The previous post listed several reasons that the PPIP would best be implemented by simply having the government purchase loan pools directly, modify any defaulting loans, and then issuing notes based on the loan pools. In addition to the reasons set forth in that previous post, there is one additional, very significant role that the government can play in addressing the problem of toxic assets that no other entities can effectively accomplish. The issue of pricing the toxic assets has apparently not yet been seriously addressed, but it will be addressed sooner or later. However, as many have noted, there is a wide gulf between estimates placed on the value of toxic assets by banks and estimates placed on the value of those assets by potential buyers. Given that gulf, it is unlikely that many assets will be sold. However, it was also noted in the previous post that the problems causing difficulty in valuing those assets can be sensibly addressed, and that sensible analyses can then be performed to better estimate the value of those toxic assets. Of course, there may still be resistance by parties on either or both sides to accepting any values so calculated. If the PPIP relies on private parties to agree on mutually acceptable prices for the toxic assets, even more accurate valuations may not result in many successful purchases of those toxic assets.
However, if the government intervenes to purchase the toxic assets directly, then reticence to accept the estimated values placed on the toxic assets will not unnecessarily hinder the purchase of those toxic assets. The government has at its disposal two powerful tools for encouraging even reticent banks to sell their toxic assets at fair prices. For the time being, I will not elaborate on either of those options. Suffice it to say, though, that either tool could effectively break the logjam that we may experience if the private parties and the banks cannot find mutually acceptable price estimates for the toxic assets.
Wednesday, April 22, 2009
The Elusive Details for the PPIP
Last month the Treasury Department unveiled the general parameters of its plan to address the toxic assets dragging down banks' balance sheets. Since that time, Treasury has been trying to figure out the particular details and structure through which to effectively implement the plan. While the PPIP plan proposes to rely on private investors to relieve the banks of their bad assets, there are several major problems that just have not been addressed. First and foremost, and a point which, curiously, is being not being directly addressed, is the fact that simply moving the toxic assets from one place to another will do nothing to alter the toxic character of those assets. Second, and the issue upon which most people have focused, is the challenge of pricing the toxic assets to be moved. And third, there are challenges in developing an effective mechanism for permitting investment in those assets.
One significant issue that nobody seems to be discussing directly is the fact that merely moving the toxic assets from one place to another will not change the fact that they are toxic. This point is always tacitly subsumed into discussions about finding an appropriate price for those assets. But it is better to address these considerations separately. First, it is unclear just how vast the problem of toxic assets is, although some estimates have been made. Second, it is unclear just how toxic those assets are (i.e. how much a particular loan will perform, if at all, or how many loans in a securitized pool will fail, affecting the value of the securities). Both of those uncertainties will affect any attempted valuations of toxic assets.
As an example, information recently released by one particular bank is rather instructive. Despite the fact that the bank in question recently reported a first quarter gain of $4.25 billion, its stock plunged 24% on that same day. Investors and analysts saw past the glitz and identified numerous troubling signs at the bank. Of predominant importance for this discussion was the year-over-year increase in non-performing assets from 0.9% to 2.65%. At the same time, the bank increased its loan loss reserve to $13.4 billion, an increase of 57% above its fourth quarter reserve. However, due to the increasing failure of loans, the ratio of the loan loss reserve to non-performing loan value actually dropped from 1.41 to 1.21. Yes, despite a 57% increase in the loan loss reserve, the ratio of the reserve to bad loans actually fell by 14%. (See articles from the Wall Street Journal, April 21, 2009: http://online.wsj.com/article/SB124021187032334351.html and http://online.wsj.com/article/SB124027011654636755.html.) This demonstrates that the breadth of the toxic asset problem has not yet been defined and that a significant portion of assets are continuing to become toxic. Due to the fact that questionable loans were originated through the first quarter of 2007 and that most of those loans had 3-year reset provisions, it can be anticipated that the toxicity of assets will continue to spread well into 2010. While subprime borrowers are likely to default quickly once the terms of their loans reset, prime borrowers may succeed in staving off default for some time by liquidating other assets. This may result in a "slow roll" wave of defaults by prime borrowers. The deterioration of the assets held by this bank - or any other bank - will not cease if those assets are simply transferred elsewhere.
Given the uncertainty in determining how many assets will may become toxic and the depth of that potential toxicity, it would be very difficult to try to place a value on those assets. Indeed, articles have consistently reported a wide disparity in the values placed on such assets by banks and by investors. Many articles report bank valuations in the range of 60-80 cents on the dollar for such assets, while potential investors place their value at closer to 20-35 cents on the dollar. With such a wide gulf between the estimated values reached by the parties to these ostensible transactions, it is highly unlikely that many sales of the toxic assets will be completed.
That brings me to the second issue: that of determining accurate values for the toxic assets. If the defaulting loans that comprise those toxic assets (or elements of loan pools on which toxic assets are based) are modified, then the breadth and toxicity of those toxic assets can be contained. This is not to say that some loans will not continue to default, but they will do so in a much more predictable pattern. Banks routinely use prepayment curves and loan default curves to estimate how many loans in a healthy loan pool are likely to prepay or default over the lives of the instruments. By relying on these estimates, the banks can closely estimate the actual performance of the loan pools, and therefore the anticipated revenue that will be generated from such loan pools. This enables those banks to place values on the assets based on those loan pools. Of course, in a bad economic environment, like the current recession, the loss of jobs means that more borrowers are likely to suffer economic hardships, and therefore to default on their loans. However, estimates can also be made to gauge the impact of these job losses on the performance of loan pools, and thereby their effect on the values of assets based on those loan pools. This may be a challenging assessment, but should be eminently feasible if conducted with diligence, intelligent assumptions, and careful analysis of relevant information from similar prior economic circumstances.
Finally, there remains the real problem of finding the proper vehicle for enabling investors to invest in these toxic assets. It has recently been pointed out that there are some regulatory prohibitions that might prevent the establishment of mutual fund-like investment funds to facilitate private investment in such assets. Currently, regulations prohibit open-end funds (like mutual funds) from investing more than 15% of their funds in illiquid assets. There is some question as to whether closed-end funds would be permitted to invest 100% of their funds in such illiquid assets. Perhaps this would be permitted, perhaps not.
As an alternative, I would propose that the government establish these funds for several reasons. First, the PPIP already proposes that most of the money used to purchase the toxic assets would come from the government. The government would stand to earn some profit if the assets recover value above that for which they are purchased, but it would also bear substantially greater downside risk if the assets do not recover. If the government simply purchases the assets in the first instance, it will commit only a small additional proportion of funds, but will obtain a much better balance of potential upside benefit against the downside risk. Second, the government could much more easily marshal all of the assets in a single mortgage pool than could private investors. This is significant because it would facilitate the modification of toxic assets to render them non-toxic, thereby making it more likely that the assets would continue to perform, and making it much easier to value those assets. Third, the government could then sell government-guaranteed shares in these pools directly to investors, essentially skipping the middleman. Demand for other types of government notes is extremely high right now, and it is likely that investors would also feel secure in purchasing guaranteed notes from these loan pools. As a significant additional benefit, the sale of those notes would replenish the government's coffers, enabling it to target additional mortgage pools, modify those toxic assets, and perpetuate the process.
One significant issue that nobody seems to be discussing directly is the fact that merely moving the toxic assets from one place to another will not change the fact that they are toxic. This point is always tacitly subsumed into discussions about finding an appropriate price for those assets. But it is better to address these considerations separately. First, it is unclear just how vast the problem of toxic assets is, although some estimates have been made. Second, it is unclear just how toxic those assets are (i.e. how much a particular loan will perform, if at all, or how many loans in a securitized pool will fail, affecting the value of the securities). Both of those uncertainties will affect any attempted valuations of toxic assets.
As an example, information recently released by one particular bank is rather instructive. Despite the fact that the bank in question recently reported a first quarter gain of $4.25 billion, its stock plunged 24% on that same day. Investors and analysts saw past the glitz and identified numerous troubling signs at the bank. Of predominant importance for this discussion was the year-over-year increase in non-performing assets from 0.9% to 2.65%. At the same time, the bank increased its loan loss reserve to $13.4 billion, an increase of 57% above its fourth quarter reserve. However, due to the increasing failure of loans, the ratio of the loan loss reserve to non-performing loan value actually dropped from 1.41 to 1.21. Yes, despite a 57% increase in the loan loss reserve, the ratio of the reserve to bad loans actually fell by 14%. (See articles from the Wall Street Journal, April 21, 2009: http://online.wsj.com/article/SB124021187032334351.html and http://online.wsj.com/article/SB124027011654636755.html.) This demonstrates that the breadth of the toxic asset problem has not yet been defined and that a significant portion of assets are continuing to become toxic. Due to the fact that questionable loans were originated through the first quarter of 2007 and that most of those loans had 3-year reset provisions, it can be anticipated that the toxicity of assets will continue to spread well into 2010. While subprime borrowers are likely to default quickly once the terms of their loans reset, prime borrowers may succeed in staving off default for some time by liquidating other assets. This may result in a "slow roll" wave of defaults by prime borrowers. The deterioration of the assets held by this bank - or any other bank - will not cease if those assets are simply transferred elsewhere.
Given the uncertainty in determining how many assets will may become toxic and the depth of that potential toxicity, it would be very difficult to try to place a value on those assets. Indeed, articles have consistently reported a wide disparity in the values placed on such assets by banks and by investors. Many articles report bank valuations in the range of 60-80 cents on the dollar for such assets, while potential investors place their value at closer to 20-35 cents on the dollar. With such a wide gulf between the estimated values reached by the parties to these ostensible transactions, it is highly unlikely that many sales of the toxic assets will be completed.
That brings me to the second issue: that of determining accurate values for the toxic assets. If the defaulting loans that comprise those toxic assets (or elements of loan pools on which toxic assets are based) are modified, then the breadth and toxicity of those toxic assets can be contained. This is not to say that some loans will not continue to default, but they will do so in a much more predictable pattern. Banks routinely use prepayment curves and loan default curves to estimate how many loans in a healthy loan pool are likely to prepay or default over the lives of the instruments. By relying on these estimates, the banks can closely estimate the actual performance of the loan pools, and therefore the anticipated revenue that will be generated from such loan pools. This enables those banks to place values on the assets based on those loan pools. Of course, in a bad economic environment, like the current recession, the loss of jobs means that more borrowers are likely to suffer economic hardships, and therefore to default on their loans. However, estimates can also be made to gauge the impact of these job losses on the performance of loan pools, and thereby their effect on the values of assets based on those loan pools. This may be a challenging assessment, but should be eminently feasible if conducted with diligence, intelligent assumptions, and careful analysis of relevant information from similar prior economic circumstances.
Finally, there remains the real problem of finding the proper vehicle for enabling investors to invest in these toxic assets. It has recently been pointed out that there are some regulatory prohibitions that might prevent the establishment of mutual fund-like investment funds to facilitate private investment in such assets. Currently, regulations prohibit open-end funds (like mutual funds) from investing more than 15% of their funds in illiquid assets. There is some question as to whether closed-end funds would be permitted to invest 100% of their funds in such illiquid assets. Perhaps this would be permitted, perhaps not.
As an alternative, I would propose that the government establish these funds for several reasons. First, the PPIP already proposes that most of the money used to purchase the toxic assets would come from the government. The government would stand to earn some profit if the assets recover value above that for which they are purchased, but it would also bear substantially greater downside risk if the assets do not recover. If the government simply purchases the assets in the first instance, it will commit only a small additional proportion of funds, but will obtain a much better balance of potential upside benefit against the downside risk. Second, the government could much more easily marshal all of the assets in a single mortgage pool than could private investors. This is significant because it would facilitate the modification of toxic assets to render them non-toxic, thereby making it more likely that the assets would continue to perform, and making it much easier to value those assets. Third, the government could then sell government-guaranteed shares in these pools directly to investors, essentially skipping the middleman. Demand for other types of government notes is extremely high right now, and it is likely that investors would also feel secure in purchasing guaranteed notes from these loan pools. As a significant additional benefit, the sale of those notes would replenish the government's coffers, enabling it to target additional mortgage pools, modify those toxic assets, and perpetuate the process.
Tuesday, April 7, 2009
The Effect of Loan Modifications
Last Friday the Office of Thrift Supervision released a report about the effectiveness of loan modifications that should not have surprised anybody. The upshot of the report found that modifications that did not reduce monthly mortgage payments resulted in re-defaults in about 51% of cases, but that loan modifications that reduced borrowers' monthly mortgage payments resulted in a substantially lower rate of re-default, either 23% or 26%, depending on source. It should not take a government study to demonstrate the correlation between lowering monthly mortgage payments and lower rates of re-default on loans, but at least the measure of effectiveness is useful.
The report found that many loan "modifications" were actually forbearances, meaning that they did not alter the terms of the loan, but simply permitted borrowers to make missed payments at a later date, or factored already-missed payments into the borrowers' current monthly payments. The latter scenario actually increases borrowers' monthly payments. If the borrowers were already having trouble making payments, it should be no surprise that the failure to lower those payments does not alleviate the burden on the borrowers. As stated above, the report found that about 51% of borrowers receiving such modifications fell back into default within six months.
By contrast, loans that were truly modified to result in lower monthly payments for the borrowers resulted in much lower rates of re-default. When loans were modified so that the monthly mortgage payments were lowered by 10% or more, the rate of re-default six months later was only about 26%. That is half the percentage of re-defaults of the other category. Most of the articles describing the study did not specify how the loans were modified, but it appears to have been accomplished through lowering the interest rates and extending the life of the loans. In many cases, it appears that the monthly payments were kept lower by reducing interest rates. (See press release from the Office of the Comptroller of the Currency from April 3, 2009, listed at: http://www.occ.treas.gov/ftp/release/2009-37.htm.)
The 10% reduction to monthly payments appeared to be an arbitrary reduction level that is not tied to any particular level of the borrower's income. In such a case, it is not surprising that such a high percentage of even the modified loans continue to re-default. Many people have recently advocated identifying viable levels of mortgage payments based on the borrowers' income and then adjusting monthly payments to those levels. Such an approach is much more likely to establish viable levels for borrowers and to prevent re-defaults. However, many of those plans require lenders to forfeit some amount of principal from the loan in order to reach the targeted payment levels. Once again, I submit that my plan would facilitate the lenders' efforts to reduce borrowers' monthly mortgage payments, but would compensate the lenders in a fair manner for writing down the principal of the loans.
The report found that many loan "modifications" were actually forbearances, meaning that they did not alter the terms of the loan, but simply permitted borrowers to make missed payments at a later date, or factored already-missed payments into the borrowers' current monthly payments. The latter scenario actually increases borrowers' monthly payments. If the borrowers were already having trouble making payments, it should be no surprise that the failure to lower those payments does not alleviate the burden on the borrowers. As stated above, the report found that about 51% of borrowers receiving such modifications fell back into default within six months.
By contrast, loans that were truly modified to result in lower monthly payments for the borrowers resulted in much lower rates of re-default. When loans were modified so that the monthly mortgage payments were lowered by 10% or more, the rate of re-default six months later was only about 26%. That is half the percentage of re-defaults of the other category. Most of the articles describing the study did not specify how the loans were modified, but it appears to have been accomplished through lowering the interest rates and extending the life of the loans. In many cases, it appears that the monthly payments were kept lower by reducing interest rates. (See press release from the Office of the Comptroller of the Currency from April 3, 2009, listed at: http://www.occ.treas.gov/ftp/release/2009-37.htm.)
The 10% reduction to monthly payments appeared to be an arbitrary reduction level that is not tied to any particular level of the borrower's income. In such a case, it is not surprising that such a high percentage of even the modified loans continue to re-default. Many people have recently advocated identifying viable levels of mortgage payments based on the borrowers' income and then adjusting monthly payments to those levels. Such an approach is much more likely to establish viable levels for borrowers and to prevent re-defaults. However, many of those plans require lenders to forfeit some amount of principal from the loan in order to reach the targeted payment levels. Once again, I submit that my plan would facilitate the lenders' efforts to reduce borrowers' monthly mortgage payments, but would compensate the lenders in a fair manner for writing down the principal of the loans.
Subscribe to:
Posts (Atom)
