Page 18 - CCFA Journal - 8th Issue
P. 18
经济热点 Economy hotspots 加中金融
停工项目规模测算
由于信息有限,对停工项目具体规模进行准确测算的难度较高,一种方法是比较水泥消费量和房屋施工面积。
在 2016-20 年期间,房屋施工面积增速和水泥产量增速的平均缺口为 2.1 个百分点,但 2021 年扩大至 10.3 个百分点,
2022 年前 5 个月进一步扩大至 15.8 个百分点(图 5)。
假设 2020 年为常规情形,以该年为基准并假设 2022 年全年增速与 2022 年 1-5 月的同比增速相同,那么 2021-22 年期
间房屋施工面积可能共计增长 9.6%,而水泥产量可能共计下降 16.3%,这表明停工项目占在建项目的比重或接近
20.3%。当然,对于许多项目而言,施工进度仅是放缓而非暂停,因此这 20.3%可作为停工项目占比上限的粗略估计。
Estimating the scale of paused construction projects
With limited information, it’s difficult to come up with an accurate and specific estimate of projects, of which construction has
been paused. One approach is to compare the consumption of cement and floor space under construction.
Between 2016 and 2020, the average gap in growth in floor space under construction and growth in cement output is 2.1
percentage points, but the gap widened to 10.3ppt in 2021 and 15.8ppt in the first five months of 2022 (Figure 5).
Assuming a normal situation in 2020 and using that year as a benchmark, and assuming 2022 annual growth is the same as
year-on-year growth during January-May 2022, then during 2021-22, floor space under construction could rise by a total of
9.6%, but cement output could drop by a total of 16.3%, suggesting the ratio of paused projects to total projects under
construction might be somewhere close to 20.3%. Of course, for many projects, construction just slowed rather than being
paused, so this 20.3% could serve as a rough estimate of the upper limit of the ratio of paused projects.
对银行资产质量的影响
截至 2022 年 3 月底,住房按揭贷款余额共计 38.8 万亿元,占银行贷款总额的 19.3%。住房按揭贷款一直被视作最安全
的资产,因为如果房价不下跌,房产即可作为保值的优质抵押物。央行在 4 月中旬举行的新闻发布会上指出,中国个
人住房贷款不良率长期维持在 0.3%左右。住房按揭贷款规模大、利率高,是银行的主要利润来源。央行数据显示,截
至 2022 年一季度末,个人房贷加权平均利率(包括首套房和二套房)已从 2021 年底的 5.63%回落至 5.49%
(现有最新数据)。
考虑到平均较高的首付比例以及目前仍颇具韧性的房价,我们认为住房按揭贷款仍然是安全资产。然而,市场或需注
意期房交付率下滑、业主主动停贷、一些城市房价下跌以及失业率攀升等因素带来的风险。
根据官方数据,我们估计 2013-20 年房企预售房屋的交付率仅为 60%左右,而在这些年间中国住房按揭贷款余额增加
了 26.3 万亿元。由于若干数据质量问题,实际房屋交付率可能存在低估,但一些房企无法按期交房,已成为当前中国
房地产市场面临的一大现实问题。此外,即使是近期预售的房屋也面临交付风险,因为部分房企因财务状况紧张而暂
缓甚至完全停止施工,这可能会对 2020 年底至 2022 年 3 月 4.4 万亿元的新增住房贷款产生一定影响。
市场或许倾向于对银行贷款进行具体测算,但此举颇具风险。鉴于情况并 不稳定、数据质量欠佳,而政策解决方案的
不确定性仍然较高,我们认为应对任何看似准确的测算数据持保留态度。对此,我们在前文测算得出的停工项目占比
或可作为基准。
Impact on banks’ asset quality
As of end March 2022, outstanding mortgage loans totalled RMB38.8trn, or 19.3% of total bank loans. Mortgage loans have
been perceived as the safest assets as residential properties have decent collateral if home prices do not decline. According to
the PBoC press conference in mid-April, the non-performing loan ratio of mortgage loans in China has been quite steady at
just around 0.3%. Mortgage loans are major contributors to banks’ profits, due both to the substantial size and decent
mortgage rates. According to the PBoC, the weighted average of mortgage rates, including both first-time and second- time
home buyers, moderated to 5.49% at end-Q1 2022 (latest data available) from 5.63% at end-2021.
In our view, mortgage loans are still safe assets mainly owing to the average high down- payment ratios and relatively resilient
home prices so far. However, markets might need to take in account some risks associated with the falling delivery rate of
pre-sold homes, the bottom-up movement of refraining from making mortgage repayments, falling home prices in some cities,
and the rising unemployment rate.
Based on official data, we estimate that developers have only delivered around 60% of homes they pre-sold between 2013
and 2020, while in those years China’s outstanding mortgage loans rose by RMB26.3trn. The real delivery rate might be
underestimated due to some data quality issues, but the failure by some developers to deliver homes on time is a real
problem in China’s property sector. By contrast, even newly presold homes face delivery risks, as some developers slowed or
even outright halted construction due to financial strains, endangering the new RMB4.4trn in new mortgage loans between
end- 2020 and March 2022.
People might be tempted to come up with a specific estimate of bank loans, which are under risk. In our view, the situation is
fluid, data are of low quality, and policy solutions are still quite uncertain, so any seemingly exact estimate should be taken
with a grain of salt. Our estimates of the ratio of paused projects in the above section may serve as benchmark in this regard.
CCFA JOURNAL OF FINANCE August 2022
Page 18 第18页