1 of 26
Slide Notes
A friend once said to me, 'There are two types of people in the world, merchants and bankers. Which one are you?
At the time, I considered myself a 'merchant' but somehow became a banker, and now I understand his point. Bankers simply think about things differently.
When you read the financial press, there are a lot of different terms etc but you have the tools to understand this. Somewhere between what you read and prevailing conventional wisdom are the actual facts.
At the time, I considered myself a 'merchant' but somehow became a banker, and now I understand his point. Bankers simply think about things differently.
When you read the financial press, there are a lot of different terms etc but you have the tools to understand this. Somewhere between what you read and prevailing conventional wisdom are the actual facts.
Banking and the Credit Crisis in Plain English
Published on Nov 19, 2015
Banking...Why do I care?
MORE DECKS TO EXPLORE
PRESENTATION OUTLINE
Banking
Bainbridge Graduate Institute, Dec 2014
A friend once said to me, 'There are two types of people in the world, merchants and bankers. Which one are you?
At the time, I considered myself a 'merchant' but somehow became a banker, and now I understand his point. Bankers simply think about things differently.
When you read the financial press, there are a lot of different terms etc but you have the tools to understand this. Somewhere between what you read and prevailing conventional wisdom are the actual facts.
At the time, I considered myself a 'merchant' but somehow became a banker, and now I understand his point. Bankers simply think about things differently.
When you read the financial press, there are a lot of different terms etc but you have the tools to understand this. Somewhere between what you read and prevailing conventional wisdom are the actual facts.
Photo by PaulyB71
A = L + E
As always, assets equal liabilities plus equity. For a bank, it can be hard to wrap your mind around a few things: you have a liability that is a product, cash, while necessary for liquidity, isn't something where more is better. In fact, its called a non-earning asset
Assets = loans, securities, buildings
Liabilities = deposits, borrowings
Equity = capital, reserves, shareholder's money
Normal companies: Maximize cash. A is good, L is bad, don't talk about E much.
Banks: Both sides of the balance sheet are good and bad. You have a liability that is a product. E gets talked about all the time.
Assets = loans, securities, buildings
Liabilities = deposits, borrowings
Equity = capital, reserves, shareholder's money
Normal companies: Maximize cash. A is good, L is bad, don't talk about E much.
Banks: Both sides of the balance sheet are good and bad. You have a liability that is a product. E gets talked about all the time.
Photo by Dvolatility
Example
Let's start....BGI Community Bank
Example: Let's start a bank or credit union. (simplifying past the fact that you would need a charter)
Assets Liabilities
$100M $90M
$40M Mortgages $10M Checking
$20M Credit Cards $20M Savings
$30M Gov't Bonds $30M CD's
$10M Cash $30M Wholesale
Equity
$10M
Capital ratio = 10% or 10x; 5% = 20x; 20% = 5x
LDR = 100%, Liquidity = 10%
Things to note:
Banks are highly leveraged
Bonds are a shock absorber on asset side; not a business model
Same with wholesale borrowings - don't want to be too dependent on them
Assets Liabilities
$100M $90M
$40M Mortgages $10M Checking
$20M Credit Cards $20M Savings
$30M Gov't Bonds $30M CD's
$10M Cash $30M Wholesale
Equity
$10M
Capital ratio = 10% or 10x; 5% = 20x; 20% = 5x
LDR = 100%, Liquidity = 10%
Things to note:
Banks are highly leveraged
Bonds are a shock absorber on asset side; not a business model
Same with wholesale borrowings - don't want to be too dependent on them
Photo by OZinOH
Income Statement
- Net Interest Income (70-90%)
- Provision for loan losses (0-10%)
- Fee income (20-40%)
- Non-interest expense (50%)
Let's say our sample bank makes 4% spread on all its assets, that's revenue of $4M. What if it loans out another $100M - it makes $8M. What's wrong with that?
1. Doesn't take very many loans going bad to be underwater
2. Liquidity issues
Differentiate between provision for loan losses and the balance sheet account
In reality, it has a 4% yield on mortgages, 15% on credit cards, 0.5% on bonds, 0 on cash, mortgage losses of 1%, credit card losses of 5% and 50% cost/income ratio
1. Doesn't take very many loans going bad to be underwater
2. Liquidity issues
Differentiate between provision for loan losses and the balance sheet account
In reality, it has a 4% yield on mortgages, 15% on credit cards, 0.5% on bonds, 0 on cash, mortgage losses of 1%, credit card losses of 5% and 50% cost/income ratio
DuPont Analysis
ROE = ROA X Leverage
Banks are a low margin business. Buying and selling the ultimate commodity - money.
Why focus on the metric ROE?
Tell Economic Profit story
What is the easiest way to boost ROE? Lever up
First nifty thing to break out at a cocktail party:
ROE = ROA * Leverage
For a normal business:
ROE = Margin * Inventory Turn * Leverage
For a Bank:
ROE = ROA * Leverage
ROA = (Net Interest Income/Ave Assets + Fee Income/Ave Assets) * Ave Earning Assets/Ave Assets * Total Income After Tax/Total Income * Total Income After Tax - Costs/Total Income After Tax * Average Assets/Average Equity
= NIM * Fee Income % * Tax Efficiency * Cost Efficiency * Leverage
Why focus on the metric ROE?
Tell Economic Profit story
What is the easiest way to boost ROE? Lever up
First nifty thing to break out at a cocktail party:
ROE = ROA * Leverage
For a normal business:
ROE = Margin * Inventory Turn * Leverage
For a Bank:
ROE = ROA * Leverage
ROA = (Net Interest Income/Ave Assets + Fee Income/Ave Assets) * Ave Earning Assets/Ave Assets * Total Income After Tax/Total Income * Total Income After Tax - Costs/Total Income After Tax * Average Assets/Average Equity
= NIM * Fee Income % * Tax Efficiency * Cost Efficiency * Leverage
ROE =ROA * Leverage
- ROE = NI/Ave Equity
- ROA = NI/Ave Assets
- Leverage = Average Assets/Average Equity
- ROA = f(Net Interest Margin, Fee Income, Earning Assets, Expense Efficienc
Example
Columbia Bank and Sterling Bank
Columbia- 6% (46/762)=.94% (46/4845) * 6.36 (4845/762)
ROE = 6% = 46/762 Net Interest Income/EA * EA/TA * (Total Income - NIE)/Net Int Income * Post Tax Net Income /Pre Tax Net Income
=(199/4845 + 11/4845 * 2848/4845) * (210-150)/210 * 46/63
=4.10% NIM + .2% fee * 58% EA level * 28% Operating Efficiency * 73% tax efficiency * 6.35x
Sterling
ROE = 8.98% (93/1047) = 1.02% (93/9214) * 8.8 (9214/1047)
= 297/6091 + 163/6091* 6091/9214 * 460 - 343/460 * 385/93
= 4.87% + 2.67% fee * 56% EA Level* 25% Op Eff * 410% tax efficiency* 8.8x
What increase in provision would wipe out profits: Columbia (46m or 1.61%EA) Sterling 1.52%(93M)
Columbia- 15% cap ratio Sterling 13%
ROE = 6% = 46/762 Net Interest Income/EA * EA/TA * (Total Income - NIE)/Net Int Income * Post Tax Net Income /Pre Tax Net Income
=(199/4845 + 11/4845 * 2848/4845) * (210-150)/210 * 46/63
=4.10% NIM + .2% fee * 58% EA level * 28% Operating Efficiency * 73% tax efficiency * 6.35x
Sterling
ROE = 8.98% (93/1047) = 1.02% (93/9214) * 8.8 (9214/1047)
= 297/6091 + 163/6091* 6091/9214 * 460 - 343/460 * 385/93
= 4.87% + 2.67% fee * 56% EA Level* 25% Op Eff * 410% tax efficiency* 8.8x
What increase in provision would wipe out profits: Columbia (46m or 1.61%EA) Sterling 1.52%(93M)
Columbia- 15% cap ratio Sterling 13%
Photo by Canadian Pacific
Risk
Credit, Market (aka Interest Rate), Operational
Now is when we get into some more bank lingo:
Now, lets get into some more bank lingo. The difference between a good bank and a bad bank is how they manage risk through the credit and interest rate cycle. Risk appears in other companies but we don't normally talk about it explicitly. Basically, there are 3 broad types of risk:
1. Credit - breaks down in counterparty, secured, unsecured, etc.
2. Market - basically interest rate risk - duration, basis, extension, etc
3. Operational - fraud, reputational, physical damage
Capital is allocated to these 3 categories; capital we have vs. capital we need. Credit is the largest usually - by about 5-10x. The hard part of lending is getting people to pay you back. Simplistically, the people that need money are the ones that don't have it. That's why its you can mis-manage a bank around interest rate risk but if you really want to sink a bank, credit will do it.
There are different capital measures and rules. Most involve giving a risk weighting got the asset side of the balance sheet.
Now, lets get into some more bank lingo. The difference between a good bank and a bad bank is how they manage risk through the credit and interest rate cycle. Risk appears in other companies but we don't normally talk about it explicitly. Basically, there are 3 broad types of risk:
1. Credit - breaks down in counterparty, secured, unsecured, etc.
2. Market - basically interest rate risk - duration, basis, extension, etc
3. Operational - fraud, reputational, physical damage
Capital is allocated to these 3 categories; capital we have vs. capital we need. Credit is the largest usually - by about 5-10x. The hard part of lending is getting people to pay you back. Simplistically, the people that need money are the ones that don't have it. That's why its you can mis-manage a bank around interest rate risk but if you really want to sink a bank, credit will do it.
There are different capital measures and rules. Most involve giving a risk weighting got the asset side of the balance sheet.
Photo by epSos.de
Interest Rate Risk
Banking is a "spread" business
What is an 'interest rate' - its the price of money. Your business as a banker is to buy and sell money. We understand price when its the supply and demand for a good or service. Same for money. Banks are a 'spread' business - borrow low and lend high.
This is a 'normal' yield curve. The longer you want to borrow, the more you have to pay. Reflects inflationary expectations.
2nd thing to break out at a cocktail party:
Cool tricks with interest rates.
Long term rates are really just the combo of market expectations for future short term interest rates.
1 year rate = 1%; 2 year rate = 3% (1 +x)/ 2 = 3%==>the 1 year rate 1 year forward is 5%
This is a 'normal' yield curve. The longer you want to borrow, the more you have to pay. Reflects inflationary expectations.
2nd thing to break out at a cocktail party:
Cool tricks with interest rates.
Long term rates are really just the combo of market expectations for future short term interest rates.
1 year rate = 1%; 2 year rate = 3% (1 +x)/ 2 = 3%==>the 1 year rate 1 year forward is 5%
Price Sensitivity
Interest Rate Risk
Duration is the price sensitivity of a financial asset - a bond. With all things being equal, (e.g. no other risks impacting price), the price of a stream of cash flows will change per this graph. Duration is also known as the geometric average of the cash flows from a bond. Basically, if you own a bond that pays a market interest rate of 5% and market rates, drop, your bond is now worth more to people
5 things to note:
1. Inverse relationship between price and yield
2. Longer term to maturity => more price sensitivity
3. Lower coupon => More price sensitivity
4. Lower market yields =>More price sensitivity
5. Capital gains always greater than capital losses for the same change in yield.
A bank balance sheet contains this type of interest rate risk. You'll read about banks with losses due to interest rate changes all the time.
5 things to note:
1. Inverse relationship between price and yield
2. Longer term to maturity => more price sensitivity
3. Lower coupon => More price sensitivity
4. Lower market yields =>More price sensitivity
5. Capital gains always greater than capital losses for the same change in yield.
A bank balance sheet contains this type of interest rate risk. You'll read about banks with losses due to interest rate changes all the time.
Credit risk
EL= PD x EAD x LGD
Remember the difference between provision and ALLL. The ALLL is the balance sheet account. Provision is the income statement manifestation of changes to ALLL.
Ending ALLL balance = Beg bal - NCO's+ Provision + Recoveries
Over time, your expected loss on a loan = the probability of default * exposure at default * loss given default
Now you can see why its hard to accurately estimate credit risk. You can be wrong on 3 separate things.
Ending ALLL balance = Beg bal - NCO's+ Provision + Recoveries
Over time, your expected loss on a loan = the probability of default * exposure at default * loss given default
Now you can see why its hard to accurately estimate credit risk. You can be wrong on 3 separate things.
Photo by woodleywonderworks
Mortgages
Hard to be a bank and not do mortgages. Yes, there are lots of other types of loans but not at the scale of mortgages.
Stats on mortgage market
US mortgage market - biggest in the world
Mortgages - biggest financial asset class
MBS - Larger than the US treasury market
Stats on mortgage market
US mortgage market - biggest in the world
Mortgages - biggest financial asset class
MBS - Larger than the US treasury market
Swaps and MBS
Very necessary
Go back to our sample bank. $60M in deposits (mostly short term) and $60M in loans (mostly long term mortgages)
The nature of most banks products is that they borrow short term and lend long term. This is called a 'carry' trade and has risk.
This is why swaps are important risk management tools. Buy a Pay Fixed - Receive Floating Swap to match the pay floating - receive fixed nature of your balance sheet.
Swaps came about in the 80's for risk management reasons. Lots of flavors: interest rate swaps, fx swap, credit default swap
Let's say you want to keep lending but you don't want to (or aren't allowed to) increase leverage. How do you do that?
Seuritization - get some loans off the balance sheet.
Is there a social good from these...yes!
The nature of most banks products is that they borrow short term and lend long term. This is called a 'carry' trade and has risk.
This is why swaps are important risk management tools. Buy a Pay Fixed - Receive Floating Swap to match the pay floating - receive fixed nature of your balance sheet.
Swaps came about in the 80's for risk management reasons. Lots of flavors: interest rate swaps, fx swap, credit default swap
Let's say you want to keep lending but you don't want to (or aren't allowed to) increase leverage. How do you do that?
Seuritization - get some loans off the balance sheet.
Is there a social good from these...yes!
Photo by quaziefoto
Financial Crisis
Terms
- Retail Banks
- Commercial Banks
- Investment Banks
- Universal Banks
- Insurance Companies
Photo by bingpoint-uk
Capital (akA Equity)
- All crises seem to start out as a liquidity problem
- They all seem to end as a crisis in confidence
- Underlying issue A-L
- Capital is king
- Powell Doctrine
Recommended books:
Too Big To Fail
The Big Short
The Lost Bank
The credit problem manifested itself as a liquidity 'issue' in August 07. It ended as a crisis in confidence.
TARP statistics: $608B outflows; To date $580B paid back
Credit stats: Pre-crisis, a 5% equity to assets ratio was pretty common. In fact, for some European banks (Barclays, Deustche Bank) it was 2 -2.5%. TARP added 2% to this. The average is around 11% today. Alan Greenspan suggests we should be at 14% or higher. Will an increase in capital requirements result in lower ROE's....maybe. Higher confidence in the system (lower borrowing costs) and greater efficiency could mitigate.
Too Big To Fail
The Big Short
The Lost Bank
The credit problem manifested itself as a liquidity 'issue' in August 07. It ended as a crisis in confidence.
TARP statistics: $608B outflows; To date $580B paid back
Credit stats: Pre-crisis, a 5% equity to assets ratio was pretty common. In fact, for some European banks (Barclays, Deustche Bank) it was 2 -2.5%. TARP added 2% to this. The average is around 11% today. Alan Greenspan suggests we should be at 14% or higher. Will an increase in capital requirements result in lower ROE's....maybe. Higher confidence in the system (lower borrowing costs) and greater efficiency could mitigate.
Photo by nuklr.dave
CDO and CDS
Thanksgiving 2007. One friend asked another why volatility was so low. He replied, 'Everyone is hedged out'.
It turns out, ultimate, that everyone was insured by AIG financial products division. (that's an oversimplification)
In normal insurance products, their are massive risk models run by actuaries with reserve requirements, etc. In this case, the State of New York would have to be the one to catch all the risk at AIG.
'Structured finance' has grown from about 20% of financial services profits in the 80's to 50% in 2006 and now it will likely go back to 20%. (need to find the reference for that)
It turns out, ultimate, that everyone was insured by AIG financial products division. (that's an oversimplification)
In normal insurance products, their are massive risk models run by actuaries with reserve requirements, etc. In this case, the State of New York would have to be the one to catch all the risk at AIG.
'Structured finance' has grown from about 20% of financial services profits in the 80's to 50% in 2006 and now it will likely go back to 20%. (need to find the reference for that)
Photo by trader_john
FCIC Report
- Widespread failures in financial regulation
- Dramatic failures in risk management
- Excess borrowing and lack of transparency
- Breakdown in ethics
- Derivatives and credit rating agencies
When everything is wrong, nothing is wrong.
The FCIC report was not really a root cause analysis.
The FCIC report was not really a root cause analysis.
Dissenting Opinions
Global nature of the issue
This chart represents the home prices in many countries.
A more focused analysis if the causes of the global crisis might be found in the first dissenting opinion of the FCIC report.
1. Everything is global. BRIC economies come of age
2. Lots of cash looking for a home in the world
3. Also, aging populations in the west and large pension pools
4. Everyone looking for yield.
5. Banks made dodgy loans because investment banks were willing to buy them
6. Investment banks bought them because they had investors who were looking for yield
A more focused analysis if the causes of the global crisis might be found in the first dissenting opinion of the FCIC report.
1. Everything is global. BRIC economies come of age
2. Lots of cash looking for a home in the world
3. Also, aging populations in the west and large pension pools
4. Everyone looking for yield.
5. Banks made dodgy loans because investment banks were willing to buy them
6. Investment banks bought them because they had investors who were looking for yield
Stats is the Most important class I took in B-school
- Low probability that housing prices would decline significantly
- Pricers were largely uncorrelated across regions
- Low level of 'strategic defaults'
- Regression models are only as good as the data set
We make big assumptions when using statistics to model things. Frequently, the limitations of these models are not well understood by the business using them.
One big assumption is that events will follow a normal distribution curve. We are having more tail events that a normal curve would suggest.
Photo by Trevor Blake
Regulation
- Can't inspect your way into more security
- Living wills
- We need to regulate risk and capital
- Global consistency
- Debit cards?
Its like raising kids: Any solution where you have to inspect your way to safety will always fall apart.
We need to regulate risk and capital....in a globally consistent way.
We need to regulate risk and capital....in a globally consistent way.
Recent Developments
- LIBOR, FX, SEC Mortgage Settlements
- Volcker Rule and prop trading
- SIFI subsidy, FNMA, and FHMC
- Money market funds
- Rating agencies and auditors
LIBOR scandal: This chart represent Barclays actual LIBOR submissions over the fall of 2008. They were the highest and so their submission was throw out...why all the fuss.....there is always more to the story.
SIFI subsidy:
If TBTF implies a funding advantage of 60-80 bps for the big guys, with 10x leverage, that is a 6 - 8% increase in ROE. Let's say sample bank earned an ROE of 15% (Net income of $1.5M on 10M in equity). In reality, that should be 7%. How much more capital should it carry to bring it back to that level?
Answer: about double - $20M. $1.5M/$20 = 7.5%
SIFI subsidy:
If TBTF implies a funding advantage of 60-80 bps for the big guys, with 10x leverage, that is a 6 - 8% increase in ROE. Let's say sample bank earned an ROE of 15% (Net income of $1.5M on 10M in equity). In reality, that should be 7%. How much more capital should it carry to bring it back to that level?
Answer: about double - $20M. $1.5M/$20 = 7.5%
Long Term Credit Cycle
Money = Credit
What is happening in the world and what is the outlook? Ray Dalio, Bridgewater: The world economy is going through a structural deleveraging. Economic growth can be explained by the combination of 3 things.
First, there is the long-term productivity trend, which has been roughly a 2%
annual increase in GDP for the US and Europe for the last 100 years. Second, there is the short-term credit cycle that takes place at intervals of about 5-8 years. The gist of this cycle is that as credit and liquidity expand, business activity expands, and inflation increases. To keep inflation under control, the central bank raises bank borrowing rates through open market operations and the discount rate reduce the amount of credit available, and business activity slows down. The primary tool in this cycle is monetary policy – changing the interest rates that banks pay to reduce the amount of money in circulation.
It's important to note much of what economists call ‘money’ in the various
definitions of the money supply is not cash and coin, but rather credit and the ability to borrow. Its important to note that a large part of what we call the money supply is really credit. When you make $100M and borrow $10M, you spend $110M. That can continue and fuel a lot of expansion...until your debt service because unsustainable no matter how low interest rates are.
Which leads to the final point: the long-term debt cycle. This is much harder to identify because
it comes at much longer intervals – about 60-70 years. The fact that this is seen only once in a lifetime helps explain why policy response can be so disjointed. Ultimately, you need to pay down debit, default, or inflate your way out of the issue.
First, there is the long-term productivity trend, which has been roughly a 2%
annual increase in GDP for the US and Europe for the last 100 years. Second, there is the short-term credit cycle that takes place at intervals of about 5-8 years. The gist of this cycle is that as credit and liquidity expand, business activity expands, and inflation increases. To keep inflation under control, the central bank raises bank borrowing rates through open market operations and the discount rate reduce the amount of credit available, and business activity slows down. The primary tool in this cycle is monetary policy – changing the interest rates that banks pay to reduce the amount of money in circulation.
It's important to note much of what economists call ‘money’ in the various
definitions of the money supply is not cash and coin, but rather credit and the ability to borrow. Its important to note that a large part of what we call the money supply is really credit. When you make $100M and borrow $10M, you spend $110M. That can continue and fuel a lot of expansion...until your debt service because unsustainable no matter how low interest rates are.
Which leads to the final point: the long-term debt cycle. This is much harder to identify because
it comes at much longer intervals – about 60-70 years. The fact that this is seen only once in a lifetime helps explain why policy response can be so disjointed. Ultimately, you need to pay down debit, default, or inflate your way out of the issue.
Why Do I care?
- We don't use cars as the basis for value
- Financial services represents 10% of GDP
- Managing risk is hard
- A stable and profitable banking system is key for sustainable commerce
- At your next Christmas party: Break out ROE = ROA * Leverage
In the middle of the Detroit bailout, someone asked me, 'Why would we bail out banks and not car makers.' My initial thought was that 'banks are different'. We don't use cars as the basis for value and exchange.
When assessing non-bank competition and whether or not they will challenge banks, I always look at their ability to underwrite and manage risk better than a bank. That is the ultimate determinate of success.
Long term impacts of the crisis?:
1. No long term investment due to uncertainty
2. Monetary policy has done everything it can; fiscal policy needs to do the reset
3. Asset bubbles can be created; also drive more income inequality
4. Europe, Japan, China
Banks are like telcos or utilities. They have the ability to be much more than that but whether or not they will is is up to them.
When assessing non-bank competition and whether or not they will challenge banks, I always look at their ability to underwrite and manage risk better than a bank. That is the ultimate determinate of success.
Long term impacts of the crisis?:
1. No long term investment due to uncertainty
2. Monetary policy has done everything it can; fiscal policy needs to do the reset
3. Asset bubbles can be created; also drive more income inequality
4. Europe, Japan, China
Banks are like telcos or utilities. They have the ability to be much more than that but whether or not they will is is up to them.
