Banking System 101 – Part 2

This is the second part in an ongoing series designed to guide you toward a basic understanding of Federal Reserve and ECB monetary policy as conducted in recent history, as well as what exactly constitutes and drives quantitative easing. Despite being on the forefront of the mainstream media since 2007 you will rarely run into anyone who has a thorough understanding of how all the pieces fit together. We are completing our basic introduction and then will explore a chronological recap of what has transpired in the past few years. If you missed part one then please make sure you read it first before continuing below.


Recall our scenario from before, with two US commercial Banks, A and B. Reserve Requirements are 10%, and the balance sheet of Bank A that we ended up with is shown below:

As explained before, required reserves are simply 10% of deposits, and also recall the changes to Bank B’s balance sheet after it made a loan of 20bn USD to a firm:

Changes to its BS after making a loan are highlighted in red: most importantly, excess reserves decrease slightly, as deposits grow (Bank B credits Firm X’s deposit account with the loaned money, and the loan itself appears as an asset on the BS).  Clearly this could not occur if Bank B held no excess reserves in the first place. You can think of the loan as being a case of Bank B “using up” only 2bn of its excess reserves to “create” 20bn of money in the real economy. This will become important later, as this is known as the (money) multiplier effect.

To understand this effect we need to define Money supply – simply the total quantity of monetary assets available in the economy. There are different measures of money supply, of which the important ones are:

  1. Starting from the most narrow measure, we have M0 which is simply all the notes and coins in circulation. This does NOT include cash stored at the Federal Reserve Banks (ie. the central bank in the US) and cash stored at commercial banks’ vaults, as it is effectively dead money.
  2. The monetary base (MB) = M0 + bank vault cash + banks’ reserve accounts. This is a broader measure, that includes notes and coins but more importantly it includes a measure of central bank money, ie required and excess reserves, which commercial banks hold at their reserve accounts with the Fed. In our example from Figure 2.1, the monetary base would equal M0 + 0 + (10+40) bn + (17+43) bn = M0 + 110bn USD.
  3. M1 = M0 + demand deposits, and simply measures the money in circulation by also including the most common form of “electronic money”, which is easily accessible deposits (“on demand”), that one can use to withdraw money at any time without restrictions or delays. A common example of this is a debit (card) account.  Note that this measure does NOT include central bank money and vault cash! M1 therefore measures commercial bank money. For our example in Figure 2.1, M1 = M0 + (100bn) + (170bn) = M0 + 270bn USD [Assuming all deposits in our example are demand deposits].
  4. M2 = M1 + savings accounts + “time deposits” (money has to be locked in for a period for time to earn interest, and cannot be withdrawn – e.g. bonds, money-market deposit accounts for individuals etc). This is a broader measure of commercial bank money. In our example of Figure 2.1, M2 = M1 as we assume there are only demand deposits. [ There are other broader measures of money which I will not discuss here for simplicity ]

The multiplier effect, which we described for Bank B above, can be measured in general by the money multiplier (MM) , which is simply a statistic, that measures the ratio of “commercial bank” money to “central bank” money. Out of the 4 measures of money supply listed above, only the monetary base can qualify as central bank money. So we can choose either   MM = M1/MB  or  MM = M2/MB  to be the money multiplier. Let’s take money multiplier, ie MM =  M2 / MB  as a broader measure. Firstly, let’s measure the MM for our example in Figure 2.1, substituting in our calculations for M2 and MB above:

MM = M2/MB = M1/MB = (M0 + 270bn) / (M0 + 110bn). We do not know the notes/coins in circulation (M0), but given that in modern economy it is very small compared to electronic money (ie deposits), we can assume for our example to be negligible. Therefore as M0 à 0, mathematicians reading this will know that MM à 270bn / 110bn = 2.45.


Above we can see the money supply for the US economy, and the changes that resulted since 2008: As Fed started QE, the monetary base grew, but so did excess reserves – but M2 expanded at a lower pace than before, thereby reducing the MM to ALL-TIME lows.

According to the multiplier effect, an increase in bank reserves, which resulted from QE, should have been “multiplied” into a much larger increase in M2, as banks should have expanded their deposit and lending activities. The expansion of deposits, in turn, should have raised required reserves until there are no excess reserves in the system. Why has this NOT happened in our Figure 2.3 above?

To answer this, consider the situation from a commercial bank’s perspective. Once it gains excess reserves via QE, it has two options:

  1. deposit them with the Fed, in which case it would earn the Interest that the Fed pays On Excess Reserves (IOER); Since there were virtually no excess reserves until 2008, the Fed only started considering setting IOER after the 2008 crisis, and after briefly playing around with it, it was set to 0.25% and has remained that ever since.
  2. attempt to lend its excess reserves to earn an interest, which is higher than IOER (0.25%).

So the bank will attempt to lend out its excess reserves at any positive interest rate, greater than IOER. This additional lending in turn decreases short-term interest rates (Why? Interest is the price of money, and if more money is supplied via lending then this price goes down).  This lending also creates additional deposits in the banking system as in Figure 2.2 and thus leads to a small increase in required reserves.  Because the increase in required reserves is small, however, the supply of excess reserves remains large. The process then repeats itself, with banks making more new loans and the short-term interest rate falling further.

This multiplier process continues until one of two things happens:

  1. there are no more excess reserves, i.e. the lending has expanded so much that all excess reserves turned into required . In this case, the money multiplier is fully operational.
  2. However, the process will stop before this happens if the short-term interest rate slips below what the bank can earn on its excess reserves with the Fed, i.e. IOER. At this point, there is no longer an incentive to lend and hence the multiplier process stops.

Above we add the Fed funds effective rate, currently 0.15%, (ie the actual short-term interbank lending rate), and the IOER, 0.25%, to our money supply picture from before.

Initially in 2008 the money multiplier shrunk as the credit squeeze kickstarted a massive contraction in lending (ie money supply fell), the Fed responded by aggressively cutting the Fed Funds target rate, which in turn anchored the effective rate lower. Once the FF effective rate fell below the interest Fed paid on reserves, the banks were no longer incentivised to lend and the multiplier process stopped, resulting in further collapse of the MM.

In the summer of 2012, the Fed considered cutting IOER to 0% or even negative to encourage lending but that never got much traction and another round of QE went ahead instead. I would argue on the basis of the diagram above that further rounds of QE will become increasingly less effective, and therefore cutting IOER is a much more effective next step in monetary policy.

Will excess reserves cause inflation?

The market consensus is that the large quantity of reserves will lead to an increase in the inflation rate unless the Federal Reserve acts to remove them quickly once the economy begins to recover. I would concur, as once grows picks up (IF IT PICKS UP), demand for lending will increase sharply, and the presence of such a huge amount of excess reserves, will encourage banks to lend them out aggressively – this can create an exponential surge in lending over a very short period of time, thereby resulting in runaway inflation.

However, the Fed has the power to control this via two ways:

  1. Under a traditional framework, ie Fed’s Open Market Operations (OMOs), it can remove nearly all excess reserves from the banking system. It can do so by selling back Treasuries it acquired via QE: recall our review of Fed ‘s OMOs in the previous article – the primary dealers (ie banks), would have to drain their reserve accounts to put up cash in exchange for Fed’s bonds. As their reserves shrink, but clearly required reserves have stayed the same, their excess reserves therefore also diminish.
  2. By raising the Interest on Excess Reserves (IOER), the Fed can manage the build-up of excess reserves more directly. As IOER is raised, it discourages banks from lending at the market rate, and instead simply park their excess reserves at the Fed. Of course, the IOER would have to be consistently set higher than the market lending rate.

The big problem here, which officials refuse to discuss, is that runaway inflation would quickly lift the market bond yields. Why? Well, imagine you are lending someone money at 5% a year, when inflation is 10%. You would be stupid, since you are effectively locking in a -5% annualized real return. Instead you could simply buy an asset, whose price grows with inflation (ie real estate) and earn 10%. Therefore, government bond yields (ie the rates at which the US Treasury borrows from the market) must also track inflation.

The chart above illustrates this relationship between 5yr and 10yr bond yields and CPI for the last 50 years.

So in an inflationary scenario above, the borrowing costs for the US Treasury would rise materially. If the Fed employs OMOs to combat inflation, it risks screwing with the Treasury even more, as its bond sales would also lift bond yields even further (yield is proportional to the inverse of the bond price). As the size US government debt is well publicised (including social security payments its over 500% of GDP), you can easily see that the costs of funding that debt will become unmanageable – tax receipts that the Treasury will earn from a growing economy will be nowhere near enough to pay its debt, but if it borrows more from the market it will have to commit to paying even higher interest rates in the future. This is a classic debt spiral that terminates in one of two ways:

  1. Formal default. That would be the last resort, as since the US government bond market is the most liquid asset market in the world, it would bring down the Financial system globally and this time no central bank alone will be able to patch it up.
  2. Pseudo-default. In this case, the Treasury would not default officially but it will effectively make the US consumer pay – by inflating away its debt. It will print an enormous quantity of USD, which will dilute the purchasing power of the consumer in real terms. This would be hugely negative for the US dollar – we are talking about a bear market of our lifetime. For historical reference, look up Weimar Germany.


Peter Levchenko

Banking System 101 – Part 1

This is the first part in an ongoing series designed to guide you toward a basic understanding of Federal Reserve and ECB monetary policy as conducted in recent history, as well as what exactly constitutes and drives quantitative easing. Despite being on the forefront of the mainstream media since 2007 you will rarely run into anyone who has a thorough understanding of how all the pieces fit together. We will start with a basic introduction and  then explore a chronological recap of what has transpired in the past few years.

Fed Monetary Policy and Quantitative Easing

Before the 2008 crisis, conventional monetary policy in the US boiled down to the Federal Reserve (the Fed) controlling interest rates. The way it achieved that was by manipulating what’s known as the Fed funds rate:

Commercial banks (banks in the US that accept deposits and make loans) have daily funding requirements. Let’s assume Bank A faces a withdrawal of deposits today totaling $1bn. Deposits are a liability of the bank, so to make that payment to its customers it needs to borrow money. It could do that in the interbank market with another commercial bank, so it negotiates an overnight rate and deals at that rate.

The weighted average rate of all such transactions in the banking system every day is known as the Fed funds effective rate.  That is essentially a proxy for an overnight interbank market rate. The Fed regularly meets via Federal Open Market Committee (FOMC) and sets what is known as the Fed funds target rate, which is the level of interest rates they consider appropriate for the economy based on macroeconomic factors and so on. Their job as part of conducting monetary policy is to ensure that the market rate (ie the Fed funds effective rate) accurately mimics the benchmark Fed funds target rate.

The Fed uses Open Market Operations (OMOs) to achieve this, by directly altering the supply of money in the economy. In simple terms, think of it as two worlds: the Fed and the Monetary System (ie banks, lending institutions, and then the real economy).  The way money reaches the economy is via banks’ lending money to firms and businesses. All the money that is sitting at the Fed does not affect the economy in any way since it is not available for lending to businesses. BUT if the Fed chooses to inject money into the economy, the quantity of money that banks have available to lend increases.

Interest rate is the price of money. And similar to a stock, if supply of money increases – its ‘price’ (ie the interest rate) goes down. During Open Market Operations, the Fed buys and sells government bonds to banks as a way of regulating money supply. Suppose the Fed funds effective rate (ie the actual market rate) is trading higher than the Fed’s target (the Fed funds target rate). To lower the market rate, the Fed can attempt to increase the supply of money. To achieve this, it buys government bonds from banks, and in exchange it credits their reserve accounts. That way the money now enters the monetary system as banks can lend that money to the economy.

During Temporary OMOs, Fed reverses its actions on money supply, i.e. if it initially bought bonds from banks it will reverse that after a period of time by selling an equivalent amount, thereby creating only a temporary change in money supply. In 2009 as response to the credit crisis, the Fed announced a longer-dated programme which became the main part of their Permanent OMOs (POMOs), as they started outright purchases of longer-dated bonds without immediate intention to sell them back. This has been labelled as QE (quantitative easing).

How Does QE Work?

To illustrate, consider two US commercial banks, ie institutions that take deposits and make loans. We have Bank A, which has been started with $10bn equity capital, and has accepted $100bn of deposits. These are liabilities on its balance sheet (BS), since it effectively “owes” that money to shareholders and depositors respectively. For simplicity, consider Bank B which has exactly the same liabilities. Let’s assume for simplicity that they are the only two commercial banks in the system.

By law a bank is obliged to maintain a certain level of reserves with the central bank, which it should hold at its reserve account. This is what is known as reserve requirements and they cover the bank’s assets and liabilities but the most basic form is that banks in the US should hold 10% of their deposits as reserves. These show up as assets, since they represent cash owned by the bank.

So total reserves in the system are $20bn and they are all required reserves, i.e. there are no excess reserves. Banks also make loans, which are assets on their balance sheet (since they are owed money by borrowers) and lets assume that they hold some securities on their balance sheet too (securities = stocks/bonds etc). Suppose Bank B has access to a larger pool of lending opportunities, and to be able to make more loans it has borrowed $40bn from Bank A. The full balance sheet of both banks is displayed in Figure 1.1 with the interbank loan in orange (Bank B owes money so it is a liability on its BS, and an asset on Bank A’s BS):

Note the importance of the interbank lending system in normal times, money has flown towards its most productive use (i.e. Bank B borrowed to service the need of its customers). In our example from Figure 1.1., suppose the financial system enters a period of turmoil like in 2008, which disrupted the normal pattern of lending and led to a “freeze”, which could be due to uncertainty about banks’ ability to pay back and also their ability to borrow for their own funding needs. Now Bank A does not want to continue lending to Bank B. This places a severe disruption on Bank B: if it is unable to quickly raise new deposits or borrow elsewhere to obtain the $40bn it owes, it could be forced to decrease its loans by $40bn. That in turn would force a contraction in deposits, as borrowers scramble to pay back loans to Bank B resulting in a sharp contraction in economic activity.

To prevent this, the Fed steps in with QE and suppose it buys $40bn of Treasuries from Bank B. Recall that in the original example the banking system, which consists of A and B, held no excess reserves. The resulting balance sheets after $40bn Fed QE are shown below (changes in red):

Fed credited Bank B’s reserve account with $40bn, which Bank B used to pay back the loan to Bank A (hence Bank A’s reserve account now holds extra 40bn). Bank B got the money by selling $40bn of its bonds (securities). Note that the total reserves in the system are now $60bn, but only $20bn are required – hence there are now $40bn of excess reserves. The chart below shows the actual US banking system total reserves (yellow) and excess reserves (red) that have resulted from QE, similar to the example above:

Numerous market observers and commentators such as ZeroHedge have used this to conclude that Fed has injected “dead money” that is sitting on banks’ balance sheets and hence its policies have been ineffective. But as our example shows, the job of the Fed here was to prevent an interbank market freeze and it was highly effective in this regard – it prevented Bank B from having to reduce its lending by $40bn!

In fact, lets assume that Fed does $50bn more QE, but this time the cash is released into the real economy, ie firms are sellers of the bonds to the Fed. The Fed can only deal with primary dealers which are banks, so suppose Bank B stands in as the intermediary and sells the bonds to the Fed on behalf of its clients:

Resulting Bank B’s balance sheet is now shown in Figure 1.3 (changes in red from Figure 1.2). I have split total reserves into required and excess reserves for clarity: the Fed deals with Bank B, which is representing the firms, so it credits Bank B’s reserve account with 50bn. Bank B is selling bonds on behalf of its clients to the Fed in return, so there is no drain to Bank B’s own stockpile of securities. To channel the proceeds of this sale to its clients, Bank B must credit the deposit accounts of its clients – hence its deposits increase by $50bn from Fig 1.2. So both reserves and deposits go up, BUT required reserves are only 10% * 150bn = $15bn, hence Bank B now gains $45bn in excess reserves (it previously held none).

Let’s further assume that Bank B directly increases lending by making a new loan of $20bn to a firm. Resulting balance sheet is shown in Figure 1.4: Bank B now has a new asset (the loan to the firm of 20bn) and an offsetting liability (it credits the firm’s deposit account at the bank by 20bn). Note that Bank B’s total reserves have not changed at all, and excess reserves have only narrowed slightly to ($60bn – 10%*170bn) = $43bn.

This is counter intuitive, as even though bank lending in the economy increased, total reserves and excess reserves still went up! Sure, the firm in our example can now withdraw that money from its deposit account and spend it, but since most of the money payments are nowadays electronic whatever they spend this money on (investment, research etc) will just end up at other firms’ accounts, which in turn would end up as another bank’s deposits.

This is the general principle: loans to banks and firms and direct purchases by the central bank all increase the TOTAL level of reserves in the banking system by the same amount. It doesn’t imply anything about the level of lending in the economy! Similarly, the level of excess reserves in the system does NOT imply that banks are not lending.  In the example above bank lending went UP by nearly 10% ( = $20bn / Total loans [$150bn + $60bn] ) But Bank B’s excess reserves also went UP by $43bn.


The Sky Is The Limit

As I’m typing this the SPX is a bit over 33 handles away from its all time high of 1576.09. Whereas just yesterday we were stuck near an inflection point which the bears could have exploited to their advantage the landscape has changed quite a bit since early this morning:

After the limbo comes the rocket. As of right now I don’t see any technical hurdles that would prevent us from continuing toward our P&F target of SPX 1615.

As a matter of fact today it triggered a triple top breakout – that’s just two days after a bear trap warning when the ES was stuck near that diagonal resistance line. The bears had a real good shot there to exploit this but once again didn’t have the balls to exploit momentary weakness. Not that I could blame them given the full roster of POMO auctions scheduled throughout March.

Crude is looking interesting tonight. That hourly candle tells me that it has a pressing appointment with its 100-hour SMA, which also matches the 100-day on the daily panel. Which would constitute a possible last kiss goodbye move – and that is a spot I would love to get positioned at.

However most of the action is happening over on the FX side today – please step into my lair:

More charts and commentary below for anyone donning a secret decoder ring. If you are interested in becoming a Gold member then don't waste time and sign up here. And if you are a Zero subscriber you get free access to all Gold posts, which gives you double the bang for your buck!

Please login or subscribe here to see the remainder of this post.


    Zero Indicator

  1. poll

    • What is your average spread on the EUR/USD?

      view results

      Loading ... Loading ...

  2. search warrant

  3. recent misdeeds

    1. Don’t buy the dip this time
    2. No Crying Over Spilled Milk (Or Beer)
    3. Entries Do Not Matter
    4. Last Chance For The Bears
    5. Three Strikes You’re Out
    6. Binary Proposition
    7. Time To Wield The Iron
    8. Zone of Fuckery
    9. Whipsaw Galore
    10. Fridays Stunning Reversal and What it Means

  4. yes we can!