Everyone Is Arguing About China's New Asset Management Rules. What Do They Actually Say?

Will they end the frenzy in the wealth management market?

On April 27, the People's Bank of China, the China Banking and Insurance Regulatory Commission, the China Securities Regulatory Commission and the State Administration of Foreign Exchange jointly issued the Guiding Opinions on Regulating the Asset Management Business of Financial Institutions. The new asset management rules are now official.

The rules cover a lot of ground, but the part that matters to most people is this: wealth management products may no longer promise "guaranteed principal and guaranteed returns." When an investment loses money, investors can no longer count on the institution to make them whole.

The rules take full effect at the end of 2020, and what they face is a wealth management market worth 100 trillion yuan.

Bank wealth management products used to sit in a legal gray zone — but the market was already enormous

Under the Commercial Bank Law, commercial banks in the People's Republic of China may not engage in trust investment or securities business. In other words, a bank wealth management product cannot be registered as a business entity. To subscribe for shares in a private fund, a bank has to invest through a trust plan or an asset management plan.

The earliest model for bank wealth management was investing through trust plans. Then in 2012, the asset management businesses of brokerages, mutual funds, fund subsidiaries, futures companies and insurers opened up one after another, and the range of asset management products that bank wealth management money could buy expanded enormously.

By the end of 2017, financial institutions' asset management business had reached 100 trillion yuan. Of that, banks' off-balance-sheet wealth management products accounted for 22.2 trillion yuan and capital trusts entrusted to trust companies for 21.9 trillion yuan; the outstanding balances of public funds, private funds, brokerages' asset management plans, fund and fund-subsidiary asset management plans, and insurance asset management plans were 11.6 trillion, 11.1 trillion, 16.8 trillion, 13.9 trillion and 2.5 trillion yuan respectively.

An asset management industry this large carries enormous weight for the whole financial system and the whole economy. Yet before this round of rules, plenty of legal and regulatory gray areas surrounded the way bank money got invested in asset management products.

Take one basic wealth management structure: bank wealth management product → trust plan → asset management plan → trust plan → private fund → underlying non-standard debt asset. That chain runs through the segmented oversight of the CBRC → the CSRC → the CBRC → the CSRC, and every individual link is very likely legal on its own. But once money is invested through layer upon layer of nested products, regulators genuinely cannot tell where it finally lands. In practice a good deal of it probably flowed into overheated sectors, working against the government's efforts to steer the economy.

This kind of structure also breeds leverage that is both murky and too high. Under the old siloed system, each regulator capped the leverage of the asset management products in its own patch. But stack products in layers and the leverage ratios multiply.

Many people still remember the wild swings in the A-share market in 2015. A lot of what went wrong was planted the moment "guaranteed principal and guaranteed returns" became the norm.

What does "guaranteed principal and returns" mean, and how was it pulled off?

"Guaranteed principal and returns" is the familiar "expected rate of return" model: when a wealth management product matures, the manager usually pays investors the expected rate of return. Plenty of financial institutions, commercial banks above all, prefer to hand investors that "expected return" — both to protect the institution's reputation and to grow their asset management business.

That 100 trillion yuan of investments could routinely deliver investors a stated "expected rate of return" is, on its face, unreasonable. To pull it off, banks and asset managers leaned on two devices: the "capital pool" on the funding side and "securitization" on the asset side.

The capital pool works like this. A bank gathers the money raised by its wealth management products into a single pool, then spreads it across a mix of underlying assets — some low-risk and low-return, others high-risk and high-return.

If a top-level asset management product matures before the underlying project does, or the underlying project misses its expected return, the institution simply draws on the internal pool to pay the maturing product's principal and expected return.

The bank's profit is the spread between what the underlying assets earn and what it pays out. If the underlying assets return 10% and the top-level product promises 5%, the institution keeps the 5 points in between.

The heart of the game is using new money to settle old money. As long as new asset management products keep selling, the pool keeps turning. Over the past several years the outstanding balance of asset management products climbed steadily and fast — Chinese households were getting richer, more of them wanted their wealth managed, interest rates were being liberalized. That the capital pool worked during the industry's upswing is no surprise.

But the biggest problem with guaranteeing principal and returns this way is that it turns asset management into lending. And unlike lending, or ordinary diversified asset allocation, a bank's capital pool does not have to publish its net value or its returns. That makes it shadow banking.

How the funding side operates at the top also shapes how the asset side is designed and valued. The reason is simple: there was more money on the funding side than there were good assets to be found on the asset side.

2012 and 2013 were exactly when China's economic structure was shifting. New sectors led by mobile internet and new energy were rising, and many companies in them needed money. But such companies tend to look the same on paper: explosive growth in revenue, historical net profit that is very low or negative. On the A-share market they fell far short of listing standards, and they could not go public through a backdoor listing either. Meanwhile they faced fierce competition and needed financing urgently.

Broadly, a company reaches the A-share market by one of four routes: an IPO, a backdoor listing, an ordinary merger-and-acquisition restructuring, or a "quasi-backdoor listing." Between 2012 and 2016, IPO timetables were full of uncertainty, backdoor listings gradually caught fire, ordinary M&A boomed, and "quasi-backdoor listings" kept appearing.

In those years, commercial banks looked past their traditional M&A lending business and paid more and more attention to the opportunities inside M&A funds. Besides putting their own proprietary capital into that business, they also steered wealth management money into M&A funds. At the same time, M&A restructuring and "quasi-backdoor listings" had become the fastest way onto the A-share market.

The two sides clicked instantly, and the M&A market became the main meeting point for capital and assets. But the premise of "guaranteed principal and returns" produced severe structural distortions, giving rise to large amounts of debt dressed up as equity. Through the structured design of asset management products, what were in substance debt claims entered equity-investment channels in the form of senior tranches, safety cushions, equity-linked debt or convertible bonds.

What is the biggest problem with "guaranteed principal and returns"?

The way bank wealth management products originally took part in M&A funds, the gray areas built into the system, and the consequences that followed can all be shown with one simple example.

Say a listed company has a market value of 10 billion yuan (10 yuan a share, 1 billion shares) and plans to buy a target valued at 3 billion yuan. The seller will accept only cash.

So the listed company's major shareholder plans to buy the target first. Later the listed company acquires the target from the major shareholder by issuing shares to pay for the assets, and the major shareholder is paid in stock.

But the major shareholder holds only 25%, worth 2.5 billion yuan. Even pledging those shares for financing, plus the shareholder's own money, does not cover a 3-billion-yuan acquisition. So the major shareholder decides to set up an M&A fund and bring in outside investors at the fund level.

The major shareholder goes to a private fund manager, who can privately solicit qualified investors willing to join the deal.

The manager quickly finds one: an investor with deep pockets, a fairly high tolerance for risk and a good grasp of how the M&A market works, who likes the deal. But this investor's money is still not enough.

So the manager also brings in a commercial bank. The bank thinks the deal is good quality. But its investment style is conservative, and it wants a solid guarantee on its return.

At this point the M&A fund can be structured like this:

The M&A fund brings in a limited-partnership private fund that raises 2.5 billion yuan in total. As a result:

(1) The major shareholder's contribution shrinks to 500 million yuan, easing the funding pressure sharply;

(2) A brokerage, acting for an asset management plan, subscribes for the private fund's senior tranche, which solves the business-registration problem. The wealth management product involved is worth 2 billion yuan, with an expected return of 5%.

(3) The subordinate LP puts in 500 million yuan for the private fund's subordinate tranche. At the private fund level, leverage is 1:4. Under the partnership agreement, once the senior LP has earned an 8% return it takes no further share of the profit, and the subordinate LP promises to make up any shortfall owed to the senior LP.

The listed company then issues shares at 10 yuan each, for a total of 3 billion yuan, to buy the M&A fund and with it, indirectly, the target. The major shareholder receives 50 million new shares and its stake rises. The private fund receives 250 million shares, locked up for 36 months.

Bringing a private fund into an M&A fund this way is exactly what is meant by "equity in form, debt in substance." Bank wealth management money, invested in the private fund indirectly through an asset management plan, is nominally an equity investment and actually a debt investment. When the lock-up ends, even if the listed company's shares have fallen below the issue price, the bank wealth management product still earns its steady 8%.

Because a listed company's private placement carries a lock-up of at least three years, a bank can use the capital pool to bridge the mismatch in maturities. But it can also take part without the pool.

The method is to put a one-year wealth management product into this three-year project, with an agreement that after one year the investment will exit by selling the private fund shares, and that if the sale does not deliver an 8% return, the other capital providers will make up the difference. After a year, then, the asset management plan sells the private fund shares to a new investor and the wealth management product exits.

But the bank still owes its wealth management investors an implicit duty of rigid redemption, so the way the product is designed pushes the risk down the chain, step by step, until it reaches the listed company's major shareholder.

And to lift the returns, the best thing to do is drive up the listed company's share price and market value.

So the capital market from 2013 through the first half of 2016 looked like this: money everywhere, huge financing demand from new industries, a clogged IPO pipeline, wide arbitrage margins. The M&A market took off overnight.

In 2013, the year bank wealth management money plugged into asset management products, A-share M&A deals were worth 293.1 billion yuan. By 2015 the whole M&A market had grown to 1.6485 trillion yuan. In the first half of 2016 alone, deal value reached 1.2363 trillion yuan.

The market moved through one so-called "hot sector" after another: mobile games, film and television, internet finance, online education, electric cars. Valuation and growth went from being judged on six months of earnings, to a discount on three years of growth, to a ten-year vision. As long as a stock carried the concept of an acquisition, or the expectation of one, its price went up.

But there is a limit to how far capital, leverage and credit can be stacked on top of one another. So when the CSRC started cracking down on margin financing and leverage in 2015 — aiming only at the secondary market — it set off a butterfly effect. Whether or not a listed company was doing deals, whether or not it was working the capital markets, share prices collapsed across the board.

The more serious damage, of course, went well beyond a volatile stock market.

Wealth management products that had turned into a lending business did, to be fair, support the real economy for years. But their very existence raised the cost of financing for the real economy and worked against innovation-driven growth. Nor can we ignore what their expected rates of return did to the public's sense of what a risk-free return looks like.

What good does regulating asset management do?

The original point of asset management, to use an old phrase, is to "be entrusted by others and manage money on their behalf." Broadly, a growing asset management industry serves two ends: it meets households' need to allocate their assets and gives them more places to invest, and it channels money into corporate bonds, unlisted equity and other assets, meeting companies' financing needs and supporting the real economy.

Over the past several years, though, both investors and the industry lost their way.

A manager running an asset management product is supposed to work to create the most value it can. An investor buying one is supposed to understand that the gains and the losses are theirs. Most important of all, a manager cannot guarantee that an investment will make money.

In a rational market, if an investment loses money, investors can stop buying that manager's products; or, if they find the manager made a mistake or broke the rules, they can sue.

Only when investors carry their own gains and losses, and managers invest in investors' best interests, will the market weed out the weak and leave behind good asset managers who can give good companies real financing support.

In practice, though, investors often behave as if they had put their money into a time deposit at a better interest rate, and expect the manager to repay principal and expected return on schedule. Managers, for their part, wrongly drift into running a bank's lending business — treating investors' money as a disguised liability, folding it into a single capital pool, and making decisions without regard for investors' tolerance for risk or their right to know.

The new asset management rules set the direction of Chinese financial regulation for the next five to ten years: no more stacking wealth management products in multiple layers, and an end to rigid redemption — no more "guaranteed principal and guaranteed returns."

The rules make clear that asset management products may be nested only one layer deep; if the underlying product invests in another asset management product, that product must be a publicly offered securities investment fund.

They also give a great deal of space to breaking rigid redemption, and put it front and center.

Breaking rigid redemption changes both the industry and the ordinary investor at the root. It is a systematic correction of a way of thinking. Once rigid redemption is gone, an investor whose investment loses money bears the loss. In return, when an investment does well, the manager can no longer keep the part of the return that exceeds the "expected rate of return."

Cover image credit: Steve Johnson on Unsplash