Risks on Investing Investment Banks
Since year 2008 is not a distant past, I believe many people still remember the time when the biggest investment banks were in deep trouble. In fact, two of the five biggest investment banks were killed, with the rest either getting acquired, or on the verge of collapse.
Among the five investment banks, Goldman Sachs (NYSE:GS) was probably the biggest and strongest. Even so, it still needed to get $5 billion cash injection from Warren Buffett to survive. It is more than just the cash though, but also a re-establishment of confidence given Buffett's reputation. As I mentioned in my last article about Leucadia (NYSE:LUK), the profitability was rarely a problem for investment banks. Even in 2008, profitability was not really the concern. Instead, it was the liquidity crisis and credit crisis that was killing these investment banks. Even for the strongest of all and very favorable terms, Buffett still wouldn't invest that $5 billion without betting on the help from government. As he said during the time, if he wasn't sure that government had to step in to help the market and economy, his bet on Goldman would be a mistake.
Since Leucadia is a conglomerate with well diversified business lines, there is much less down side risk comparing to many other companies, except that Jefferies is the biggest subsidiary (50% of Leucadia) and it is an investment bank. Therefore, it is very important to understand the risk behind Jefferies in order to have an educated guess on the worst case scenario for Leucadia.
Leucadia is generally conservative on the liquidity side. With diversified businesses in its holdings, one would think they don't need much liquidity on the parent company level, but still they got almost $3 billion liquidity at the parent company (this amount may have been reduced to $2.2 billion after several new investments in 2014). On the other hand, at the parent company level, they have $456 million debt due in 2015, and $1 billion debt due in 2023. People may question whether this kind of long-term financing is needed, since they still have a lot of cash on hand, and so many other companies are using short-term credit facility for their financing needs.
I think this shows the conservatism principle and long-term view of Leucadia, and may also reflect their expectations on interest rate trends.
At the Jefferies level, as shown below, there is $6.6 Billion long-term debt with an average maturity of more than 8 years. (Note that these debts are non-recourse to the parent company, which means the parent company doesn't have to be responsible for them if the subsidiary is not able to pay them back.)
From the 2013 10-K, we can see that Jefferies has $16.7 billion trading assets, and $7.3 billion trading liabilities. Adding these two together, this is $24 billion. (For sure, much of the trading assets are hedged by the liabilities, but to be safe, I will just add them instead of subtracting them.) Given the $4 billion equity (after removing $1.3 billion goodwill) in Jefferies, $8 billion long-term debt on the parent company and Jefferies, and the extra $5 billion equity on the parent level, the total capital cushion is $17 billion which is about 1:1 ratio to the trading asset and liabilities.
Since the trading asset and liabilities are usually hedged against each other, the net exposure to some market risk or macro risk is usually much less. This is seen from the more detailed disclosure on European exposures. For example, Jefferies' long debt securities in Italy is $1.18 billion, short debt securities is $1.01 billion, and long derivative notional exposure is $74 million. Adding these together, if there is a credit crisis in Italy, the net exposure is only about $250 million. Similarly, exposure in Spain is $200 million, and $130 million for Puerto Rico. Also, if we assume the credit crisis happens in the entire Europe, not just one country, the total net long exposure in the more risky European countries (including Greece, Italy, Spain, Ireland, and Portugal) is $483 million; and the total net sovereign debt securities in these countries is $194 million.
This basically shows that the exposure to Europe is less of a concern, and it is completely wrong to just mention the long side of the exposure, as the short attacks did in 2011.
Compare Derivatives Exposure with Goldman Sachs
In my opinion, when talking about risks, liquidity is just one side of the equation. The other side is the potential losses. Even with a lot of capital, if the loss mounts to an even bigger number, there is still no safety and confidence.
The following table summarizes the derivative exposure for Jefferies and Goldman Sachs.
|(all numbers in millions)||Jefferies||Goldman|
|Derivatives (asset + liabilities)||$5,000||$1,640,000|
|Derivatives after counter-party netting and cash/security collateral (asset + liabilities)||$441||$80,000|
|Credit Derivatives (asset + liabilities)||$110||$56,000|
|Level 3 asset + level 3 liabilities||$488||$40,000|
|Level 3 derivatives||$10||$7,000|
|Equity (after removing most of the goodwill)||$4,000||$79,000|
|Net Credit Derivatives (asset - liabilities)||$7||$16,000|
When we try to estimate the potential losses during a credit crisis, it is not just the amount of assets and liabilities that matters, but also the quality of that asset. During quiet times, no one will pay much attention to the credit derivatives, but in 2008, it was the credit derivative that almost killed AIG (NYSE:AIG) (without Government's help, it would have been killed for sure).
When we compare the derivatives in Jefferies with Goldman Sachs, we can see Goldman has about 9 - 16 times more derivatives than Jefferies, and 4 times more level 3 asset, after adjusting for their sizes (assuming Goldman is 20 times larger than Jefferies based on their equity ratio). When it comes to credit derivatives and level 3 derivatives, Goldman has 25 - 35 times more exposure than Jefferies.
Another interesting thing I have observed is that Goldman only listed the net derivative asset ($57 billion) on their balance sheet, but Jefferies listed the original "raw" derivative asset ($2.5 billion) on the balance sheet, when its net derivative asset is only about $200 million.
Since Jefferies is considerably safer than the other large investment banks, I believe it should be justified to use a higher P/E multiple for its valuation.
Although the tighter regulations will help to control the investment banks in some degree, my opinion is that the tighter regulation won't be that effective when facing a real crisis again.
When we are talking about the complex derivatives business, it is often extremely difficult to estimate the value of the contracts. It is even more difficult when it comes to estimating the potential risk during a credit crisis. People often seek VaR (value at risk) for a guideline of risk exposure. What they don't know (or often forget) is that the VaR only gives the minimum risk during the bad times, not the maximum risk. The fact is that there is no way to model the maximum risk or the "average" risk during a crisis, at least not right now. We hear people are talking about "fat tails" and "black swans," but how fat should the tail be? Looking back to history, we have clearly seen that VaR underestimates the true risk by many folds. If not, why would something modeled as an AAA asset (senior tranches of subprime MBS) before the crisis turned out to be even worse than the junk bond? How much confidence can we still have in those models?
Given the difficulty of estimating the risks, how do we control our risks? I believe the answer should be "common sense" and "conservatism." In the financial world, "common sense" is often uncommon. When stocks are high flying, everyone loves them even though the history has shown the cycles again and again for so many times and in such a regular (almost predictable) pattern. When mortgages were lent out with no down payment or even negative equity, I believe some people must have known this is clearly wrong, but that common sense didn't stop them from doing it.
If the first defense or the minimum we need is the "common sense," then the next step should be the "conservatism," or we may even have to be a little paranoid. As Buffett said, when he invests in a bank, he has to be sure the management is good, since the risk control can't be done by regulators or credit rating agencies, and the CEO of a bank should also be the chief risk officer at the same time. His method proved right in 2008 when his two banks (Wells Fargo (NYSE:WFC) and U.S. Bancorp (NYSE:USB)) easily survived the crisis without much losses (Wells Fargo's trouble was more related to the acquisition of Wachovia, but even after that the risk was still within control.)
Again, I think at the investors' level, we should not rely on credit ratings or regulations; instead, we have to look at the balance sheet ourselves, and more importantly, we have to find a good management team that cares about shareholders and tends to be conservative on risk taking. I believe I have observed some signs of that conservatism in Jefferies' balance sheet.
Disclosure: The author is long LUK, UBNT, MKL. The author wrote this article themselves, and it expresses their own opinions. The author is not receiving compensation for it (other than from Seeking Alpha). The author has no business relationship with any company whose stock is mentioned in this article.