Gorman’s points were correct in a narrow sense, but missed a deeper reality. I told a story of an encounter with legendary Citibank CEO Walter Wriston in 1981, at a time when confidence in the U.S. dollar was waning and oil-rich Arabs were rumored to be pulling petrodollars from U.S. banks, including Citi. Wriston explained to me the banking system was a closed circuit. Arabs could pull deposits from banks and buy other assets, including gold, but the sellers of those assets would put the money on deposit with another bank, which lent it back to the original bank in the interbank deposit market. There were small costs in terms of rates and fees, yet the money ended up where it started. Wriston told me, “Banks don’t need capital; they just need to borrow from other banks.”
The truth is, banks don’t need capital when markets are calm, interbank liquidity is readily available, and collateral is well-bid. Yet the opposite is true. In a panic, when liquidity dries up, when assets go no-bid, and when everyone wants his money back, no amount of capital is enough. Banks, even good banks, are leveraged and it only takes modest declines in asset values on a leveraged balance sheet to wipe out capital. My point to Gorman was that his improved capital cushion was more than enough for most market conditions, but far from enough for a replay of 2008 on a larger scale.
Kelleher had been biding his time and now pounced. He repeated Gorman’s points about capital adequacy and risk, but went further. He asserted that derivatives exposures had declined significantly (I asserted the opposite). Even under my complexity analysis, diminution in derivatives notional amounts should result in a nonlinear reduction in risk that made the system safer. Kelleher was tough, but his analysis was expert.
The gross notional value of all derivatives held by banks, typically off-balance sheet and disclosed only in footnotes, had declined since 2008 as Kelleher asserted, according to comprehensive statistics reported by the BIS. This was because the banks assigned swaps and other derivatives to centralized clearinghouses, which net out offsetting exposures as urged by the G20 leaders in September 2009. When clearinghouse footings are included, the gross notional value of derivatives has increased since 2008, consistent with my claim. Clearinghouses provide transparency and netting, and in a panic, problems are spotted more readily. Still, derivatives risks have not disappeared; they have simply been moved from banks to clearinghouses, which raises the question of their capital adequacy. What happens if a major clearinghouse participant is in financial distress and cannot perform its obligations? It’s not comforting to know that the major clearinghouses, including CME, ICE, and the London Clearing House are using the same flawed VaR and stress-test risk-management methods that missed the coming catastrophe in 2008. Mutualization of losses among all clearinghouse members is the prescribed remedy when a member defaults, but this simply acts as a conduit for contagion. It’s as if a patient with an infectious deadly disease escaped from quarantine and spent the day at Starbucks. As with AIG in 2008, net exposure rapidly morphs into gross exposure once counterparty performance is in doubt.
Gorman’s and Kelleher’s critiques suffered from what Keynes called the fallacy of composition; that components of a system can be added up to describe the system as a whole. This is not true in complex systems, where scaling metrics means emergent properties come out of nowhere and cannot be inferred from perfect knowledge of the system’s parts.
I leaned forward, looked straight at Kelleher, and said, “Look, Colm, you’re right. Morgan Stanley is safer. Still, the system is not. Morgan Stanley is part of the system; when the system collapses, Morgan Stanley goes with it. It’s not enough to look at your own balance sheet; you have to look at the global balance sheet. It’s all connected.” Of course, there was no need to single out Morgan Stanley. The same could be said of all the big banks. Their balance sheets were safer in a narrow sense. Yet they were more vulnerable than ever to systemic risk.
It was almost time to head back to the bar. Jami Miscik asked the smartest question of the evening: “What would you do if you were in my position as a director of Morgan Stanley?” I said I would work with colleagues to break up the big banks, including Morgan Stanley, ban most derivatives, and adopt Bayesian statistics and complexity theory as new risk-management tools. Like the CIA veteran she was, Miscik showed no reaction. Still, she clearly understood every point.
I sympathized with the reality that a bank director has a duty to that bank and its stockholders, not to the system as a whole. Systemic risk is more the purview of central banks, finance ministers, and the IMF. Still, directors and CEOs are powerful. If they urged a risk-reducing agenda on government policymakers, the policymakers might listen. There’s no evidence this is happening. Inertia rules.
Glaciers
In writing and public speaking, I frequently use a snowflake-avalanche metaphor to describe complex system dynamics and the way systems collapse. Capital markets are complex systems, but their workings are incomprehensible to everyday citizens. Attentive investors know if the Dow Jones index went up or down or if their 401(k) balance showed a gain or loss from the last monthly statement, but that’s all. People are busy; unless they’re finance professionals, there’s no reason they should know more than the latest prices. When one explains how complex systems work and why capital markets are vulnerable to total breakdown, people are intrigued, yet their eyes glaze over at the mention of density functions, power curves, and hypersynchronicity. That’s understandable and is why the avalanche metaphor is useful. It begins with the buildup of an unstable snowpack on a mountainside. Billions of individual snowflakes form an interconnected lattice. A new snowflake falls, it hits the snowpack in a way that shakes loose a few other snowflakes, those snowflakes begin to slide, the slide gains momentum, and soon the entire snowpack rips loose from the mountain and buries the village below. The image is vivid and more than a metaphor; the math and dynamics behind the avalanche are exactly the same as a capital markets collapse, adjusted for the idiosyncrasies of snowflakes versus live traders. Breaking up unstable snowpacks with dynamite to reduce danger has the same rationale as breaking up big banks.
Yet an avalanche is not the only metaphor that might apply to capital markets. In some cases, a glacier is a better way to describe the economic processes that drive securities prices, exchange rates, and interest rates. A glacier is a complex dynamic system, yet glaciers move more slowly and cause more lasting change than an avalanche. Glaciers gouge valleys, move boulders, and push obstacles aside with ease.
While glaciers are reputed to move slowly, they can also surge. In 1956, the Muldrow Glacier near Denali in the Alaska Range surged at a pace of fifteen hundred feet per day. Surface ice levels dropped three hundred feet as the ice moved rapidly to lower elevations. The glacier moved four miles down the mountain by the end of the surge.
Events shaking capital markets today are better likened to glaciers than avalanches. They are unrelenting and slow, yet sometimes produce dramatic surges. They are less dramatic than avalanches in the short run, but more destructive in the end. Some of the glaciers grinding down the system today are Chinese debt, trade wars, Fed monetary finesse, and an emerging-markets debt debacle. There are other threats, but these are among the greatest. What follows is a précis of some glaciers gaining ground to push capital markets into a new Ice Age and leave us in the aftermath.
China Is Madoff
1 percent, 1 percent, 1 percent, 1 percent, 1 percent …
That’s a close approximation of the time series of monthly returns reported by Bernie Madoff over the twenty years he ran his wealth management business.
When you gain 1 percent per month, that compounds to 12.7 percent per year, year after year. That return more than doubles your money in six years, and doubles it again in another six years. After eighteen years, about the time between the birth of a child and when she goes off to college, you would have made eight times your money with Madoff. One million dollars invested with Madoff in 1990 would have been worth $8 million by 2008.
There was only one problem. It was all a fraud. There was no pool of investable assets. There were no above-average returns, no compounding, and no profits. It was all fake accounting and looting by Madoff. Sometimes new money was used to cash out old money that wanted to redeem, but most of the money stayed in. Madoff’s Ponzi collapsed in 2008.
The amount lost in the Madoff fraud varies depending on one’s calculation method. If you use the amount the investors believed they had, even though the account statements were bogus, the losses were about $65 billion. If you use the amount of invested money that was lost without counting fake profits, the loss was $17 billion. Either way, Madoff set the record for the biggest Ponzi scheme in history.
The fraud was discovered in conjunction with the financial panic of 2008. Global investors were losing money in stocks, mortgages, derivatives, and other asset classes. Leveraged investors were getting hit with margin calls. Money market funds and banks experienced runs as investors tried to get their money back any way they could. It was the worst global liquidity crisis in history.
In this panicked environment, Madoff’s investors knew they could count on Bernie as a liquidity source. They began to make redemptions from Madoff’s fund. That’s when the Ponzi unraveled. Madoff didn’t have the funds to meet the redemptions, he began to default, rumors spread, the SEC and FBI moved in, and the rest is history. On June 29, 2009, Bernie Madoff was sentenced in federal court to 150 years in jail.
Readers familiar with the Madoff story may also know there were numerous suspicions and warning signs as early as the mid-1990s that Madoff might be running a Ponzi. These warnings were never properly investigated by the SEC or other agencies.
The most famous warnings were given by forensic analyst Harry Markopolos. What first tipped Markopolos off to the fact that Madoff might be a fraud? It was those steady returns, the 1 percent, 1 percent, 1 percent month after month, year after year. A graph of Madoff’s returns over time rose at a near perfect 45-degree angle. Markopolos knew it’s impossible to produce those returns in finance.
It is possible to produce positive returns on an annual basis over long periods of time. Some of the best hedge-fund managers have done it, although most have not. But even superstar hedge-fund managers have a bad month or a bad year now and then. And the positive years are not all the same. You might be up 10 percent one year, 25 percent the next year, then down 3 percent, and up 7 percent in year four. That’s a pretty good track record, but it’s not repetitive and it doesn’t move in a straight line.
The technical name for a time series of returns similar to what Madoff was reporting is serial correlation or autocorrelation. This happens when a signal contains a feedback function that causes it to produce the same signal over and over, sometimes with amplification. Serial correlation exists in physics, mathematics, and acoustics, but it does not exist naturally in finance. Markets are complex dynamic systems with emergent properties that disrupt the steady feedback needed to produce serial correlation. The fact that Madoff reported returns exhibiting autocorrelation was a dead giveaway to Markopolos. Unfortunately, the SEC did not understand what Markopolos was saying.
Here’s another time series of economic returns: 1.8 percent, 1.7 percent, 1.5 percent, 1.8 percent, 1.8 percent, 1.6 percent, 1.4 percent, 1.8 percent. That’s the time series of quarterly growth in China’s GDP from the second quarter of 2016 to the second quarter of 2018. It’s not as smooth as Madoff’s returns, but it’s close. It’s also impossible. China can only produce those returns by cooking the books, the same as Madoff. China reports steady, positive returns quarter after quarter, like clockwork. Those numbers aren’t real; they’re manufactured to appease gullible investors, policymakers, and the media.
What does a real economy look like? Here’s the annualized U.S. growth rate for GDP for the same eight quarters as the China example: 1.9 percent, 1.8 percent, 1.8 percent, 3.0 percent, 2.8 percent, 2.3 percent, 2.2 percent, 4.1 percent. Notice that the U.S. growth rates exhibit far more variance than China’s from a high of 4.1 percent to a low of 1.8 percent. Notice that weak quarters, like 1.8 percent, are adjacent to strong quarters, like 3.0 percent.
If you take the time series back even further you discover that China has not had a negative quarter in over five years, while the United States has. In short, the U.S. data exhibits the mix of weak, strong, and negative quarterly data that one expects from a complex economy, while China exhibits the autocorrelation that one expects from a financial fraud.
There is no question that China is manipulating its growth data. Real growth in China is closer to 5.5 percent per year than the 6.8 percent per year China claims. Growth is even lower once wasted investment is stripped out. The policy question is why China feels compelled both to lie about the data and present it as an improbable autocorrelated time series.
The reason is that China is a Ponzi like Madoff. China has trillions of dollars in external dollar-denominated debt, wealth management products, bank loans, intercompany loans, and other financially engineered arrangements that can never be repaid. If everyone with a claim on China wanted her money back, China couldn’t come close to satisfying even a small portion of those seeking liquidity.
This doesn’t mean China does not have a real economy. It does. It’s just that the real economy is tangled in a web of leverage, unpayable debt, bogus accounting, and the vain hope that Communist Party leadership can keep a lid on dissent until the global economy improves.
That’s not happening. The global economy is sinking into trade wars, currency wars, and fights over intangibles such as intellectual property. Shooting wars in the South China Sea, Taiwan Strait, Korea, and the Middle East may not be far behind.
China cannot win a trade war because it exports far more than it imports, especially on a bilateral basis with the United States. Trump wants the U.S.-China bilateral trade deficit reduced by several hundred billion dollars. China cannot easily do that without hurting its economy, so the trade wars will drag on and get worse.
China does have one financial weapon it can use to alleviate the pressure from the trade wars: currency devaluation. China has about $3 trillion in reserves. About $1 trillion is illiquid; invested in hedge funds, private equity, and other alternative assets that cannot easily be redeemed. Another $1 trillion is held as a precautionary reserve to bail out the banking system when the time comes. That leaves only $1 trillion to defend the currency peg with the dollar. It’s not enough. In 2016, China used up $1 trillion in reserves defending its currency. China was losing reserves at the rate of $80 billion per month at one point. It would have been broke by the end of 2017 if it had not closed its capital account and trapped the reserves inside China.
By devaluing its currency, China can take pressure off the capital outflows, buy time, import inflation to reduce the value of local currency debts, and make its exports more attractive. Devaluation is a simple solution to China’s financial imbalances. Imagine if Madoff had been able to “devalue” his liabilities to investors. He might still be in business. China will still be in business a century from now, but that doesn’t mean there won’t be enormous investor losses and global economic disruptions along the way.
China’s risks go far beyond liquidity and exchange rates. It is now reaping the bitter fruit of its one-child policy of the 1980s, 1990s, and early twenty-first century. The ban on two children, sometimes enforced by drowning newborn girls in bedside buckets, has left China with a rapidly aging population and insufficient younger workers to maintain growth or provide benefits for retirees. Relaxing the policy as China has done recently will not have an impact on workforce participation or productivity for another twenty years. Labor-force participation and productivity are all there is to economic growth. China has shortsightedly truncated its labor force and lags badly in productivity. Apart from borrowed money, wasted infrastructure investment, and fictitious accounting, there is no Chinese economic growth miracle. In short, China is growing old before it grows rich. In the en
d, it is just another emerging-markets economy stuck in what the IMF calls the middle-income trap with no easy way out.
These economic and demographic headwinds come on top of an increasingly confrontational geopolitical relationship between China and the United States. This confrontation, referred to by experts as the “gray rhino,” is summed up by University of Hong Kong scholar Andrew Sheng:
In addition to structural and cyclical risks, China must address the “gray rhino” (highly likely, but often ignored) strategic risks arising from the intensifying Sino-American geopolitical rivalry.2 Here, the emerging trade war is just the tip of the iceberg. The U.S. and China are set to become immersed in a long-term competition for technological and strategic supremacy. To stay ahead, they will use every kind of leverage and instrument at their disposal. If this competition is left unchecked, it will surely have far-reaching spillover effects.
Introduction of the geopolitical struggle is critical to the analysis because it marks a change from the era of globalization, when economic growth trumped all other policy considerations. Wars are not free, even cold wars, and if the price of containing Chinese ambition is slower growth, that is a price the United States is prepared to pay to protect its intellectual property and national security. This is a rude awakening for younger bankers and scholars who have known only a golden age of globalization (1989–2017). More senior analysts acquainted with the first Cold War (1947–89) will find the elevation of geopolitics over growth to be familiar ground.
Trade Tango
In May 2018, a high-level delegation of Trump administration officials traveled to Beijing for a critical round of negotiations intended to avoid an all-out trade war between the two largest economies on the planet.
Aftermath Page 28