Of -6% Fed Fund Rates and $9.3 Trillion in Troubled US Assets

Zero Hedge makes fun of sell side research often. After all who doesn't. But sometimes we are pleasantly surprised, such as when we read Goldman's weekly economic analysis piece for the week of January 23. Several issues discussed in the report include an overview of the remaining arsenal of governmental responses to the mega crisis we are in, both in the fiscal and monetary realm, a sober estimation of just how bad the real troubled asset situation in the U.S. may be if we remove the wool from our eyes, as well as what needs to be considered before making an appropriate policy move.

A key issue in the current market environment is the sole reliance on fiscal stimuli to push the economy out of this slump, as monetary policy will be a non factor for a long time due to the zero lower bound on nominal rates. Nothing new there. What is interesting, however, is that John Hatzius of Goldman Sachs, referring to the original Taylor formulation of monetary policy definition, extrapolates that based on data that the Fed is likely seeing, fund rates in 2010 "should" go negative, to the tune of -6% !

(If you don't care about the reasoning behind this conclusion, skip ahead to the next part).

A little math for those curious - Taylor's thesis is that the Fed Open Market Committee (FOMC) sets its funds rate target according to the rule:

i=r*+p+0.5(p-p*)+0.5(y-y*)

where i is the nominal fed funds rate, r* is the equilibrium real funds rate, p is the inflation rate, p* is the Fed's desired inflation rate, y is the (log) level of real GDP, and y* is the (log) level of real potential GDP. Using core PCE index for the inflation reading, and CBO estimates for real GDP (tangentially, Q4 2008 GDP number is out next Friday, consensus at -5.5%; GS says -5.9%, Zero Hedge is at -6%), and assuming 2% for both equilibrium real funds rate and desired inflation rate, equation implies -2% Fed funds rate for Q4 2010. Goldman takes the Taylor formulation one step further and replaces the GDP gap with the unemployment gap, and allows coefficients on the inflation and unemployment terms to be estimated based on empirical data, resulting in the following formula:

i=2.7+p+0.3(p-p*)-1.9(u-u*)

This relationship puts more weight on the GDP/unemployment gap, and implies a much lower funds rate of -6% for late 2010. Intuitively this makes a lot of sense: as debt is currently perceived as an overwhelming liability, the government should compensate people for accepting credit. Plus all the talk about stimulating lending will cease once people receive free money to borrow. Of course this is all a hypothetical argument - monetary policy, at least conventional, can not go below zero fed fund rates. Maybe unconventional policy is required.

So what type of response will be needed if Goldman's assumption is correct? If an "appropriate" funds rate is -6%, then "appropriate" LT Treasuries rates, mortgage rates, high grade (and maybe even high yield) corporate rates also have to be negative. In order to drive yields on these securities to much lower levels (if not necessarily negative yields) the Fed would need to engage in massive purchases of much lower quality assets, including HY bonds, non-agency mortgages, CMBS and even equities! And we are talking Massive, not what is contemplated in terms of current policy measures in either size or scale. Which explains why the fiscal response is critical (more on that later) - monetary stimulus is not only powerless to do anything at this point, but monetary logic implies that a fiscal push is needed now more than ever.

What are the practical implications? Deflation is staring us in the face. This is unavoidable, and will become more and more evident as consumers eliminate inflationary expectations from their behavior (i.e. not selling a house whose mortgage is underwater until "the market turns"; recent housing sales patterns in cities such as Greenwich already indicate this is happening). Additionally, the fiscal response will have to be much more pronounced if it is to be effective: GS is worried that the effective stimulus package in 2009 is only $259 billion, of which just $54 billion is infrastructure spending (more below). Unless the Senate bill expands significantly on the House version, it will be impossible to push the economy back to positive growth by the end of 2009.

So what are the key issues as a much more comprehensive fiscal response is contemplated? They can be summarized most neatly as follows: 1) severe shortage of capital; 2) opaque and illiquid balance sheets; and 3) complete policy uncertainty.

Severe shortage of capital. GS estimates that US institutions will bear loan losses of over $1 trillion (this number could be much larger as will be noted later). As such, the $250 billion of new bank capital provided by TARP last year and the $350 billion raised from private investors falls well short to cover these losses. Even assuming banks are able to exploit tax loss benefits, the system is undercapitalized. It is estimated that at least $200 billion is needed to return the system to a minimal acceptable ratio of tangible equity, and that much more than minimum is needed to spur lending to consumers and corporations. and to insure against possible larger losses if the base forecast is wrong.

Opaque and illiquid balance sheets. Bernanke said it best: "a continuing barrier to private investment in financial institutions is the large quantity of troubled, hard-to-value assets that remain on institutions' balance sheets." As Zero Hedge has harped on previously, the arcane asset marking methodologies used by different financial companies have made investors disbelieve any disclosure about "true" value. The GS Financial team recently estimated that major banks were holding subprime mortgage loans at values ranging from 57 to 90 cents on the dollar. The questions about balance sheet values only raises questions about bank solvency, making investors further unwilling to invest capital. The original TARP was geared at addressing this concern, by purchasing bank assets, however it was subsequently perverted into its current form emphasizing bank recapitalizations without addressing the balance sheet issue.

Policy uncertainty. As further government intervention is inevitable, its form and timing will have important consequences for financial firms, owners, creditors and the entire financial system. The biggest concern here have been the constant policy shifts in 2008 as the crisis spread. Of biggest import is that the government has been unable to make up its mind where in the capital structure of a troubled company it will inject capital, and as such, which balance sheet liabilities are safe for investors to participate in. The table below shows just how dramatic the lack of a clear intervention template has been.

The conclusion is that private capital will not be a major aid to the financial system until the uncertainty around regulatory flip-floping has diminished. As long as private capital is unwilling to be allocated, the government will be forced to shoulder the effort of recapitalizing the system alone.

So what is the correct approach to a fiscal, or otherwise, rescue?

The complexity of the problem, and its resolution, has been constantly highlighted by swoons in the market on any potentially adverse news, by the TARP's initial skeptical reception and its subsequent transformation to a bank recapitalization vehicle, by different models of dealing with the "systemic failure" risks, including the US, the British and the Swiss model, and many other aspects. Fed Chairman Bernanke has suggested three ways of dealing with bad assets: 1) an asset guarantee program along the lines of the UK approach, as was done in the "asset pool" creation at Citi and Bank of America, 2) a direct government purchase of assets, and 3) the creation of an "aggregator"/"bad bank" that would buy assets from financial institutions, which would receive shares in the bank and cash. First, however, key questions that must be answered before the correct policy approach is decided upon are among the following.

1. Just how much in troubled assets is out there. The answer depends on the scope one is willing to attribute to this, of course. The narrowest definition involves just the most troubled assets: subprime residential mortgage loans and securities, which would imply bad assets with a face value of roughly $1 trillion, of which half is held by US banks. If one expands the definition to include option-ARM loans, second liens, CMBS, credit card debt, and consumer auto loans or in other words all assets with a cumulative loss rate in the near double digits, the total face value of such asset would reach almost $10 trillion, of which $5.7 trillion is held by US banks. One can easily see the dilemma here: a narrower definition would limit upfront government payments and be easier to implement but would leave more residual uncertainty after the rescue is "complete."

2. How to value troubled assets. As most securities are marked to market already, despite attempts by many lawmakers to change the accounting rules, the emphasis here is on whole loans which are a major component of the mortgage market. And again a dilemma emerges: the more transparent market pricing mechanism policymakers decide upon, the larger the capital shortfall in the system will be. Valuation options include existing bank marks, "marking to model", a spin on Level 3 asset marks where asset managers and banks work together to estimate losses, the recently popular "hold-to-maturity" value, or most transparently, complete mark-to-market approaches which would result in the lowest valuations.

3. What to do with troubled assets. The choice is whether to keep the assets on the banks' balance sheets or to transfer them to government controlled, special purpose entities. The first choice implies a system of "ringfenced" guarantees, which has a low upfront government cost. We have previously written about this issue here. The more expensive option would leave banks with transparent balance sheets and might enable the government to modify loans more rapidly.

4. Where in the capital structure to inject capital. The government must choose whether to protect the taxpayer or the institution itself. If the common equity base is larger, the more losses can be absorbed without insolvency, however the more diluted and less insulated from losses taxpayers become.

5. How to deal with banks that are insolvent at realistic values for troubled assets. There are four main choices here: 1) recapitalization via government purchases of assets at the banks' marks: this would be most detrimental to taxpayers who have to fund the different between the artificial and real marks on bad assets; 2) direct recapitalization of banks while diluting shareholders, as was done with TARP V2; 3) debt-to-equity swaps (forced or otherwise); 4) forbearance, or allowing banks to earn their way back to adequate capital levels, which has the most risk if insolvency is a real risk. And then there is nationalization and shuttering: two options which few have breached but might become more and more attractive.

6. Who pays for all this. Main costs would be associated with asset purchases and recaps. It is likely that trillions in purchasing power will have to be made available, as that amount will definitely be necessary to address an median assumption of troubled assets around $3 trillion (see above) which is much, much more than has been authorized or is currently on the table. This means that policymakers will need to request much more from congress very soon, ironically contrary to what Tim Geithner has naively said will be the case, or have to decide on a strategy of asset guarantees, due to the lack of an upfront cost.

One things is sure: piecemeal, gradual, indecisive, subjective and nontransparent fiscal policy decisions will only exacerbate the problem, while increasing the final cost to taxpayers and investors, and prolonging the term of the current Great Depression V2.