China and the Monetary System

I had a discussion with the "China expert" for Peter Thiel's hedge fund years ago; I argued that China would significantly underperform coming out of Covid lockdowns (which turned out to be the case). He strongly disagreed and stated that he didn't understand the topics I was referring to and that there was an "asymmetry in knowledge." I thus want to present the indicators and phenomena I was looking at to illustrate why it is so important to understand the "financial plumbing" in order to forecast macro outcomes. We will build to understanding all this material.

---

“What’s referred to as a ‘global dollar shortage’ is bound to get worse in the future... none of this is about a shortage of dollars per se, but rather a shortage of balance sheet to intermediate dollars” ~ Zoltan Pozsar (2016)

Introduction

On March 15th, 2023, Goldman Sachs raised China’s GDP forecast from 5.5% to 6%. Expectations for a strong rebound prompted the firm, like most others on Wall Street, to call for CNY/USD to fluctuate between 6.7-6.5 throughout the year. In the subsequent months, Goldman revised its GDP forecast downwards and the yuan hit 7.25 against the dollar.

In Goldman’s 2023 Outlook for China the firm did not mention the word “collateral” once; the condition of dollar funding markets and an exposition of how dollar funding could affect China was also omitted.

It is of particular irony that Goldman raised its GDP forecast for China immediately following the collapse of SVB. This drives home the point that even the most prestigious financial institutions are decades behind in understanding the monetary system.

The following analysis aims to elucidate why the author called for an underperformance of China’s economy and a surge in the dollar vs. yuan utilizing a financial plumbing centric framework. Most ex-post analysis of the yuan’s underperformance has focused on interest rate differentials. The analysis in this note will focus on the effects of dollar funding markets on the yuan to provide a different perspective.

Money Markets and the Yuan: A 2018/19 Case Study

Let’s begin with a brief case study in order to illustrate how financial plumbing stress has affected China in the past.

The first image illustrates that the dollar weakened vs. the yuan throughout 2017 but then spiked in value immediately following May of 2018 and April of 2019 (burning heavily skewed dollar longs in 2017 and shorts in 2018 from traders too focused on policy rates).

The following three images illustrate the stress in unsecured money markets during these periods.

The remaining images illustrate the conditions of collateral markets.

*Note that the Federal Funds market has both arbitrage and dollar funding components to it. Arbitrage occurs at the lower percentile rates, while the dollar funding needs of non-US banks are met by the higher percentile rates. The arbitrage activities turn off at month ends for regulatory reasons, thus drops in EFF volume at month ends illustrate the amount of Fed Funds activity being driven by arbitrage vs. dollar funding. Note that prior to SVB, EFF volume actually rose on month end, illustrating dollar funding stress.

Note that the first bank takeover in China in decades- Basoshang Bank – occurred in May of 2019. While a detailed analysis of this event is outside the scope of this note, the important takeaway is that the liquidity “tide” is of pivotal importance in China as it is in the US. When there are bank failures, the blame will almost always be placed on the institutions themselves- SVB’s treasury holdings and poor interest rate risk-management, Baoshang’s ownership, etc... There is always an element of truth in these claims, but they often miss the forest for the tress.

The more important analytical focus- as the previous note on the March banking crisis evidences- is the ‘tide’, not those who have been swimming naked.

China and the Global Dollar System

Having illustrated that there is a correlation between the yuan and money markets, let’s now turn to the causal mechanisms at play.

The PBOC and Chinese financial institutions lack connections and credibility in the global dollar system. This means that Chinese banks are at the “outer- rim” of dollar funding markets.

Being at the outer-rim doesn’t matter much as long as your country is isolated; however, China is intimately intertwined with the global economy. This puts the country in a unique position: no other country’s fate is so tied to conditions in the dollar system.

Unfortunately for China, post-GFC the dollar system has been functioning poorly (primarily the result of balance sheet inelasticity and risk-aversion on the part of dealer banks). In large part due to this, China’s growth has slowed drastically over the past decade and a half.

But let’s back up a bit- What exactly does it mean to be at the “outer-rim” of funding markets?

It means that Chinese banks cannot access wholesale funding in the same way as, say, a French or Japanese bank. For example, they cannot utilize FHLB advances, repos from MMFs, or interbank lending from American banks. These are the cheapest sources of funding and they are by and large unavailable to Chinese banks.

Compounding this problem is the fact that, post-GFC, Chinese banks have increasingly desired to “go global”, expanding greatly their USD denominated assets (thus creating a need for dollars).

So, what options are available to these banks?

One option is turning towards your more reputationally pristine neighbors- Japanese banks- who can intermediate dollars on your behalf.

However, these banks began retrenching from the dollar intermediation business in 2017-

Another option is to issue offshore USD denominated debt.

However, the same reputational problem affects Chinese entities in these markets too, meaning that only the largest (state-backed explicitly or implicitly) entities are able to issue debt offshore. Smaller Chinese banks are thus reliant on the larger banks to redistribute dollars to them.

In fact, it is not just the large banks that are in this dollar redistribution business, but also large non-financial corporations that are able to tap offshore dollar markets and act as surrogate financial intermediaries.

This gives us some insight into the vulnerability of the country to strains in wholesale dollar funding markets. *Note that small and medium sized banks are crucial for the funding of the private sector and NBFIs in China.

Let’s run through a brief example:

Dollars may start from an American financial institution who lends via o/n repo to a dealer bank in France, who lends term to a bank in Hong Kong, who buys a Chinese corporate USD offshore debt issuance. This corporate may then lend to a large domestic bank, who in turn lends to a smaller regional bank, who then lends to a small / medium sized Chinese company (note that collateral is involved in every link of this chain).

Any strain in dollar funding markets gets passed along, affecting the last legs of this chain the most.

FX Swaps

Similar to their Japanese neighbors, Chinese banks have large reservoirs of domestic currency deposits: thus, another option for these banks to obtain dollars is via FX swaps (again requiring the intermediation of their more reputationally pristine regional neighbors).

Taking up less balance sheet space (they are not completely balance sheet neutral as is often claimed), FX swaps have become the ‘new repo’ post-GFC: conceptualize the yuan as being ‘collateral’ for dollars in FX swap transactions.

The BIS estimates that the use of FX swaps by Chinese entities has doubled in size every three years since 2007 and that the dollar is on one side of 99% of Chinese swaps (the global average is 90%).

Here we see what the BIS’ $65 trillion in off-balance sheet FX swap debt is all about... ‘funding’, not necessarily ‘hedging’.

But the FX swap market is significantly affected by conditions in dollar funding markets- stress in wholesale markets necessitates more ‘yuan’ collateral to be posted in order for Chinese institutions to obtain USD loans (putting pressure on spot rates).

FX swaps are intermediated via the balance sheets of dealer banks and these balance sheets have become structurally much more expensive post-Basel III- limits were effectively put in place on the size of balance sheets, thus making low-margin, high volume intermediation (repo and FX swaps) much less attractive.

Prime money market fund reform in late 2016 also made dollar funding significantly more strained: this reform did not necessarily affect the aggregate flow of dollars, but rather affected the veins through which dollars flow. Specifically, foreign banks lost hundreds of billions of dollars in unsecured funding; this funding was redirected to American dealer banks via FHLBs and repo via government MMFs.

Below is a chart illustrating the shifting of funds from prime to government MMFs (the blue bars are government MMFs)-

Assets must be unencumbered to qualify as HQLA, thus a decrease in unsecured lending from prime funds and an increase in repo intermediated lending from government funds to dealers increases the marginal cost and therefore price of dollar swaps.

American banks have access to deep pools of retail dollar deposits, while non- American banks rely on wholesale markets for dollar funding; this wholesale funding became structurally more expensive post MMF reform. As such, global American banks play an increasingly crucial role in setting the marginal price of dollar funding. Retrenchment on the part of American dealers causes dollars to be priced via the now more expensive footing of non-American dealers who fund via MMFs. *Expensive in terms of pricing (repo vs. retail deposits), utilization of balance sheet space (repo takes up space, retail deposits are “gold” from a regulatory perspective), and balance sheet cost (the cost of utilizing the balance sheets of American banks is higher for regulatory reasons).

So, what does this entail for China? It has become even more expensive to obtain dollars, and the country has become much more dependent on conditions in the American banking system.

The PBoC’s Balance Sheet

If dollars are more difficult and expensive to come by, one option- of course – is for China’s central bank to direct its FX reserves to the banking system.

However, this comes at a cost-

The counterpart of FX reserve assets on the PBoC’s balance sheet is the yuan denominated reserves of Chinese banks. Thus, supplying dollar reserves means (ceteris paribus) a decline in yuan reserves. A decline in yuan reserves can then lead to volatility in money markets, which is what the following image (copied from above) illustrates-

China’s government is often discussed in terms of having complete control over everything that happens within the country. Thus, increased volatility in markets like Shibor is often discussed in terms of Chinese authorities “allowing” volatility in unsecured markets in order to reduce leverage, risk- taking etc. Similarly, a weakening yuan is often discussed in terms of Chinese authorities allowing depreciation in order to spur economic growth.

However, this mindset often causes one to miss the bigger picture: a depreciating yuan and volatility in money markets are often the effects of stress in the dollar system, not the deliberate intentions of Chinese authorities.

Take a look at the image below-

There was a huge injection of yuan liquidity at the end of February / early March of this year (note that China’s reverse repo is the opposite of the Fed’s RRP, meaning that this is the government accepting collateral and making yuan loans). Many consider increases in reverse repo volume or decreases in the RRR as being “stimulus” for China’s economy.

In the author’s humble opinion, this is incorrect. As was astutely pointed out by a Financial Times writer a decade ago, these liquidity injections are simply making up for liquidity outflows- i.e., it is the PBoC trying to get back to a neutral position.

In short, decreases in the RRR (which theoretically “allows” Chinese banks to do more with less reserves) and increases in reverse repo volume are responses to the same thing- a shortage of dollars- which causes the PBoC to supply dollars, reducing yuan reserves in the domestic banking system.

Note the close correlations of the following three images-

The first image is of swap spreads, the second is of the yuan / USD exchange rate (inverted, so ‘down’ means a stronger dollar), and the third is of China’s reserve requirement ratio.

Note that the yuan strengthens in 2013 and 2017, during which the RRR is held steady and swap spreads move towards positive territory; the yuan weakens from 2014-16 and 2018-19 as the RRR is reduced and swap spreads become more negative.

Negative swap spreads can proxy for balance sheet conditions (it takes balance sheet space to arb away spreads), and because dollar funding is intimately connected to balance sheet space, more negative swap spreads can proxy for more difficult funding conditions. As such, the yuan weakens vs. the dollar during periods of negative swap spreads and the RRR is cut as the PBoC must offset its internal yuan reserve drain caused by its supplying of dollars (draining FX assets) to domestic entities.

In this way, China has a trilemma: the objectives of supporting the yuan, stimulating the economy, and maintaining money market stability often run against one another (with the actions of achieving the first running against the latter two goals).

Lastly, it is important to appreciate that China has a bank-dominated monetary system and that smaller banks and the shadow banking system are heavily reliant on secured funding chains stemming from the largest banks. As such, repo markets and collateral play an enormous role in China (this is true in the ‘informal sector’ as well, with trillions of dollars of repo lending occurring via unofficial ‘daichi’ agreements).

Notably, asset managers are some of the largest repo borrowers, and Chinese government bonds are considered pristine collateral and are utilized commonly for repo funding.

As such, note that there is a strong correlation between the Hang Seng Index and Chinese government bonds (lower yields due in large part from increased demand for repo funding), and that both of these indicators were flashing warning signs at the turn of this year-

Breaking down this relationship and providing an exposition of China’s repo markets is outside the scope of this note (26 pages is probably long enough!), but the important takeaway is that collateral matters as much, if not more in China, and that the intertwining of China’s official banking sector with its shadow banking sector makes money markets particularly important. These markets and China’s economic well-being are intimately affected by conditions in the global dollar system, which now more than ever relies on the health and risk-appetite of the largest American financial institutions.

Conclusion

To sum up the logic behind the author’s calls...

It was evident leading up to March that there was a dire scarcity of collateral and stress in dollar funding markets. It was likely that March was when this stress would manifest, given this is one of the annual low points in liquidity. It was the author’s belief that stress in the American banking system would affect China the most and that the yuan would significantly underperform relative to expectations due to strain at the furthest ends of dollar funding chains.

*Note the author believed that European currencies would be spared in the near term given the influx of MMF assets / FHLB advances but that the next great trade opportunity would be a long dollar position vs. these currencies in the latter half of this year.

The analysis provided in these two notes illustrates the general contours of the author’s macro framework, utilizing simple publicly available data.

Hopefully readers found this analysis to be interesting and beneficial- please contact the author if more details are desired!