House under construction

Housing Perspectives

Research, trends, and perspective from the Harvard Joint Center for Housing Studies

America’s Housing Finance System in the Pandemic, Part 3: Q&A on the Ugly Fight over Servicer Advances

4/16/20 – This version of the blog has been updated to reflect that there are policy requirements in the Ginnie Mae program to finance servicer advances.

There is a very ugly fight going on right now between the mortgage industry and the US government over the tens of billions of dollars of pandemic-related advances that mortgage servicers, especially non-bank ones, are projected to be required to make. It is far more contentious than anything I saw in my seven years running Freddie Mac. The fight reflects fundamentally different views of how the government takes extraordinary actions in times of extraordinary stress to “rescue” industries and markets for the greater good, but is so contentious because the survival of hundreds of companies in the mortgage industry is believed to be at risk, along with major disruption to the country’s housing finance system if they fail.

To understand the fight requires expertise about how the mortgage system works, and the nature of the topic does not lend itself well to being fully understood via typical daily press coverage. Therefore, in this piece, I will explain the issues behind the fight in a plain English Q&A format, concluding with some recommendations of how to move forward.

What are servicer advances?

Of the $10-plus trillion of single-family first mortgages outstanding in America, about 70 percent – or almost $7.5 trillion – is securitized. About $7 trillion of that amount is securitized through government-related entities, specifically through the two government-sponsored enterprises (GSEs) of Freddie Mac and Fannie Mae (about $5 trillion together) and through a direct government agency located in the Department of Housing and Urban Development known as Ginnie Mae (about $2 trillion). For the GSEs and Ginnie Mae, the investors in the resulting mortgage-backed securities (MBS) are promised that, every month, they will either get their principal and interest payments in full (or have the mortgage paid off), regardless of whether individual homeowners are paying or not. This guarantee of the credit risk through these three government-supported agencies is at the heart of how the housing finance system in America can deliver a 30-year fixed rate, free-prepayment mortgage to the broad middle and working class.

Mortgage servicers, who handle the payments and many other operational features related to securitized mortgages, are the entities required to temporarily advance funds to MBS investors to make up for any payments not received from borrowers, until the GSEs or Ginnie Mae intervene to pay the servicers back as part of their implementation of the guarantees they have given to the MBS investors. The rules by which this happens are different between Ginnie Mae and the GSEs, and even somewhat different between Freddie Mac and Fannie Mae. (There are also obligations to pay property taxes and home insurance when defaulting homeowners do not. This is a smaller amount to date and in the interest of simplicity will not be highlighted in this article.)

Why is this a major problem right now?

To help relieve stress on homeowners in the current pandemic-induced downturn, the government has announced that people with mortgages securitized through the GSEs and Ginnie Mae have the option to ask for “forbearance.” This means they can defer their monthly payments for up to six months (with a further six months as a possible extension) upon attesting to the fact that they are economically struggling, through job loss or other factors, to make their mortgage payments. No proof beyond this attestation is required, which is short-handed in the industry as “the honor system.” This forbearance option for distressed borrowers had been part of the mortgage system for a few years but little known or used until now. It was originally constructed with disruptions related to natural disasters in mind, and the deferred amounts were assumed to be repaid at the maturity of the loan so monthly payments could resume unchanged at the end of the forbearance period. (The forbearance program was enshrined in law in the CARES Act just passed; previously it was only mortgage agency policy.)

Given the depth of the current economic downturn, the industry has assumed that forbearances like this could apply to an unprecedented 25 percent of all mortgages. (This estimate seems a bit high, inconsistent with 1) the prediction of maximum 30 percent unemployment rate, which will impact homeowners far less than renters and 2) that the CARES Act is ameliorating so much current income distress associated with unemployment. One recent research report from the securities industry assumed a 10 to 20 percent take-up rate. The rate currently is 3.7 percent, up from 2.7 percent in the last week.) Such a large take-up of forbearance would generate the requirement that servicers advance funds to MBS investors in amounts far beyond what was contemplated when the forbearance program was developed in 2017-18. It could be in the tens of billions of dollars, depending upon how long and deep the economic downturn will be and how much lost household income the government will replace via various programs.

For servicers that are banks, the burden of this cash requirement should be, in most cases, absorbed because banks have many sources of liquidity, including the ability to borrow from the Federal Reserve. Servicers that are not banks (about half of the industry) are more limited in their access to sources of liquidity; these are arranged by each individual servicer with its banks or other financial firms.

It is these non-bank servicers which are at the heart of the fight between the industry and the government. If not helped by the government to face this unprecedented call upon them to advance funds, the majority is believed (under the assumption of a 25 percent forbearance take-up rate) likely to collapse en masse, leading to bankruptcy and destabilization of the mortgage system, with major negative impacts on homebuilding and homeowners. (Even a lower rate of forbearance take-up in the 10 to 20 percent range will probably be enough for destabilization.)

What does the industry propose as a solution?

The industry is proposing a single, extraordinary action by the government to address the problem: a line of credit at the Federal Reserve where servicers can borrow the funds needed to make the required advances. The general concept is for the line to finance 100 percent of their need, and cover not only the advances to replace principal and interest on all Ginnie Mae and GSE loans, but also make property tax and homeowner insurance payments that need to be made when the homeowner is not paying, as well as other smaller items. The industry proposal does not get into specifics of the interest rate or other terms for the line of credit; those presumably will be developed over time, and articles from various individuals and organizations, with recommendations as to what those specifics should possibly be, are now appearing. The industry justifies a solution of this type – which amounts to a hold-harmless response – because the problem is not of their making but instead stems directly from the government-imposed solution of forbearance for six or twelve months. In other words, because they are “innocent bystanders.”

How does this solution compare to other extraordinary actions taken by the government?

It would be quite unusual, in two ways.

First, as a matter of history, extraordinary actions by the Federal Reserve, including most being implemented right now, broadly take the form of buying bonds in the secondary market to counter panic selling and restore some market normalcy, rather than lending money directly to individual companies (with the obvious exception of lending to banks). Under the pressure of the sharp and major downturn taking place right now, the Federal Reserve will get closer to doing just such lending, especially given certain requirements by the CARES Act, but it will generally do so as little as possible. I note the Federal Reserve does not intend to take credit risk in such extraordinary actions – Treasury provides the funds to absorb any losses from that risk. And yet the industry proposal is for there to be such direct lending by the Federal Reserve on a major scale to hundreds of non-bank servicers, almost all of which are privately held.

Second, and again as a matter of history, when the Treasury or the government utilizes taxpayer funds to help an individual company or companies in an industry, the actions have been a last resort at significant “pain” to those companies – sometimes extremely significant. As but one example, the rescue in 2009 of General Motors and Chrysler (which did not create the financial crisis and therefore can claim “innocent bystander” status) required they go through bankruptcy, wiping out their shareholders’ investment. Such rescues always seem to be accompanied by other policy-based restrictions placed on the rescued firm, such as limiting executive compensation or layoffs, or the taking of an equity position so that the taxpayer shares in the recovery of the company being rescued, to compensate for the risk taken. The industry’s proposal has no such painful or policy-based restrictions mentioned.

What is the government’s view of the industry’s proposed solution?

While the government has not spent much time explaining its views on the industry proposal, leaving the public to hear only one side of the story, there has been enough communication nevertheless to discern what its views are, given a significant knowledge of the history of other government rescues in distressed situations. There are five key points that summarize that view, as best as I can determine them:

  1. There is full agreement there is likely a major problem to solve and that it would be very bad for housing and financial markets to have large swaths of the non-bank servicer industry collapse and declare bankruptcy.
  2. However, putting in place the broad-based program suggested by the industry is premature, as the need for advances has not yet materialized at the size which would necessitate a rescue of the nature suggested. (Forbearances today are estimated by the industry to be only about 3.7 percent of Ginnie Mae/GSE loans versus the assumed 25 percent.) The gap in timing between a forbearance being granted and servicers needing to make advances is large enough that any rescue actions can be sized to fit the actual need as it develops and not preemptively based upon an assumption in order to calm industry nerves.
  3. The proposal by the industry goes well past what is a rescue for the greater good of the housing finance system and economic recovery, and instead looks like a full and complete bailout of the owners of every single servicer impacted, regardless of how poorly capitalized or illiquid they might be. This is not acceptable from a taxpayer perspective.
  4. The proposal has no policy restrictions placed upon the servicers that would receive extraordinary support from the government, such as on layoffs or dividends. This is also not acceptable from a taxpayer perspective, and is contrary to all the other rescues taken in recent memory. (And, historically, other supposed “innocent bystanders” have been required to have policy restrictions or other penalties placed upon them in exchange for taxpayer funded rescues.)
  5. The proposal is contrary to the policy and tradition that government funds be used only as a last resort; it instead seems to be using them as a first resort, with no requirement that private capital sources cover some portion of the required advances.

In summary, as best as I can tell, the government’s view is that the industry’s proposal is very unbalanced: significantly too generous to the industry, dismissive of the taxpayer’s interest, and also somewhat premature.

So, what action is the government actually taking?

As background, the amount of funds needed to make advances, as forbearances grow, fall into three “buckets”: 1) principal and interest on Ginnie Mae loans, 2) principal and interest on GSE loans, and 3) all the other items (most prominently property taxes and insurance on defaulted loans). While there is no definitive measure of how big each bucket is – some will start small but grow large where others will start larger but not grow so much, and there are other measurement complications as well – the first two buckets are generally regarded as the large ones, and the third is materially smaller.

The government has already directly addressed one of the two larger buckets; Ginnie Mae, citing obscure legal authority, has already announced and is beginning to implement a program to cover the advances on principal and interest payments being made by non-bank servicers, which provide the vast majority of servicing on its loans. The non-bank servicers will have to individually put this in place with Ginnie Mae. Reflecting the extraordinary nature of this program, it has several reasonable  (from the taxpayer’s perspective, if not the industry’s) policy requirements that prohibit 1) dividend or equivalent payouts to shareholders and 2) increases in executive compensation.

Is that all the government is doing?

Unfortunately, it’s not clear. On April 8, the Director of the FHFA, Mark Calabria, gave press interviews in which he seemed to announce a program to address the second of the two large buckets identified above: GSE principal and interest advances. I say “seemed” because it was a very poor-quality communication, widely criticized in the industry and the housing press. He simultaneously announced two things:

  • First, that any non-bank servicer could go to Freddie Mac and/or Fannie Mae to request aid in making advances on that GSE’s loans if they still needed such help after utilizing their existing private market sources of liquidity supplemented by the just-announced Ginnie Mae program (referenced above). This could take one of two forms: 1) transferring servicing away to another (presumably bank-affiliated) servicer, or 2) entering into an agreement on some or all of their serviced GSE loans to sell that servicing to the relevant GSE, which would then turn around and enter into a sub-servicing contract with that servicer on the same loans, with compensation on the usual sub-servicing fee-for-transaction basis. Both alternatives relieve the non-bank servicer from having to make forbearance advances and there are existing commercial terms upon which to do such transactions. (Note: Sub-servicing is quite common; the largest non-bank servicer just announced that its servicing book was in fact slightly over 50 percent in the form of being a sub-servicer.)
  • Second, that he (i.e. Director Calabria) had a very strong disposition and attitude towards doing less rather than more in providing aid, as the GSEs could only play a limited role because they were so thinly capitalized, and he had to be concerned for their own safety and soundness. He also indicated a bias toward a program of servicing transfers (again, presumably to bank-affiliated servicers) rather than sub-servicing.

The entire communication effort was muddled – the noise of the latter comments, which were poorly received in the industry, virtually drowned out the considerable import of the former. Many people did not recognize that sub-servicing (which is far less economically disruptive to servicers and homeowners than a servicing transfer) was even an alternative. As a result, given the confusion of the FHFA communication, it is not clear right now exactly what the government is doing to relieve need for cash by non-bank servicers with respect to the GSE principal and interest bucket.

How does this all stand now?

Instead of the government instituting a single extraordinary action via a Federal Reserve line of credit, as proposed by the industry, the administration instead seems to be implementing a strategy of utilizing multiple limited programs that address specific parts of the non-bank servicer’s need for liquidity. Presumably, the government has a willingness to grow or add to these limited programs over time if and when the need arises based upon actual, rather than projected, forbearances. Such programs, of course, appear less “major rescue” in terms of their image.

Using this strategy, we know that the first of the two major buckets of cash need – Ginnie Mae principal and interest payments – seems to be adequately addressed. If there is additionally a robust program at the GSEs (the sub-servicing or servicing transfer alternatives announced on April 8), then a mass failure of non-bank servicers should be off the table – as both of the two major buckets would have been addressed (and the third, smaller bucket can likely be addressed for now by private capital sources of liquidity each firm should have in place). However, if the GSE programs are more theory than fact due to Director Calabria’s stated bias to help less rather than more, then mass non-bank servicer failures are not yet off the table – which has the industry in an understandable uproar and also represents a systemic risk.

Predictably, this has all become quite political. The mortgage industry is now engaged in a major lobbying and public relations effort, with elected officials in Congress calling for the Federal Reserve line-of-credit solution and articles in industry press and think-tank journals almost overwhelmingly sympathetic to the industry, especially in reaction to Director Calabria’s recent comments. In short, the industry is heavily engaged in – and seemingly winning – a PR war.

And then, late last week, the Federal Reserve, when putting out information about new initiatives to aid the financial system, specifically omitted a line of credit for mortgage servicing advances. When questioned about this, the response was that it was monitoring the situation, which is consistent with the notion that the government’s strategy is to help via several programs as needed, rather than being based upon a single extraordinary action rescue by the Federal Reserve.

Can this all be resolved and, if so, how?

Yes, it can be resolved – and it needs to be. There are two fundamental paths to do so.

One path for resolution is to effectively implement the GSE programs for sub-servicing or servicing transfer (or maybe something similar to what Ginnie Mae is doing), and make it clear that, despite Director Calabria’s earlier comments, the program will be robust and designed to fully address the burden of advancing principal and interest on GSE loans. This may be a climb-down for the FHFA, but it does seem necessary to take off the table the threat of mass bankruptcies by non-bank mortgage servicers.

A statement by Treasury that more aid would be forthcoming if forbearances grew to the upper range of expectations would additionally help fully explain the government’s strategy and calm the nerves of the industry, since so many servicers are worried that they will not survive through the end of the year.

This path is one where, strategically, the industry gets a great deal of relief – but clearly not 100 percent – through different (and somewhat paperwork-intensive) programs which are established over time as the actual size of the problem becomes known. It also has the image of being based more on commercial transaction structures, rather than something easily categorized as a “bailout.”

A second path for resolution is for the administration to change its strategy and agree to the Federal Reserve line of credit alternative, but with the addition of usual extraordinary-action policy restrictions rather than the hold-harmless proposal of the industry. Such conditions could include burden sharing (e.g. that the line of credit would finance perhaps 90 percent of advances made, with the other 10 percent coming from private market sources) and also policy requirements such as those called for under the CARES Act, like a limitation on layoffs or executive compensation.


It is not good for the economy or homeowners or anyone for the ugly fight between the mortgage industry and the government to continue. The government should recognize that its inadequate communications have taken a difficult situation and made it a festering problem that needs to be addressed promptly, especially remembering that hundreds of mortgage servicers are facing an existential threat in the next few months. The industry should recognize it has overplayed its hand and reduced its credibility with the government by looking for such an unprecedentedly-generous rescue, i.e. a full 100 percent hold-harmless line of credit at the Federal Reserve that looks too much like a rescue of every single owner of a non-bank servicer rather than a rescue of the mortgage system and the economy – and with no policy restrictions to boot.

If both can come down from their positions, taking one of the routes of resolution I have outlined above, the ugly fight can hopefully be remedied in relatively short order.