余额宝和存款准备金的那些事-《中央银行与货币供给》读后感之二/Yu'ebao and the Matters of Reserve Deposits - Reflections on Central Banks and Money Supply, Part Two

本文于2015年6月7日发表于央行观察

人民银行调查统计司盛松成司长与翟春老师的新书《中央银行与货币供给》围绕着经典货币银行学的脉络展开,其中存款准备金机制是这一脉络的核心要点,而余额宝和准备金的关系则是新出现的现实问题,因此,盛司长特将两篇关于余额宝与准备金的文章附于本书篇末压轴,亦即《余额宝与存款准备金管理》与《什么是存款准备金管理?》。这两篇文章,都是在2013年夏天余额宝横空出世之后,中国金融界对余额宝高企的收益率、活期存款般的流动性、与纸币几无二致的购物能力、以及快速成长为中国第一大基金的瞠目结舌下写就的。

 文章的主要观点,是认为非存款性金融机构在存款性金融机构的存款应该进入存款准备金的交存范围。所谓存款性金融机构其实就是银行和财务公司,而非存款性金融机构则相当多,证券公司、保险公司、基金、信托公司、金融租赁等无一不包。这一思路的结果,当然就是2014年底的银发[2014]387号文,即将证券及交易结算类存款、银行业非存款类存款,SPV存放,其他金融机构存放以及境外金融机构存放纳入各项存款口径,并计入存款准备金交存范围。而从央行的统计标准来看,余额宝作为货币市场基金,其在银行的存款应该属于SPV存放。 

货币市场基金有自己的资产负债表。其资产包括债券投资,买入返售金融资产,应收利息,以及银行存款等。这一资产组合,与商业银行通常持有的贷款,债券,买入返售金融资产,应收利息,准备金存款并无本质区别,只不过银行持有的贷款不像债券一样可以转让,银行的准备金存款存放在央行,银行持有的贷款和债券质量有好有坏。货币市场基金也有其负债和所有者权益,负债包括卖出回购金融资产以及很多应付项目,但规模较小,而所有者权益则非常巨大,因为基金投资者买入基金份额,在会计上是持有基金股权,这也是为什么基金投资者面对的是不断波动的净值。银行的负债和所有者权益则与基金不同,最大的部分是客户存款,对银行而言是负债,而少部分是股权。 

这里展现了两个货币市场基金的资产负债表的构成情况,一个是余额宝(对接的是天弘基金增利宝货币基金),一个是宝盈货币市场基金(宝盈货币A)。前者资产以银行存款为主,而后者则以债券为主;前者资产对应的几乎都是股权,而后者则有少许债务杠杆。银行的资产负债表,则更像是将宝盈货币A的银行存款替换为准备金存款,债券替换为贷款,所有者权益替换为存款,而卖出回购替换为所有者权益。

余额宝的厉害之处在于两点。首先,余额宝的净值并不像股票基金净值那样上蹿下跳,也不像一般基金那样按月发红利,而是每天返收益,加上其两小时内赎回到存款账户的功能,使得余额宝份额在其用户看来,其“份额净值”是稳定不变的,1元份额兑换的虽然不是1元纸币(央行的负债),但已经是差不多好的1元银行存款(银行负债)。用户已经不用区分这份额代表的是股权,还是存款一样的债权,只要1元份额净值能随时转化成1元银行存款,就可以当作银行存款来花。

其次,余额宝成为了一家“影子银行”。用户都很清楚,在买火车票的时候,可以选择储蓄卡支付、信用卡支付以及余额宝支付。当然,铁总收到的都是银行存款:用户选择储蓄卡支付时,自身存款减少,铁总的存款增加;信用卡支付时,用户对银行的负债增加,铁总的存款增加;余额宝支付时,用户在余额宝的份额减少,余额宝在银行的存款减少,铁总的存款增加。我们注意到,余额宝以商业银行存款为准备金,余额宝余额相当于客户在商业银行的存款,余额宝的使用者只要用余额宝互相转帐,无论买卖多少东西支付多少服务,对余额宝而言都只是在股权侧改改记录,而在资产侧完全没有变化。用户以为自己在余额宝转帐的是银行存款/现金/钱,但其实只是基于余额宝“一比一兑换银行存款”承诺的记帐单位,余额宝在银行的存款根本没有转。

余额宝用户之间转帐,单个商业银行的存款客户之间转帐,以及人民银行的准备金存款客户(其实就是商业银行)之间或者纸币持有者之间转帐,都不会影响到各自的资产侧。只有当余额宝用户需要支付给铁总或者微信理财通的用户(目前操作尚不可行)时,工商银行的客户需要支付国库或者中国银行的客户时,人民银行需要支付给美联储时,才会影响到自身的资产。这里的铁总、微信理财通用户与国库、中国银行客户并非随意列举:铁总在银行有存款帐户,国库在央行有存款帐户,而微信理财通用户持有微信的红包余额,微信在银行有存款,中国银行客户在中国银行有存款余额,银行在央行有准备金存款。

由于余额宝份额能够在实体经济中广泛使用,余额宝可以实现传统银行式的信贷扩张和货币创造。这里的信贷扩张和传统货币基金买债券不同。传统货币基金买债券,债券发行人得到银行存款,并不一定会存回该货币基金中,相当于传统货币银行分析中的现金漏出。但余额宝可以给淘宝上所有的商户都提供余额宝份额作为贷款,得到贷款的商户用上述份额可以在淘宝上采购各种货物和服务,事实上在万能的淘宝上也确实可以做到,最后商户以余额宝份额偿还贷款,信用货币由此经历了从创造到消失的全过程。了解银行运作的读者一定发现了,这就是一家银行的运行机制。当然在货币基金的管理框架下,目前恐怕还没有达到这一境界,但一旦放开,余额宝就可以成为一家银行。

余额宝的一元份额,并不是人民币的一元,只是表示“正常情况下”能够转出而得到一元人民币的存款,而在挤提的“非正常情况”下还得不到一元。这和金本位发展之后纸币的面额也用1镑等重量单位表示可以提取含有某确定重量黄金的金币是完全一样的模式,也和银行存款用人民币元作为单位来为存款计价是一样的模式。当银行如果出现危机,存款也会打折,一元存款只能兑换7角人民币纸币,这正是塞浦路斯金融危机中银行和储户所处的情况。余额宝的一元份额与一元银行存款的关系,是靠余额宝的稳健经营来维持的。当然它现在十分稳健,因为其份额对应的资产有80%以上都是银行存款。

在揭示了如何用银行的视角来分析余额宝的运行之后,这里讨论一下准备金制度。盛司长在其文章中指出了准备金制度的两个功能,即防范单个银行的流动性风险(防备挤提),以及控制整个银行业的存款总额(货币总量)。对货币基金而言,如果其在银行的存款不在交存范围中,那么当货币基金投资者大量赎回基金份额,使原先交存范围外的存款进入了交存范围内,银行存款准备金规模不变,但交存范围内的存款总量暴增,银行的准备金比率会低于法定要求,这时候顺势一挤提(比如纷纷转走存款到其他银行),银行就会出现流动性问题。由于存款保险制度的存在,上述风险会被弱化。当然基金公司直接将存款转帐至其他银行,更会减少银行的准备金,使流动性风险增加,但这应该是银行出了问题而导致的,不像是余额宝的问题。此外,盛司长的文章讨论的主要是大规模赎回下的情况,这里也仅讨论这一情形。

而在存款总额与货币总量的问题上,银行发现交存范围内的存款跑到交存范围外了,超额准备金增加了,便又可以发放信贷和结汇,结果导致交存范围内的存款增加,当货币基金投资人赎回时,之前新增的存款不会减少,而赎回使交存范围外的存款又回到交存范围内,银行发现超额准备金少了,法定准备金率要失守了,又需要售汇或者紧缩信贷了。由于在货币量的统计上,非存款性金融机构在存款性金融机构的存款也计入货币量,因此货币基金的购买与赎回本身大概对货币量统计没有影响,但由于货币基金在银行的存款进出准备金交存范围会影响银行在资产侧的操作,因而对总存款量和总货币量会产生干扰。当然目前由于存贷比等其他监管要求的缘故,银行操作并不能像上面描述的那样自如。

如果货币基金在银行的存款纳入存款准备金交存范围,那么上述问题当然就都可以避免了。银行不会因为储户购买基金而扩张资产负债表,因为基金在银行的存款和储户存款一起受到法定准备金率和准备金规模的限制。因此,当储户赎回时,银行也不会发现需要更多的准备金存款,因为法定准备金要求并没有因赎回而提高。盛司长还有一个关于余额宝应该缴存准备金的观点在于市场公平竞争。如果任由基金的银行存款游离于交存范围外,银行会不喜欢那些在交存范围内的存款,会希望这些存款通过基金这一通道变身为交存范围外的存款,而保险公司存款这种在交存范围内的存款就会被歧视。因此,像本文开头提到的,盛司长主张把其他非存款类金融机构在银行的同业存款也放入交存范围内。

至于如何计算余额宝进入存准范围对其收益的影响,盛老师的公式是:“假定余额宝-增利宝基金投资银行协议存款的款项须缴存20%的准备金,按照6%的该基金协议存款利率和我国统一的1.62%的法定存款准备金利率计算,拥有5000亿资金规模的余额宝一年成本将增加约42亿元(5000×95%×20%×(6%-1.62%)≈42),收益率下降约1个百分点”。公式的意思应该是说,余额宝有95%也就是4750亿都是某家银行的存款,这些存款的回报率是6%,对应了银行的一块儿净收益为6%的资产。忽然,这4750亿进入了存准交存范围,于是这净收益6%的资产中中有20%也就是950亿需要卖给央行或者其他商业银行换存款准备金,收益率一下子就跌到1.62%了,收益也少了42亿。

这里特别指出,作为银行负债的存款是不能交银行在央行的存款准备金的,即不能理解为余额宝有950亿的银行存款直接变为了准备金存款,这种情形只能发生在余额宝在央行开有帐户的前提下,而且余额宝上缴存款准备金的同时还会造成银行在央行持有的存款准备金的转移,但现在余额宝在央行没有帐户。所以盛司长的公式,需要从单个银行调整余额宝存款的对应资产这一角度理解。当然单个银行未必给余额宝单划了一个资产,反而从担心余额宝另寻其他银行合作的角度,银行自己会承担这42亿的成本,或者均摊到其他债权人上。

而如果我们不考虑单个银行,而把视角投向整个银行业,我们发现,由于准备金存款总量由央行确定,这也是《中央银行与货币供给》一书的重要观点,因此面对余额宝在银行的存款进入存准交存范围一事,在准备金总量不变的情况下,法定准备金部分增加,超额准备金部分减少,整个银行体系的收入变动,其实是超额准备金率和法定准备金率之间的差额带来的。事实上4750亿存款进入交存范围,使950亿存款准备金由超额状态变为法定准备金状态,利率反而从0.72%变为1.62%,整个银行体系从央行得到的收入增加了8.55亿元每年,而并不是减少了42亿元。单个银行的考虑思路往往和银行体系视角下的分析不同,因为单个银行分析往往假定其他银行不变,比如上面提到的银行出售资产给另一家银行时,就没有考虑另一家银行会怎么样调整自己的资产负债表,它的超额准备金肯定减少了它怎么办之类的。

当然如果银行体系想要维持余额宝存款进入存准交存范围前的超额准备金率或超额准备金量,则收缩资产负债或者要求央行支持就成为必需了,这又会导致新一轮对净收入的影响。至于上面讨论的银行间买卖6%的生息资产,只是收入在不同银行间分配的变化,而不影响整个银行体系的收入。而资产卖给央行,使央行扩张资产负债表,则使银行体系失去了一部分收入,因为卖给央行的资产的利率是6%,而央行给出的法定存款保证金的利率是1.62%,利差是4.38%,这或许不如直接找央行借款或者卖外汇划算。

而如果把余额宝看作是准备金存款建在银行侧的银行,那么除了余额宝在银行的存款应该放入银行的存款准备金交存范围外,余额宝在银行的存款也应该和余额宝的总规模保持一个“准备金率”,因为除了传统上货币基金为应对流动性风险应保留的备付金之外,还应该考虑到余额宝份额本身的货币流通属性。越是使用广泛的货币,其需要的准备就越少,只要对数量控制得当,流通本身就可以支持这个货币运转起来。余额宝份额和其对应的银行存款,就可以像130万亿银行存款对应7000亿库存纸硬币那样,达到千分之五的备付率,而余额宝在银行目前的存款已经达到5000亿,余额扩张规模还有很大的想象空间。当然有人会说需要考虑超额准备金存款所代表的7000-10000亿元,但这部分的产生是由于多个银行的存在造成的。如果只有1家银行存在于市场,那么超额准备金大可不必存在,唯一需要预防的就是提取纸币,这和支付宝面临的唯一需要预防的就是赎回成存款一样。

因此,如果人民银行还采取盯住货币量的政策取向,以存款准备金制度为调控的手段,那么余额宝的份额也应该进入货币量的统计范围当中,余额宝在银行的存款应被看作其准备金,准备金应该与余额宝总规模存在比例下限。这不是货币基金的属性导致的,而是余额宝的广泛流通性导致的。当然人民银行肯定会遇到一个问题,在于货币基金资产组合配置的风险控制是证监会的工作。这一监管范围的划分当然是历史合理的,毕竟传统货币基金份额并不能用于余额宝这样广泛的购物和支付,基金持有者只有赎回份额成银行存款后才能和实体经济发生关系,也才同时进入货币政策的考虑范围。

可以考虑的另外一个问题是,既然余额宝份额已经获得了银行存款般的流通能力,那么是否应该要求余额宝像银行一样,直接在央行开设准备金和备付金账户?这样其在银行的存款就不用交准备金了,准备金政策对余额宝份额规模的约束就更为直接,不用像上一段提到的那样,余额宝在银行的存款适用准备金率,而余额宝自身规模和在银行的存款间也适用准备金率,余额宝和央行之间还要通过银行来传导。但这就意味着一家由证监会监管的银行的诞生,一家没有存款保险的银行的诞生。这当然也涉及到如何看待余额宝份额的货币性的问题,如果否认其客观具有的流通性,坚持“份额不是存款,存款才是货币,所以份额不是货币”的论点,自然也就不能接受“余额宝已经成为事实上的银行”这一结论,更不用说让余额宝和银行一样在央行开户存放准备金了。

总之,余额宝作为一个新生事物,一方面它是货币市场基金,带有货币市场基金的性质,同时它又联系了淘宝,具有了传统货币市场基金不具备的性质。很多人用美国上世纪末货币基金的红火与没落来预测余额宝的未来,这是没有看到本世纪新出现的余额宝,已经与实体经济的紧密结合。基金份额的货币化和流通化,使货币基金焕发了新生机。余额宝对银行体系的挑战才刚刚开始。

This article was published on June 7, 2015, in "Central Bank Observation"

Director Sheng Songcheng from the People's Bank of China's Survey and Statistics Department and Professor Zhai Chun have released a new book titled "Central Banks and Money Supply," which delves into the context of classical monetary and banking studies. Within this context, the reserve requirement mechanism holds a central role. The relationship between Yu'ebao and reserves is a new emerging real-world issue. As a result, Director Sheng has appended two articles concerning Yu'ebao and reserves at the end of this book. These articles are titled "Yu'ebao and Deposit Reserve Management" and "What is Deposit Reserve Management?" Both articles were written after the sudden emergence of Yu'ebao in the summer of 2013, capturing the attention of the Chinese financial sector with its high returns similar to current deposits, liquidity akin to demand deposits, shopping capabilities closely resembling paper currency, and its rapid growth to become China's largest fund.

The primary viewpoints of the articles emphasize that non-deposit financial institutions' deposits in deposit-taking financial institutions should fall within the scope of deposit reserve requirements. Deposit-taking financial institutions refer to banks and financial companies, whereas non-deposit financial institutions encompass a variety of entities, including securities companies, insurance companies, funds, trust companies, financial leasing, and more. This line of thought ultimately led to Document No. 387 of 2014 issued by the Banking Regulatory Commission, which incorporated various deposit categories such as securities and trading settlement deposits, non-deposit deposits from the banking sector, SPV deposits, deposits from other financial institutions, and deposits from overseas financial institutions into the deposit reserve requirement range. From the central bank's statistical standards, Yu'ebao, as a money market fund, should fall under the category of SPV deposits within banks.

Money market funds have their own balance sheets. Their assets include bond investments, repurchase financial assets, interest receivables, and bank deposits, among others. This asset composition is fundamentally similar to that of commercial banks, which typically hold loans, bonds, repurchase financial assets, interest receivables, and reserve deposits. The primary distinction lies in the fact that loans held by banks cannot be transferred like bonds, bank reserve deposits are held at the central bank, and the quality of loans and bonds held by banks can vary. Money market funds also have liabilities and owner's equity. Liabilities include repurchase financial assets and various payable items, although on a smaller scale, while owner's equity is substantial. This is because fund investors buy fund shares, which, from an accounting perspective, represent ownership in the fund. This is why fund investors face fluctuating net asset values. In contrast, banks' liabilities and owner's equity differ from those of funds. The largest portion is customer deposits, which are liabilities for banks, while a smaller fraction represents equity.

Here, two examples of money market fund balance sheets are presented: one is Yu'ebao (associated with Tianhong Fund's Money Plus Fund), and the other is Bao Ying Money Market Fund (Bao Ying Money A). The former's assets are primarily bank deposits, while the latter's are mainly bonds; the assets corresponding to the former's shares are predominantly equity, while the latter possesses a slight amount of debt leverage. A bank's balance sheet is more similar to that of Bao Ying Money A: bank deposits are substituted for reserve deposits, loans are substituted for bonds, and customer deposits are substituted for owner's equity, with repurchase transactions being replaced by owner's equity.

What sets Yu'ebao apart is twofold. Firstly, the net asset value of Yu'ebao's shares doesn't fluctuate as dramatically as stock funds' net asset values or pay dividends on a monthly basis like regular funds. Instead, it returns earnings daily. Coupled with its functionality to redeem to a deposit account within two hours, Yu'ebao's shareholders perceive its "net asset value per share" as consistently stable. Although 1 yuan in shares doesn't equate to a 1 yuan banknote (a liability of the central bank), it is nearly equivalent to a 1 yuan bank deposit (a liability of a bank). Users no longer need to differentiate between whether the share represents equity or deposit-like debt. As long as the net asset value of 1 yuan per share can be instantly converted into a 1 yuan bank deposit, it can be used as a bank deposit.

Secondly, Yu'ebao has become a "shadow bank." Users are well aware that when buying train tickets, they can opt for payment via savings cards, credit cards, or Yu'ebao. Of course, the railways only receive bank deposits: when users pay with savings cards, their deposits decrease, and the railways' deposits increase; when they pay with credit cards, users' liabilities to banks increase, and the railways' deposits increase; when paying with Yu'ebao, users' holdings decrease, Yu'ebao's deposits in banks decrease, and the railways' deposits increase. It's worth noting that, with commercial bank deposits as reserves, a user's Yu'ebao holdings essentially represent customer deposits within a commercial bank. When users transfer money between Yu'ebao accounts, regardless of the amount spent or the number of services purchased on platforms like Taobao, the records merely alter on the equity side, while the asset side remains unchanged. Users assume they're transferring money/cash when using Yu'ebao to transfer funds, but they're essentially using units based on Yu'ebao's "1-to-1 conversion of bank deposits" promise. Yu'ebao's deposits in banks haven't actually been moved.

Transfers between Yu'ebao users, between individual bank deposit customers, or between reserve deposit customers of the People's Bank of China (which are essentially commercial banks) or between paper currency holders have no effect on their respective asset sides. Only when Yu'ebao users need to pay other users of platforms like WeChat Financial Services (currently not feasible) or when Industrial and Commercial Bank of China customers need to pay the treasury, or when China Bank customers need to pay the Bank of China, would this influence their own assets. The mentioned Industrial and Commercial Bank and China Bank customers were not arbitrarily selected: Industrial and Commercial Bank holds bank deposit accounts, the treasury holds deposit accounts at the central bank, while WeChat Financial Services users possess balances in WeChat's "red envelopes," WeChat holds deposits in banks, and Bank of China customers hold bank deposit balances in Bank of China. Banks hold reserve deposits at the central bank.

Because Yu'ebao shares can be widely used in the real economy, it can achieve traditional bank-style credit expansion and money creation. This credit expansion differs from that of traditional money market funds buying bonds. When traditional money market funds buy bonds, the issuer of the bond receives bank deposits, which might not necessarily return to the money market fund, creating a cash outflow similar to the traditional money and banking analysis. However, Yu'ebao can provide Yu'ebao shares as loans to all merchants on Taobao. These merchants can use the shares to purchase various goods and services on Taobao. Afterward, the merchants repay the loans using Yu'ebao shares. Thus, credit money goes through the entire process from creation to disappearance. Within the current framework of fund management, this level has probably not been reached, but once relaxed, Yu'ebao could effectively become a bank.

A 1 yuan share in Yu'ebao doesn't equate to 1 yuan in Chinese currency. It merely represents the deposit that, under "normal circumstances," can be converted into 1 yuan in Chinese currency. In abnormal situations, such as a run on deposits, it might not be converted into 1 yuan. This is similar to the pattern where, after the development of the gold standard, the denomination of banknotes represented a specific weight of gold, and the practice of pricing bank deposits using Chinese yuan. When banks face a crisis, deposits can also be discounted. A 1 yuan deposit can only be exchanged for a 70-cent Chinese currency banknote. This is exactly the situation during the Cyprus financial crisis, where banks and depositors were situated. The relationship between a 1 yuan Yu'ebao share and a 1 yuan bank deposit is maintained through the sound operation of Yu'ebao. Currently, its operation is robust because over 80% of the assets corresponding to its shares are bank deposits.

After unveiling how to analyze Yu'ebao from a bank's perspective, the discussion shifts to the reserve requirement system. Director Sheng's articles highlight two functions of the reserve requirement system: preventing the liquidity risk of individual banks (preventing runs) and controlling the total amount of deposits in the entire banking industry (money supply). For money market funds, if their deposits in banks fall outside the scope of the reserve requirement, then if fund investors extensively redeem fund shares, causing previously out-of-scope deposits to fall within the scope, the scale of bank deposit reserves remains unchanged, but the total amount of deposits within the scope dramatically increases. The bank's reserve ratio would fall below the statutory requirement. If a run occurs at this point (such as transferring deposits to other banks), the bank could face liquidity issues. The presence of deposit insurance weakens these risks. Of course, if fund companies directly transfer deposits to other banks, it would further reduce the bank's reserve deposit and increase liquidity risk. However, this should be a result of bank issues, not a problem with Yu'ebao. Moreover, Director Sheng's articles primarily discuss scenarios involving large-scale redemptions, and this discussion is limited to that context.

In terms of the total amount of deposits and the total money supply, banks have observed that deposits within the reserve range have flowed outside of it. As excess reserves increase, banks can then extend credit and make foreign exchange settlements. This results in an increase in deposits within the reserve range. When investors redeem money market fund shares, the previously added deposits do not decrease. However, redemptions cause deposits outside the reserve range to return to within the range. Banks then discover that excess reserves have decreased, and there is a risk of breaching the statutory reserve requirement, necessitating foreign exchange sales or credit tightening. Due to statistical considerations of money supply, non-deposit financial institutions' deposits within deposit-taking financial institutions are also included in the money supply. Therefore, while the purchase and redemption of money market funds probably do not have a direct impact on money supply statistics, the movement of money market fund deposits in and out of banks affects the banks' asset-side operations and thus creates interference with the total deposit and money supply levels. Of course, due to other regulatory requirements such as the loan-to-deposit ratio, banks' operations may not be as fluid as described above.

If money market fund deposits in banks are included in the reserve requirement range, then all the aforementioned issues can naturally be avoided. Banks will not expand their balance sheets due to customers purchasing funds since both the fund deposits in banks and customer deposits are subject to the statutory reserve requirement and reserve scale. Therefore, when customers redeem, banks will not need more reserve deposits since the statutory reserve requirement does not increase due to redemptions. Director Sheng also puts forth the view that Yu'ebao should contribute to fair market competition. If banks allow fund deposits to exist outside the reserve range without restriction, banks may dislike deposits within the reserve range and may prefer these deposits to become outside-the-range deposits through funds. This could lead to discrimination against in-range deposits like those from insurance companies. Therefore, as mentioned at the beginning of this article, Director Sheng advocates for including non-deposit financial institutions' interbank deposits in the reserve requirement range.

Regarding how to calculate the impact of Yu'ebao's entry into the reserve range on its earnings, Director Sheng's formula is as follows: 'Assuming that the funds invested by Yu'ebao-Money Plus Fund in bank agreement deposits require a 20% reserve requirement, and based on a 6% deposit rate for this fund agreement and China's unified 1.62% statutory deposit reserve rate, with a fund size of 500 billion, the annual cost of Yu'ebao will increase by about 4.2 billion yuan (500 × 95% × 20% × (6% - 1.62%) ≈ 4.2 billion), and the yield will decrease by about 1 percentage point.' The meaning of this formula is that 95% (475 billion) of Yu'ebao's deposits belong to a certain bank, and these deposits yield 6%, generating 6% net income assets for the bank. Suddenly, these 475 billion fall into the reserve deposit range, and thus, 20% (95 billion) of the 6% net income assets (475 billion) need to be sold to the central bank or other commercial banks to exchange for reserve deposits. This immediately reduces the yield to 1.62%, and the income is reduced by about 4.2 billion yuan.

It is particularly pointed out here that as bank liabilities, deposits cannot be used to deposit reserve deposits at the central bank. In other words, it cannot be understood that the 95 billion bank deposits of Yu'ebao directly turn into reserve deposits. This kind of situation can only occur when Yu'ebao has an account with the central bank, and at the same time, when Yu'ebao submits reserve deposits, it will also cause the transfer of the reserve deposits held by the bank at the central bank. However, currently Yu'ebao does not have an account with the central bank. So, Director Sheng's formula needs to be understood from the perspective of adjusting the corresponding assets of individual banks for Yu'ebao deposits. Of course, a single bank may not necessarily allocate a separate asset for Yu'ebao. Instead, from the perspective of worrying that Yu'ebao might cooperate with other banks, the bank itself may bear the cost of 4.2 billion or distribute it among other creditors.

If we do not consider individual banks and shift our perspective to the entire banking industry, we find that since the total amount of reserve deposits is determined by the central bank, this is an important point in the book "Central Banks and Money Supply". Therefore, facing the issue of Yu'ebao's deposits in banks entering the reserve deposit range, in the case of a constant reserve deposit total, an increase in the statutory reserve deposit portion will lead to a decrease in the excess reserve deposit portion. The change in income for the entire banking system is actually due to the difference between the excess reserve requirement rate and the statutory reserve requirement rate. In fact, the 475 billion deposits entering the reserve range cause the 95 billion deposit reserve deposits to change from an excess state to a statutory reserve state, and the interest rate changes from 0.72% to 1.62%. The entire banking system's income from the central bank increases by 855 million yuan per year, rather than decreasing by 4.2 billion yuan. The considerations of an individual bank are often different from the analysis from the perspective of the banking system, as the analysis of an individual bank often assumes that other banks remain unchanged. For example, the sale of assets by a bank to another bank mentioned above does not consider how the other bank will adjust its own balance sheet. If its excess reserve deposits have decreased, what would it do, etc.

Of course, if the banking system wants to maintain the excess reserve requirement rate or excess reserve deposit amount before Yu'ebao's deposits enter the reserve deposit range, it would require shrinking its assets and liabilities or seeking support from the central bank, which would lead to a new round of impact on net income. As for the buying and selling of 6% interest-bearing assets between banks discussed above, this only represents a change in income distribution among different banks and does not affect the entire banking system's income. Selling assets to the central bank and causing the central bank to expand its balance sheet would lead to a reduction in income for the banking system because the interest rate on assets sold to the central bank is 6%, while the interest rate on statutory deposit reserve guarantees provided by the central bank is 1.62%, resulting in an interest spread of 4.38%. This might not be as cost-effective as directly borrowing from the central bank or selling foreign exchange.

And if we consider that since Yu'ebao's shares have gained circulation capability similar to bank deposits, should we require Yu'ebao, like banks, to directly open reserve and reserve fund accounts with the central bank? This way, its deposits in banks would not need to pay reserve deposits, and the reserve policy's constraint on Yu'ebao's share size would be more direct. It would not be necessary, as mentioned in the previous section, for both the deposit held by Yu'ebao in the bank and Yu'ebao's own size and the bank's deposit to be subject to reserve deposit rates, and the transmission between Yu'ebao and the central bank to go through the bank. But this would mean the birth of a bank regulated by the China Securities Regulatory Commission (CSRC) and a bank without deposit insurance. This certainly also involves how to perceive the currency attributes of Yu'ebao shares. If the objective circulation capability is denied, and the argument 'Shares are not deposits; only deposits are money, so shares are not money' is upheld, then it's natural that 'Yu'ebao has effectively become a bank' cannot be accepted as a conclusion. Not to mention the suggestion of having Yu'ebao open an account with the central bank for depositing reserves like banks.

In summary, as a new entity, Yu'ebao has characteristics of both a monetary market fund and a linkage with Taobao, which traditional monetary market funds lack. Many people use the rise and fall of U.S. monetary funds at the end of the last century to predict the future of Yu'ebao. However, they have not recognized that the Yu'ebao of this century has already closely integrated with the real economy. The monetization and circulation of fund shares have revitalized the money market fund. The challenge to the banking system from Yu'ebao has just begun.