberg article saying:

“This saved me 10 years of hard work… Where we lived before there were muddy roads outside with chickens and ducks running around. It was so dirty.”

This Mrs Li was compensated for her old self-built home with three apartments and 300,000 yuan in cash – a massive amount of money for a small-town resident. The cash compensation used in shantytown redevelopment was essentially fiscal stimulus to prop up the property sector and enable excess inventories to be digested. It is not clear what led the government to take such drastic measures. It might have been related to the appointment of Vice President Liu He as an economic tzar in charge of “supply-side reform” – and his drive to to deal with the oversupply in many sectors of the economy. It might have been related to an effort by Xi Jinping and his cronies to win political favors from local government officials in the run-up to the 19th National Congress in 2017, when the Politburo and Central Committee were reshuffled.

The floor space sold in lower tier cities more or less doubled in the two years following the introduction of cash subsidies. Transaction volumes might have increased in top-tier cities as well, had it not been for the home purchase restrictions imposed in cities such as Shenzhen and Beijing since 2016.

The property market has now reached a point where it is mature at best. The potential for Chinese residential property development seems limited:

• China’s average space per person was 6 sqm in 1985. Today, China’s per capita housing stock is 37sqm vs 35sqm in Japan and 16sqm in Vietnam.
• China’s sqm built per capita of (1.15) is higher than East Asian peers such as Korea (0.88), Taiwan (0.48) and the United States (0.45).
• China’s real estate investment / GDP is around 15% vs 6% for US and 12% for Spain prior to the Great Financial Crisis in 2006
• Only 31% of new homebuyers do not currently owned an apartment vs 70% in 2008
• 70% of Chinese millennials own property vs 35% in the US, 31% in the UK
• The median age of urban housing in China is < 10 years. When Japan's housing bubble burst in 1989, the average age of a housing unit was 12 years. In the US the median age of a house is 37 years.
• It probably doesn’t make sense to buy property in China, even if you’re unable to repatriate your capital out of the country.

In the mid-2000s when nominal income growth was 15-20% and bank deposits yielded 2-3%, while rental yields were over 5%, a mortgage costing 6% essentially paid for itself while rents grew 15pp faster than the yield on bank deposits. The odds heavily favored buying over renting.

Today, nominal wage growth in a steady state is likely to end up closer to 7%. Deposits give you about 1%, a difference of 6pp. Top-tier city rental yields are 1.5% vs mortgage rates of 5.5%, giving you negative carry of 4pp. Add maintenance expenses and fast depreciation from the poor quality of construction, you may not be getting much upside from buying over renting anymore. Especially not if you consider the considerable risk of rental yields reverting to historical levels.

Rental yields in first-tier cities are now about 1.5%, lower than during any other property bubble in history that I’m aware of. Even during the South Sea Bubble in London in 1720, rental yields never dipped below 2.0%. The upside from these levels should be limited. Lower tier city rental yields are not quite as low, but still far lower than international benchmarks.

The value of the housing stock as a % of disposable income, puts China in line with previous bubbles at 500%, though lower than Ireland in 2006 at 620% and Japan in 1990 at 650%.

Affordability ratios in most major cities are now higher than 10x, compared to historical levels of 3-4.5x for the United States. Tier 1 cities have ratios in excess of 20x, higher than any other cities globally.

According to Soufun, the median price per square foot in the top 100 cities are now 38% higher than the US, where per capita income is more than 700% higher than in China. These prices seem to be out of whack with international norms, especially considering that construction costs are roughly 6x higher in the US than in China. Housing is a commodity industry, yet there doesn’t seem to be any amount of supply that can satisfy Chinese demand.

China’s household debt to GDP at 50% is lower than developed market peers, such as the level of the United States in 2005 at 100%. But China’s household consumption / GDP is only 35% vs 60% in most developed economies. Mortgage rates are also higher at 5.5% than most other markets. The result is that China’s households now spent 16% of their total income on interest payments vs 10% in the United States. The upside in household borrowing isn’t as high as sell-side analysts would have you believe.

The property market has been overvalued and seemingly defied economic reality for many years. So why should we expect the situation to change?

1. Xi Jinping has cemented his power, and a 2018 change to the constitution allows him to remain President for the remainder of his life. The 2020 social credit system will allow the Communist Party to exercise unprecedented control over the population, quelling any protest that might otherwise come from a reallocation of resources within the economy. Now that Xi has consolidated his power, reform is finally possible.

2. Since 2017, reformer Guo Shuqing was put in charge of the banking regulator to clean up the shadow banking sector. His job is to alleviate the imbalances that exist in terms of maturity mismatches, under-provisioning of credit losses and increase transparency. While these efforts may have be delayed somewhat due the ongoing trade war, total shadow banking assets are still shrinking. This deleveraging is particularly damaging to private sector borrowers such as non-SOE developers, who might have to sell land assets or completed housing units in order to repay their loans.

3. Excess inventories have largely been digested, with months of inventories back to 2008 levels. Since inventories have now been digested, it would be foolish to extrapolate the last few years of higher-than-average contract sales to continue far into the future.

4. The central government’s support of cash compensation for shantytown redevelopment is being scaled back. In July 2018, the housing ministry said it would restrict cash subsidies in cities with hot property markets. In October 2018, the housing ministry said it would scrap cash subsidies in cities with either hot property markets or low inventories. In Mid-October 2018, the government said it would accelerate shantytown redevelopment, but curbing the cash handouts that accompanied it. Since 2017, President Xi Jinping has said many times that housing should be for living in, not for speculation. The shantytown redevelopment program will be finished in 2020, with the ultimate aim of building new homes for 100 million people who used to live in shantytowns and urban villages.

5. The speculative euphoria that has built up since 2016 reached a crescendo in early 2018, when roughly 50% of all property purchased in China was for investment rather than actual use. Roughly 2/3 of buyers in the first quarter of 2018 already owned one or more homes.

6. A consumer survey by China Reality Research in August 2018 shows that 30% regarded property as the best investment option over the next 12 months, compared to 20% for stocks during at the height of the 2015 stock market bubble and 35% for wealth management products in mid-2014 at the height of their popularity.

7. According to another survey from China Reality Research, discounts have started to increase again. There are plenty of anecdotes on WeChat of sales offices of property developers being smashed after developers dropped prices to counteract a fall in demand. According to CLSA, the nationwide sell-through rate in September dropped to 75%, the lowest level since 2015.

8. The land purchased by developers hit an all-time high in the last twelve months, implying a coming supply shock. While new the YTD increase in new residential starts may be due to developers accelerating pre-sales to make up for a lack of other funding sources, the higher land purchases certainly sets the market up for an unhealthy level of supply.

9. Renewed pressure on the Renminbi may reduce the government’s willingness to engage in further fiscal stimulus to support the housing market. China’s real effective exchange rate implies that the currency may be 20-30% overvalued, even before further fiscal stimulus. The trade war, the low level of reserves in relation to the broad money supply (<10%) and persistent capital outflows makes the scope for significant fiscal stimulus more difficult. China’s current account is on track to become negative in the next 12-24 months. A current account deficit would quickly create a funding problem. The only near-term options are restrictions on outbound tourism, even-tighter capital controls, or lower interest rates, which would have the effect of devaluing the yuan. There are signs of a dollar shortage in the Chinese banking system, suggesting an oversupply of Renminbi vs US Dollar. Chinese developers have been the most active issuers of foreign currency bonds. A devaluation of the Renminbi would put further pressure on these borrowers.

10. New regulation such as pre-sale restrictions and property taxes – if implemented – would put further pressures on developers. Such reforms are still in early stages of implementation.

Of course, with China being a planned economy, it is difficult to say whether the leadership wants the construction / GDP ratio to increase further or whether it wants market forces to determine supply and demand. But what we do know is that corporate and household debt levels are already at close to early-1990s Japan and 2006 Spain. So demand would have to come from government spending, and such spending would inevitably be bearish for the Chinese currency and prices of Chinese assets denominated in foreign currencies.

China Evergrande

Guangzhou-based Evergrande is one of China’s largest property developers with contract sales of around CNY 600 billion per year (close to US$100 billion), evenly distributed across 223 cities across China. The company expanded aggressively in 2013-2016 through a debt-fueled landbanking strategy that enabled it to grow multi-fold within the scope of a few years. The company now has a land reserve of 305 million square meters and sells roughly 60 million square meters worth of housing each year, equivalent to more than half a million apartments, housing a population of 1.5 million people. The company’s average selling price of RMB 10,000 ranks somewhat lower than peers, reflecting the company’s focus on lower tier cities. Anecdotally, Evergrande seems to be just as exposed to recent years of speculative fever as any other developer. J Capital’s Anne-Stevenson Yang retold in one of her notes what an Evergrande salesman had said to her during a site visit:

“We do not recommend living here. We build the units small to be convenient for speculation”

Almost all Evergrande’s revenues come from property development, with the remaining 3-4% from rental income, property management services and other business. A sub-segment of Evergrande’s real estate arm is Evergrande Tourism Group, building “tourism-related” property projects such as the Hainan Ocean Flower Island, a group of artificial islands meant to rival Dubai’s Palm Jumeirah development. Evergrande also owns a 17% stake in Shangjing Bank, presumably due to the access to financing that owning a bank provides. Lastly, Evergrande also owns a majority stake in Evergrande Health Industry Group, a property developer masquerading as a hospital and media business, listed on the Hong Kong Stock Exchange under the ticker 708 HK. Around 1/3 of profits belong to minority shareholders in projects consolidated into Evergrande’s financials.

The problem with Evergrande is a consistent cash burn, misleading accounting, a high expense ratio, an exposure to lower-tier city markets, high debt levels and a reliance on foreign-currency borrowing. Let’s go through them one-by-one:

The company has burnt cash every single year since 2009, despite going through a goldilocks period in terms of the development of the Chinese property market. Most of the negative cash flows came from heavy investment in new properties under development. But not only: roughly 1/4 of profits in each of the last few years have come from revaluation gains of investment properties – a dubious source of profits given how overvalued the assets on Evergrande’s balance sheet seem to be. Receivables have grown much faster than revenues, with receivable days growing from 58 days in 2015 to 158 days in the first half of 2018. This could suggest delinquency issues or accounting shenanigans when it comes to recognition of revenues. Despite the increase in receivable days, Evergrande’s provisioning for credit losses was just 0.5% in the first half of 2018. Recurring acquisitions of subsidiaries of roughly US$5 billion per year raises questions of round-tripping of cash via related parties, and whether these acquisitions were made at arms-length prices.

The only way to survive and prosper in a true commodity industry is to enjoy a cost advantage vs other companies in the sector. A structurally low SG&A, that enables you to undercut the competition on gross margins and take market share through lower prices. But Evergrande’s SG&A/sales is the second-highest in the industry after Sunac, suggesting that the organization is far less efficient than competitors such as Vanke and CR Land.

Given the government’s guidance that cash subsidies used in shantytown redevelopment is going away, Evergrande’s reliance on lower tier cities is problematic. 91% of Evergrande’s landbank by value is in lower tier cities – those that benefited the most from the last few years of supply-side reform. A return to 2016 levels of GFA sold for those lower tier cities would imply a roughly 50% drop in volumes and an even greater drop in profits.

In line with the increase in developer discounts and lower sell-through rates, Evergrande’s average selling price has started to weaken over the past few months, reaching a six-month low in September after two straight months of falling prices.

In addition, the balance of sales deposits that Evergrande has received from customers shrank in 1H18 for the first time since 2010:

Evergrande’s debt level has come down since the beginning of the year, but still stands at about US$100 billion. US$21 billion of that debt was foreign currency debt, as of year-end 2017.

The company’s stated net debt/equity ratio is misleading, since many of the assets on the its balance sheet are overvalued. As Hawkeye901 pointed out in his 2017 write-up, Evergrande’s investment properties are valued at very low cap rates of just 0.5%. Adjusting their valuation to a 5% cap rate causes more than half of Evergrande’s equity to be wiped out, significantly increasing the company’s net debt/equity ratio. 1H18 equity was also boosted by the introduction of HKFRS 15, which allowed for the company to recognize revenues earlier and give a one-off boost to equity.

In the first half of 2018, Evergrande enjoyed positive operating cash flow for the first time in almost the decade, with a debt service coverage ratio of 1.7x. Given that 1H18 was one of the strongest periods in recent years in terms of margins and cash flow generation, a normalized DSCR should be significantly below this estimate.

45% of Evergrande’s borrowing comes from trust loans. That may have worked fine when the shadow banking sector was growing but presents a funding issue now that the shadow banking sector has started to shrink.

In October, Evergrande raised a 2yr USD bond at 11% when similar maturity bonds outstanding were trading at 7%. This suggests a certain level of stress. Chairman Hui Ka Yan bought $1 billion of the $1.8 billion notes himself, as he seeks to lend support to the company. His recklessness when it comes to financial affairs can be observed from the fact that he borrowed US$500 million from Japan’s Sparxx Group at a 22% interest rate using his Evergrande shares as collateral a few years ago. What appears to be happening is Evergrande is paying a hefty dividend to prop up the share price, and the Chairman then recycles the dividend back into the company to meet short-term maturities. Far more worrying are the company’s 2020 maturities, which will require over US$6 billion to be rolled over, at a time when the shadow banking sector itself is shrinking gradually.

In 2018, the price of Evergrande’s 2025 USD bond has dropped from 103 to 84, reflecting bondholder concern about the company’s solvency.

Foreign-currency borrowing may not be a panacea given Evergrande’s already-high reliance on USD bond issuance.

Using a 5% cap rate for the company’s investment properties when calculating an adjusted book value yields an asset/equity far higheer than any other Chinese developer. I believe that I am not exaggerating when I say that Evergrande is going into the downturn with significant leverage and would suffer immensely in the event of a crash.

Valuation

A scenario analysis highlights the significant operating and financial leverage inherent in the business. In a base case where 2018 contract GFA falls 25% while ASP drops 20% due to Evergrande’s lower-tier city exposure, GPMs would fall from the current 35% level closer to 15%. In such a scenario, assuming a PE multiple in line with historical levels of 7x would yield a fair value of HK$5.5, a downside of 70% from current levels. In an upside, worst-case scenario of continued speculation and further cash subsidies, with land prices rising at the same rate as overall property prices, yielding a normalized GPM of 25% level, as well as assuming a PE ratio of 10x, Evergrande’s share could potentially double. My base case is for lower tier city markets to suffer from the removal of government cash subsidies, returning to pre-2016 transaction volumes.

Risks

• Monetary and fiscal stimulus, thus prolonging the cycle for a few more years. There are early signs of the broad money credit impulse improving. On the other hand, the property market itself is coming off a period of significant speculative activity and support from cash subsidies from shantytown redevelopment.
• Evergrande being “too-big-too-fail”, especially considering Chairman Hui Ka Yan strong relationship with the leadership of the CCP. The recent list of top 100 outstanding entrepreneurs included Hui Ka Yan. He was also one of 52 delegates to a recent “symposium” for private entrepreneurs hosted by Xi Jinping. On the other hand, Hui Ka Yan’s status as one of the wealthiest people in China may also make him a target. There are signs that the leadership is getting impatient with entrepreneurs who have overreached. As an example, it is said that Geely’s Li Shufu is finding it difficult to refinance the EUR 8 billion loan it used to acquire a stake in Daimler. This is a new development.
• The short interest is high at 18% of free float. The short interest ratio has remained persistently high due to share buybacks that have reduced the float.

Catalyst

• An end to the company’s share buybacks
• Continued weakness in the property market
• RMB devaluation
• 2020 debt maturities