Splitting between quantitative and qualitative easing

23-12-2008: Splitting between quantitative and qualitative easing

Special report by DBS Group Research
Email us your feedback at fd@bizedge.com

WHEN the Fed cut its interest rate target to between zero and 0.25% last week it marked the end of interest rate easing and the beginning of what most people refer to as quantitative easing - the outright purchase of assets to influence liquidity conditions in financial markets.
In fact, even an interest rate policy is really a quantitative policy when you get right down to it. The only way the Fed controlled its policy rate, Fed funds, was by buying and selling assets, mainly repos and reverse repos, in the market.
(The day-to-day conduct of monetary policy is not typically done via buying/selling US Treasuries in what are called open market operations but rather by buying repos or reverse repos. A repo (repurchase agreement) is a temporary loan to a securities dealer, often overnight and usually for a few days or less.
The Fed credits the dealer's bank with an increase in reserves at the Fed and the borrower places high quality collateral (government or agency securities) with the Fed. At the end of the period, the collateral is returned to the borrower and the bank's reserves are appropriately debited.
The repo is thus a temporary injection of liquidity into the market. A reverse repo works the other way. The Fed temporarily drains liquidity from the market by debiting the account (reserves) of a security dealer's bank and replacing the reserves a few days later.) Buying repos injected liquidity and lowered the Fed funds rate. Buying reverse repos did the reverse. By hook or by crook, quantitative policy has been the reality all along.
Still, there's a perception that with short-term rates at zero, we've now entered a new phase of monetary easing. A quantitative phase. The Fed is going to buy large quantities of assets and continue to expand its balance sheet.
Quantitative versus qualitative easing Isn't this expansion of the Fed's balance sheet - this quantitative easing - just printing money? The answer over the past year has been yes, mostly, but not entirely.
Fed purchases of troubled assets or loans to troubled firms injects liquidity into the system and raises the monetary base. But that rise could be offset by sales of other assets like government bonds.
How much offsetting (sterilising) the Fed does when it buys a chunk of dubious assets determines whether its purchases are translated into quantitative easing (monetary expansion), or qualitative easing - an extension of liquidity to a firm, which does not raise the monetary base but dilutes the assets' quality on the Fed's books.
In the past three months, the Fed's balance sheet has grown by US$1.25 trillion (RM4.38 trillion) or 2.5 times and the monetary base has doubled. With the Fed planning to buy another US$800 billion of assets, how it pays for these assets will influence the monetary base and could have important consequences for inflation and the value of the dollar. The quality/quantity split is key.
Splitting the cake How has the Fed chosen to split the cake so far? Very inconsistently. Between August 2007 (when the crisis erupted) and Sept 3, 2008, the Fed injected some US$339 billion into markets, sterilising 93% of its purchases of troubled assets with sales of government bonds. Seven percent (US$28 billion) was paid for with an increase in the monetary base. In September, though, the Fed's balance sheet grew by a whopping 50%, or US$428 billion.
About US$157 billion (37%) of that was financed by a rise in the monetary base. Then things went wild. Between Oct 1 and Dec 17, the Fed bought US$841 billion of troubled assets, growing its balance sheet by another two-thirds. Eighty percent of that expansion (US$672 billion) was financed by an expansion in the monetary base - by printing money.
Since Sept 3, the Fed's balance sheet has expanded 2.5 times thanks to US$1.6 trillion of new credit extended loans to illiquid firms. Some 53% (US$857 billion) has been paid for via quantitative easing (an expansion in the monetary base) and 47% via qualitative easing (sterilising, or swapping good bonds for bad).
Back to qualitative easing? The Fed's Dec 16 FOMC statement that large scale quantitative easing was on the cards raised a lot of eyebrows. After all, the monetary base has already doubled since September.
So at a subsequent press conference, "senior Fed officials" hastened to add that the split going forward would favour qualitative easing (sterilised asset swaps) over quantitative easing (money printing á la Japan).
But that would be a big U-turn after the past two months of rolling the printing presses. How can you know for sure? You can't. The best you can do is to watch the data week by week.
The Fed's balance sheet - up close and technical Below, we discuss the evolution of the Fed's balance sheet since August 2007 and the present. Amongst it is the mechanics of some typical and, more recently, non-typical Fed transactions so that readers may see their effects on the monetary base and the quantitative/qualitative split in Fed policy. Readers familiar with the Fed's balance sheet and its mechanics may wish to skip ahead to the final section.
The good old days and Fed funds: Back in the good old days, the Fed's balance sheet was a pretty simple thing. On the asset side, the Fed held a lot of government bonds and not much else. Repos and reverse repos, used to fine-tune monetary policy on a day-to-day basis, were small. On Aug 8, 2007, reverse repos outnumbered repos by US$13 billion. Use of the discount window was extremely tiny and the Fed held US$40 billion of miscellaneous assets.
On the liability side were two items: currency in circulation and reserve balances, or simply reserves. Reserves are the deposits that banks must hold at the Fed. Minimum levels must be maintained on a daily basis. Banks which hold excess reserves may lend to banks temporarily short. The interest rate on these loans is called the Fed funds rate. As most are aware, this rate is the Fed's policy rate and through it the Fed controls, or used to anyway, other short-term market rates.
Open market operations: The most generic way to increase the money supply is called an open market operation (OMO). The first step is not necessary but typically begins with the Treasury issuing say a US$100 bond to a securities dealer in return for cash that it uses to cover its deficit. The Fed purchases the bond from the dealer, crediting the dealer's bank's account at the Fed for US$100. The bank may withdraw the funds from its reserve account at any time and since, in the past, reserves did not pay interest, it would typically do so fairly quickly. The US$100 bond issue has led to a US$100 increase in currency in circulation. If the Fed wanted to tighten liquidity, it would do the reverse, selling a Treasury bond into the market. The Fed would debit the dealer's bank reserves for US$100 and this would soon result in a reduction in the amount of currency in circulation. As member banks may always demand currency for reserves, the monetary base is the sum of these two Fed liabilities.
Repo injection: The Fed would typically engage in open market operations when it had heavy lifting to do, such as when it had changed the target for policy rates, Fed funds. For day-to-day fine-tuning of the Fed funds rate, repos and reverse repos would typically be used. Repos are loans extended for very short durations, usually a few days or less, secured with high quality collateral (UST or agency debt). As one would expect, a syphoning of liquidity via reverse repos would simply reverse the signs in the balance sheet entries. Interest rates on Fed funds would rise and would be transmitted out to the rest of the market.
Term Auction Facility (TAF) injections: By June 25, 2008, the Fed had extended US$150 billion of loans via the Term Auction Facility, or TAF. Other things equal, those TAF loans would have increased the monetary base just like the OMO or repo injections shown. But the Fed was worried that the TAF loans (and other injections that, between August 2007 and June 2008, amounted to US$339 billion) would have inflationary consequences. The Fed decided to offset these injections with sales of government bonds. Total assets have remained the same, as have total liabilities. This is, in effect, a simple asset swap. Again, the purpose is to provide liquidity to the TAF party in need without increasing the monetary base. The Fed's balance sheet becomes too small? Contrary to much market talk back in October, there are no constraints on the size of the Fed's balance sheet. The Fed can print money, which means it can buy any amount of assets. The sky's truly the limit. But the Fed's balance sheet can be constrained if it tries to buy too many assets without printing money. Once the Fed runs out of Treasuries, the sterilised asset swap can no longer occur. By September 2008, the Fed's supply of Treasuries had fallen by nearly half, thanks to sterilised asset swaps. In fact, it had fallen far lower than that because the Fed had already swapped another US$250 billion of Treasuries more formally in its Term Securities Lending Facility (TSLF) which the Fed records as an off-balance sheet item. In reality, 70% of the Fed's supply of Treasuries had been swapped by Sept 3, 2008. Unless the Fed could come up with some additional Treasuries, its rapidly growing portfolio of loans made through the TAF, the Primary Dealer Credit Facility (PDCF) and eight other programmes listed in the balance sheet would soon have to be financed by printing money alone.
On Oct 7, the Fed and the Treasury announced the Supplementary Financing Programme. What this amounted to, in effect, was a joint effort whereby the Fed would purchase the dubious assets as it had been and, since it was out of Treasuries, the Treasury would do the mopping up. A sterilised asset swap still obtained the result of two agencies being better than one.
There are various ways of viewing this from a balance sheet perspective. The most transparent way is probably to view the joint transaction in three distinct steps.
In step 1, the Treasury sells US$100 of bonds directly to the Fed. The Fed pays for the bonds by crediting the Treasury's supplementary reserve account. The Treasury promises not to withdraw or lend these funds so they do not affect the monetary base or the Fed funds rate.
In step 2, the Fed loans US$100 to a primary dealer, as per before, which leads a monetary base increase of US$100. In step 3, the monetary increase is sterilised by the sale of the newly acquired government bond. In net terms, primary dealers have received US$100 of new liquidity but the monetary base has not changed. In this case, the Fed's balance sheet has grown by US$100 but there is no monetary impact.
A more straightforward way of accomplishing the same thing would be for the Fed to issue its own bonds. The Fed's US$100 loan to a primary dealer broker leads to a rise in the monetary base of the same amount.
The Fed sterilises this injection with the sale of a Federal Reserve bond. This would be a pure asset swap although new Fed bonds would be issued (the size of the Fed's balance sheet would grow). If the Fed needs to expand its balance sheet to finance the acquisition of "large quantities" of assets without monetary consequences, then, in practical terms, it would seem to make little difference whether the Fed issued its own bonds for sterilisation purposes or bought them from the Treasury for onward sale to the market.
Why the quantity/quality split? And will it work? Besides avoiding the monetary consequences of large purchases of troubled assets, the Fed's qualitative easing is aimed at lowering the spread of various interest rates over the risk-free Treasury rate.
The Fed apparently feels that risk-free rates are low enough (10Y UST yields are currently at 2.13%). The hope is that by buying troubled assets and selling Treasury's, the yields on the two securities would come closer together.
Whether it will work remains unclear. Many consider the spread to be a function mainly of the probability of default on the part of the troubled borrower, not on the relative amounts of the securities bought and sold.
On this view, unless Fed loans actually lower the probability of default, the credit spread will remain. What matters to the borrower, though, and for that matter to the economy more generally, is not the spread he has to pay over the risk-free rate, but the actual rate at which he can borrow.
If the Fed wants to lower that rate, it may have to content itself with lowering the risk-free rate, for example, pursuing quantitative easing over qualitative.
Second, one cannot not forget the adage that leading a horse to water does not make it drink. The Fed can buy assets and credit the reserves of the banks. But unless the banks withdraw those funds and lend them onward in the market, it's all for naught.
This is currently a problem. The monetary base on Dec 17 was US$1.67 billion but half of that comprised reserves. Banks are not withdrawing the funds and putting them to use. This may be good from an inflation perspective but plainly not from a get-the-economy-going perspective.
Finally, one must not forget that many financial institutions have reported large losses and it is reasonable to assume that some, perhaps many, are insolvent. In such cases, Fed easing will not solve the problem but only postpone the day of reckoning. When it comes to putting off until tomorrow what can be done today, there is no quantity/quality split.

http://www.theedgedaily.com/cms/content.jsp?id=com.tms.cms.article.Article_62478269-cb73c03a-53897400-a8659b88