Home > News & Events > Conferences & Events > FDIC:
New York Region Regulatory Teleconference
FDIC New York Region Regulatory Teleconference:
HMDA Data Validation Procedures
Thursday, June 16, 2011
Moderator: George Small
Coordinator: Welcome and thank you for standing by. At this time all participants are in a listen only mode. During the question and answer session please press star then 1 on your touchtone phone to ask a question. Today’s conference is being recorded.
If you have any objections you may disconnect at this time. And now I’d like to turn the meeting over to John Vogel, Deputy Regional Director. Sir, you may begin.
John Vogel:Thank you. Thanks (Marla) and thank you everyone for joining us today. As (Marla) mentioned my name is John Vogel and I’m Deputy Regional Director in the New York Regional Office and it’s my pleasure to welcome you to today’s regulatory conference call.
Today’s 90 minute teleconference will focus on the Home Mortgage Disclosure Act or HMDA, validation procedures and discuss the importance of HMDA data accuracy and the regulatory expectations for HMDA compliance.
In addition, we’ll cover some common pitfalls and violations that examiners in our region have identified in the recent past. And we’ll discuss some of the best practices for insuring compliance.
Lastly we’ll go over some information on the HMDA (large) resubmission process and on the assessment of civil money penalty. There will be a question and answer session following the formal presentation.
And please note that you can also email questions at any time during the presentation to NYCalls and that’s one word, NYCalls@FDIC.gov. At the end of the presentation the operator will provide procedures for calling in questions.
We appreciate very much your participation on today’s call. Your telephone confirmation notice included the PowerPoint presentation for today’s topic. The slides should assist you in following today’s presentation and can be used for future reference.
Joining me today to deliver this presentation is senior compliance examiner, (Erin Skillman) and financial institution examiner, (Eliyah Shin). Thanks once again for participating in this call. And without further ado, I will turn the program over to (Eliyah). Thanks.
(Eliyah Shin): Thank you again. Good afternoon everyone. First I would like to quickly go over the agenda for today’s call and that can be found on slide number two. I will start by going over some background information on HMDA data such as the purpose and importance of data accuracy.
I will then go over the new data validation procedures and what institutions can expect in relation to the new HMDA data validation testing.
(Erin) will then cover the second half portion of today’s presentation and she will go over some common HMDA data violations that examiners have detected during the on site exams. In addition, she will also provide some information that may help in preventing those common HMDA data errors.
Lastly (Erin) will walk through the steps involved during the resubmission process and some information regarding the assessment of civil money penalties. Once again, I would like to remind everyone that - everyone to take note of any question that may come up during the presentation.
And as our deputy regional director stated, there will be a question and answer session at the end of the presentation so you can submit your questions to the email address provided for this presentation.
So to start please turn to slide number 3 as we will go over the purpose behind HMDA data and what accurate data mean to regulators and other such as private investors and various community organizations.
For regulators or examiners like ourselves, HMDA data are heavily depended upon and accurate HMDA data are necessary to draw accurate conclusions about an institution’s lending performance, specifically the bank CRA performance during the examination period.
HMDA data are heavily monitored by regulators to identify any potential failed lending risk at an institution as well. In addition, the HMDA data reported by your institution becomes a part of our aggregate data which are used for comparison purposes.
One example that I can provide of such data comparison is when examiners are evaluating an institution’s CRA performance. During a CRA evaluation examiners measure on institutions’ lending performance to peer or aggregate lenders’ lending data.
Now although various factors are taken into consideration when assigning a rating for CRA performance evaluation, one of the most important factors considered is how an institution fares with its peers or other lenders in terms of lending in certain geographies or lending to certain borrowers.
All of this important information derives from HMDA data reported by institutions. So just from the CRA standpoint alone it is important for examiners to have accurate HMDA data.
Otherwise an institution may be assigned with a CRA rating that may be unfavorable or unfair such as being rated low satisfactory under lending tests when accurate HMDA data may have resulted in an assignment of high satisfactory rating.
Another reason why HMDA data are important is because they are valuable to the public and data argues by various government entities and officials to evaluate housing activities.
Most notably, Federal Reserve uses HMDA to monitor housing and lending activities in order to determine whether housing market activity has strengthened or weakened.
In addition, HMDA data are used by private investors and community organizations to identify areas in need of investment.
Among other things these entities use HMDA data to identify certain neighborhoods such as low and moderate income areas to see if local institutions are meeting the market’s credit opinions. Moving onto the next slide, slide number four.
So those were some reasons why HMDA data and accuracy are important to regulators and other entities. But I would like to emphasize a point that accurate HMDA data are just as important to institutions that report those data.
The benefits from maintaining accurate HMDA data can be seen from two major - once again I apologize for the technical errors. I believe I left off at slide number 4. I was about to discuss some of the benefits for institutions for maintaining accurate HMDA data.
Again, the benefits from maintaining accurate HMDA data can be seen from two major perspectives starting from the CRA perspective. An institution can benefit from conducting an internal analysis of its lending pattern.
Specifically by using the accurate HMDA data a compliance officer at an institution can analyze the institution’s lending performance by geography and by borrower’s income.
In addition, the HMDA data can also be used to measure the institution’s community development lending activities such as loans secured by multi family dwellings that are located in a low or moderate income area.
And with the results of the internal analysis using the HMDA data and comparing the results to the institution’s most recent CRA performance evaluation or even the CRA performance evaluation of other institutions in the market area you can learn more about how HMDA lending activity affects your institution from a CRA perspective.
In addition, by conducting such internal analysis using HMDA data you can also detect potential weaknesses in terms of HMDA lending performance well ahead of your next scheduled CRA evaluation. This may even help in minimizing a risk of poor CRA rating.
For instance, by proactively conducting internal analysis using the HMDA data, a compliance officer or management may be able to tell that the institution fell short in terms of lending let’s say in a low income area when compared to institutional peers or to the previous CRA performance evaluation.
And by identifying such a weakness prior to the next CRA performance evaluation management can take appropriate actions such as implementing an outreach program designed to lend in those low income areas and possibly boost the institution’s lending performance.
We can also see some benefits from a fair lending perspective as shown on slide number 5. Just as it was with CRA, a compliance officer by using the accurate HMDA data, can conduct an internal fair lending analysis to identify any potential risk.
Disparity in HMDA data may alert management of potential discriminatory lending patterns. To illustrate, a relatively high number of declined applications based on rate, sex or other prohibited basis will warrant management to conduct further investigation.
And consider implementing an outreach program, strengthen the secondary review programs for denied applications or take other appropriate measures.
And in addition, by plotting the HMDA data on the map a compliance officer can detect any gap in lending inside or just outside the defined assessment area. Such gap in lending may warrant additional research by management as it can pose a potential risk of red lining.
By conducting internal analysis using the accurate HMDA data, an institution can minimize certain fair lending risks including a risk of being categorized as a HMDA outlier. Now today I won’t go into too much detail about the FDIC HMDA outlier fair lending project.
But I would like to go over some key points about the project and how HMDA data plays a significant role. The outlier project began with the review of 2004 HMDA data to identify institutions with a higher risk of pricing or denial discrimination.
When identifying those institutions that are outliers, FDIC fair lending team of economists, they rely on their reported HMDA data to identify institutions with large pricing or denial rate disparities. So in other words, HMDA serves as a starting point for identifying such institutions.
In most cases the historical show that the outlier institution has received increased scrutiny from regulators as they determine whether large disparities are due to legitimate pricing or underwriting factors or a discrimination.
Although a small number of FDIC supervised banks are identified as outliers, a small number of those outliers actually result in discriminatory findings or are in violation of fair lending rules and regulations.
So does that mean your institution is in a lot of trouble if the institution is categorized as an outlier? Well not necessarily, because a large percent of outlier reviews result in no evidence of discriminatory pricing or underwriting.
However, the key point here is that it never hurts to be not designated as an outlier. And furthermore, inaccurate HMDA can lead to your institution being incorrectly designated as an outlier. To illustrate, one example that I can provide is from 2004.
Based on the 2004 HMDA pricing data FDIC identified 47 institutions as outliers. However, further HMDA data validation process review that the pricing disparities for three out of those 47 outlier institutions resulted from the bank’s filing of incorrect HMDA data.
So after the data was corrected the disparities have disappeared. So in other words, those outliers - those three outliers were designated as outliers when in fact they were not.
Moving onto the next slide, so I discussed some benefits of maintaining accurate HMDA data and I started to touch on some negative impacts from inaccurate HMDA data. Slide 6 goes into detail about some additional negative impacts on inaccurate HMDA data.
Again, the first bullet point indicates inaccurate HMDA data can reflect an appearance of discriminatory lending at an institution when in fact that is not the case.
It is important to note the fact that HMDA data are public and the data are not just monitored by regulators but they are also monitored by various consumer protection groups.
These consumer protection groups have been using more sophisticated software to analyze HMDA data in order to identify unusual patterns and trends in lending by local financial institutions.
So you probably don’t want to be surprised by their findings, especially if their findings derive from inaccurate HMDA data. Furthermore, if your institution conducts internal analysis for CRA or fair lending compliance it is critical that accurate HMDA data are used.
Otherwise the conclusion from the analysis may be drastically different than what examiners conclude during the examination. And lastly, inaccurate HMDA data can lead to resubmission of the data which (Erin) will go over that in a few minutes.
And resubmission of data can lead to examination delays. In addition, significant HMDA data errors can result in an assessment of civil money penalties. So to sum up, the resubmission examination delays and civil money penalties are all costly to an institution.
Now I’ll go over some information on HMDA data validation testing that will be conducted by examiners and what institutions can expect in the near future.
So as shown on slide number 7 the purposes of the HMDA data validation testing can be divided into two major categories - insuring completeness and insuring accuracy.
To insure the completeness of HMDA LAR that is to be used during the exam, the examiners will focus on identifying application from a range of sources that may be missing from the LAR. So in other words, examiners will be looking for any possible omissions on the HMDA LARs.
To insure the accuracy of the data, examiners will be reviewing a sample loan file which the sample size will vary based on the total number of lines reported on the LAR. This is to determine whether the data reported on the LAR accurately reflects the information found within the loan files.
So what can your institution expect in the upcoming weeks? We turn over to slide number 8. Now I would like to emphasize a phrase “in general” before I share the information on this slide.
And the reason why I say this is because the HMDA data validation may occur prior to the scheduled on site exam or during the on site exam depending on the various factors.
Slide number 8 divides institutions into two separate categories but this does not imply that if you fall in this category then this will occur. The (planned) scheduling of the testing can still be determined on a case by case basis.
The first bullet point indicates that if your institution is considered a significant HMDA reporter which are generally considered institutions that report over 500 lines on the most recently submitted HMDA LAR or if the total assets are greater than $500 million then prior to the start of the scheduled on site exam examiners will most likely contact your institution to schedule an on site visit to conduct the HMDA data validation testing.
Again, if your institution is a significant HMDA reporter then the HMDA data validation testing will most likely take place prior to the on site examination. The purpose behind conducting data validation testing prior to the on site portion of the exam is to expedite the overall examination process.
So for instance, if an examiner finds data errors that may have an impact on the fair lending or CRA portion of the upcoming exam, conducting the data validation prior to the on site examination will allow both the examiners and the institution to promptly resolve the issues and proceed with the actual exam without significant delays.
As for those other reporters which are generally those institutions that report a fewer than 100 lines on the most recently submitted LAR. The HMDA data validation will most likely be conducted during the on site portion of the exam.
However, a visit prior to the on site examination date may still be scheduled considering various factors. Some of these factors include a scale and complexity of an institution’s mortgage lending activities and the bank’s history of complying with the HMDA data reporting requirements.
Such as the - if the institution had a history of multiple HMDA violations, resubmissions or even assessed with the civil money penalties then examiners may conduct the data validation prior to the on site examination date.
And as for the institutions that fall in between the two categories, whether validation testing will occur prior to or during the on site examination will be determined on a case by case basis.
Moving onto the next slide, slide number 9, in connection with the HMDA data validation testing the following items or information may be requested by the examiners.
When reviewing for LAR completeness the examiners may request a complete loan trial balance that includes all commercial, consumer and mortgage loans.
In addition, they may also request an explanation of codes used by the institution for specific loan categories to identify the types of collateral securing the loan. And finally, they will also request a reconciliation or pipeline reports that reflect all loan applications that are in process.
And as for validating LAR data accuracy and this is pretty much common, but as usual, the examiners will initially request for the most recently filed HMDA LAR and the year to date LAR.
Depending on the results of the initial review examiners may also request LARs from prior years dating back to the previous examination date. And when validating for completeness and data accuracy examiners will request a sample loan or application files.
And the sample requests will be random. And also the size of the sample will vary depending on the (plural) lines reported on the respective HMDA LARs. So now I will turn it over to (Erin) who will cover the next half of the presentation.
(Erin Skillman): Thank you (Eliyah). Let’s move along to the final four objectives of the call. I’m going to review some common HMDA data violations, discuss some ways in which an institution can prevent errors and maintain compliance and then briefly touch upon the resubmission process and the assessment of civil money penalties.
Let’s move along to slide 10 please. Let’s discuss a couple of common violations related to omissions. As (Eliyah) noted, examiners will ensure that all HMDA reportable loans are included on the submitted LAR.
Some common omissions that we’ve seen stem from the institution’s commercial lending area. These generally involve loans in which the purpose meets the Regulation C definition and the loan is secured by residential property. So the purpose is home purchase, re-fi or home improvements.
And these omissions typically involve loans secured by multi family or by one to four family properties. An institution must have procedures in place for collecting and maintaining accurate data regarding each loan application, origination and purchase.
And all types of dispositions are required to be reported. This includes originations and non-originations. The only optional category is those pre-approvals that were approved by the institution but then not accepted by the applicant.
Some institutions have found that side-by-side reviews of the LAR, the trial balance and call reports to look for any glaring differences, is helpful. This can be especially helpful if entire categories of loans are inadvertently omitted.
Another omission that examiners have seen is pre-approvals. For example, institutions have omitted pre-approvals because possibly a loan officer assumes that a particular transaction is just a pre-qualification when it is actually a pre-approval.
Remember that a pre-approval is when the institution makes a commitment with limited conditions such as the identification of a suitable property. Let’s move along to slide 11. The next couple of slides discuss violations related to data accuracy.
In preparation for this call, I reviewed the past 12 months of examinations. I noted the percentage of exams where HMDA violations were cited both on a national level and within the New York region.
On a national level approximately 46% of our examinations had some HMDA violations while in the New York region it actually was over majority. Over 54% of our examinations had HMDA violations. This New York statistic includes 11 of the 13 field offices in our region.
Of those HMDA violations cited 92% of them were cited under Section 203.4A. This is the section of HMDA where it references the accuracy of the data field as prescribed in Appendix A. Clearly violations related to accurate reporting are of concern for bankers and regulator alike.
Some of the violations we’ve seen relate to the rate spread field, the newest of the data requirements. I’d like to mention a couple of points here. First, the rate spread is not a reportable data field for loans not subject to Regulation Z.
For example, the rate spread field is automatically not applicable or N/A, for an investment property that is business purpose. Also the rate spread would be not applicable or N/A for non originations.
These are of course incomplete, withdrawn, denied or approved not accepted applications as well as purchase loans and unsecured home improvement loans. Secondly, the relevant date is the date on which the institution sets the rate for the final time before closing.
In many institutions this may be in the form of a rate lock agreement. A best practice that assists in identifying proper use of the FFIC calculator is to maintain documentation of the input field by way of a screen print. This will assist the institutions and their regulators with the regular data scrub.
Lastly, we’ve cited accuracy errors related to overreporting. In other words, reporting loans that are not HMDA reportable. For example, institutions should not be reporting pre-qualification, renewals and/or modifications and temporary financing.
Please refer to page 9 of the FFIEC HMDA Guide to HMDA Reporting for more information. Moving along to slide 12 we highlight some additional accuracy violations. Examiners have cited violations related to the inaccurate reporting due to a misinterpretation of the broker rule under Section 203.1.
Essentially, in a nutshell really, whoever makes the credit decision must follow the HMDA reporting requirements. Although HMDA does provide some flexibility with application dates the institution must establish a clear policy that is consistently adhered to.
Oftentimes, errors arise when there’s not a clear distinction of a HMDA application. There may be confusion relative to other regulations where there are different definitions of an application. These include ECOA, RESPA, and Truth in Lending.
Some best practices that we’ve seen include ensuring that loan officers’ signatures and dates are on the application and/or simply utilizing a date stamp to notate the application date. Documentation is really key here. Lastly, we’ve cited instances of an inaccurate action taken code.
In most cases the errors involve non-originated applications and stem from a misunderstanding of what is considered withdrawn or incomplete or approved, not accepted. Again, documentation is key here also. It’s important for a loan file to demonstrate the disposition.
Before discussing how to prevent errors let’s remember that HMDA is both a collateral and purpose driven Act.
The one optional exception is that HMDA does allow for you to report home improvement loans regardless of the collateral if you classify them as home improvement loans.
And with refinances, the purpose of the loan makes no difference. Another item, before moving along, worth mentioning, is reporting the proper action taken code for multiple purpose loans. We continually see violations with loan purpose as well.
A simile hierarchy to remember is first, home purchase; second, home improvement; and third, refinance. We’re going to move along to slide 13. You’ll see here a sample of the types of questions we asked to determine how adequate the institution’s procedures are and to identify any high risk areas.
In establishing an effective compliance management system for HMDA, the responses to these questions are considered. We’ll review the business lines at your institution including commercial, consumer, residential, mobile homes, ensuring that HMDA procedures are in place in all of these areas for identifying and reporting HMDA applicable loan types.
For each of these areas where HMDA reportable applications are handled, we’d inquire as to who was responsible for gathering the 25 - 30 data fields for each file. We would discuss how the LAR is actually populated, whether it’s an automated process with your system or if it’s manually completed.
In most instances the process is automated to a certain degree. So we discussed the level of monitoring that is completed to ensure the accuracy of the of the automated program.
We will review any data collection worksheets that are used, to insure that all the fields are appropriately captured. Another query would be how the institution evaluates audit and examination findings.
Does the institution adequately respond to findings and recommendations, correcting prior violations and identifying the root causes of violations to ensure future compliance where needed enhancements and improvements made?
Lastly, is the institution staying on top of the various types of changes that may occur that will affect the HMDA reporting function? Moving onto slide 14, I believe we’ve touched upon some of these issues on prior slides. However, the intent of this slide is to address omissions.
Institutions must have adequate tracking procedures for ensuring that non-originated files are reported. Again, to the extent your process is automated, some level of validation of this procedure must be completed to ensure it’s fully capturing all of the needed loan types.
Examiners have noted issues when an institution begins to offer a pre-approval for the first time and the software or the internal system that’s used is not updated to reflect this change and therefore, for some reason, it’s not capturing all of the reportable loans.
Some institutions’ internal systems tie directly to the HMDA reporting software. However, as part of regular monitoring the integrity of this must periodically be verified, particularly in the example I gave, if any updates are made to the internal systems or any changes in your loan offerings.
As previously mentioned, a comparison of new loan lists against the LAR, may also be an easy way to identify omissions. Slides 15 and 16 discuss some best practices regarding the prevention of accuracy violations.
The HMDA data collection procedures should be in writing and available to all applicable employees. These procedures should be detailed and provide the user with enough information to accurately complete the data fields.
Often, institutions utilize a conversation log or some sort of notepad to provide a timeline or rundown of the interaction with the applicant. This is particularly important when accurately reporting non-originated applications. Slide 16 please.
Staff must know and follow the intent of the definitions but need not cite chapter and verse. Institutions often designate an individual as their HMDA specialist and this person oversees the process and receives the detailed specialized training.
Individuals assigned the responsibility for preparing and maintaining the data must be given the resources and tools needed to produce and complete accurate data. A job aid with data fields explained would be helpful for ensuring proper collection from the file.
In addition to the completion of this job aid a second review of all loans or a sample is another method to prevent reporting errors and to maintain compliance.
Of course, similar to our data validation procedures, the extent to which you can and should review HMDA and the number of LAR entries depends upon the size of your LAR and the complexity of your lending operations. And finally, just a couple of comments relative to audits.
Institutions may consider having audits completed closer to the submission date which of course we all know is March 1st. And as a side note, and I’d like to emphasize the review must compare the LAR to the actual source document in the file.
Moving along, the next three slides, 17, 18 and 19, discuss resubmissions. Slide 17 reviews in general the factors that are considered in determining if resubmission is required. In other words, what triggers the requirement to resubmit a LAR?
Clearly the results of the HMDA validation as well as the number and type of violations noted, are part of the process. Other factors include whether or not the violations are in a key field and if the inaccuracy impacts CRA and fair lending data.
Again, generally, depending upon the severity of the errors consideration is also given to the impact of those errors and the institution’s previous history of HMDA compliance. Next slide please, slide 18, this slide reviews the requirements once it’s determined that you must resubmit.
For reporting errors, whether omissions and/or accuracy, the institution will be required to correct, review and verify the data and maintain the corrected data for public disclosure in accordance with Regulation C.
Generally, if one or both of the most recent calendar year LARs needs to be resubmitted, the institution shall do so in a timely manner and as (Eliyah) alluded to, the CRA and fair lending portions of the examination cannot proceed with inaccurate data.
Remember that in all cases of resubmission the entire LAR must be resubmitted and the public LAR must be updated. Finally, to conclude the resubmission point, please turn to slide 19. Every year the Federal Reserve Board updates the HMDA database.
I believe this is done in November of each year. They will generally accept resubmissions for up to two years after the date of the LAR. The final emphasis for resubmissions is that before an institution submits a revised file, a revised LAR, the data must first be revalidated by the examiners.
The final two slides of our prepared presentation, slides 20 and 21 present some of the components that examiners review when making recommendations for civil money penalties. All of these items here on slide 20 are taken into consideration when CMPs are recommended.
Examiners consider the percentage of accuracy violations in whether or not a resubmission was needed. For example, with regard to a late filing, examiners will want to know why the submission was late and how soon after March 1st it was sent.
The institution’s response to audit findings or internal monitoring concerns are also considered. Please turn to slide 21. Again, here are some additional items that are reviewed when considering the assessment of CMPs.
The size of the LAR and the significance of the errors and omissions compared to the total number of entries as well as the level of management oversight of the HMDA process at the institution.
And this includes ensuring that adequate procedures, training, monitoring and audits are in place as well as the bank’s response or the institution’s response and good faith of management to address any prior issues.
This concludes the prepared portion of today’s call. We may now take any questions you may have.
Coordinator: Thank you. At this time if you would like to ask a question you may press star 1. You’ll be prompted to record your first and last name. To withdraw a question please press star 2. Once again, if you have a question at this time please press star then 1. One moment for our first question.
Our first question is from (Roxanna Chapman). Your line is open.
(Roxanna Chapman): I don’t have a question.
Coordinator: Oh. Thank you. One moment. Our next question is from (Michelle Johnson).
(Michelle Johnson): Hello. Can you tell me if your data field validation error rate calculations have been changed or the competence interval sample sizes have been changed at all?
(Eliyah Shin): So I’ll try to answer that question. First in general the sample size and overall like data validation sample size, statistical samples that we pull have changed slightly.
(Michelle Johnson): Can you elaborate on that?
(Eliyah Shin): Basically we have statistical table - the sample table that we refer to. Now again, In terms of exact changes I can’t really elaborate right off the bat. I’m sorry but I can only give you the generic answer for that question.
(Michelle Johnson): Oh, okay. Are they going to be published or...
(Eliyah Shin): That will not be published.
(Michelle Johnson): Oh, okay.
(Erin Skillman): The one thing that is similar is that the random samples are - continue to be a statistically significant number based upon the actual application universe of the institution. So, you know, in general the more entries you have the larger the sample size clearly.
(Michelle Johnson): Thank you.
Coordinator: Our next question is from (Tracy Murphy).
(Tracy Murphy): Good afternoon. I just have a quick kind of question. We are considering doing a participation loan where the other loan is already on the books of another institution and we’re going to go into it for - purchase it for 50%. They’re the lead bank, it is a HMDA reportable loan.
So what I am wondering is they probably are that other bank who’s going to be the lead bank and will retain 50% of the loan. We’re just behind the scenes. The customer doesn’t even know about us.
They’re going to report as HMDA reportable and they keep the whole - they’ll have the outstanding balance of the loan and then we’ll have the 50%. Do we both report it as HMDA reportable or would that be - that would be double counting in my viewpoint.
(Erin Skillman): Correct. I believe the HMDA Getting It Right speaks to a participation loan. If you’re only acquiring a partial interest in that loan you would not report. I would refer to I believe page 58 of the HMDA Getting it Right Guide.
(Tracy Murphy): Okay. Thank you very much.
(Erin Skillman): You’re welcome.
Coordinator: Once again if you have a question please press star 1. Our next question is from (Tracy Hallesy). Our next question is from (Tracy Hallesy). Please go ahead. Your line is open. Mechanics Cooperative Bank. I’ll move onto the next...
Woman: Oh hello?
Man: Yes, hello?
Man: Hello. I’m sorry. I’m calling for (Tracy Hallesy). My question is I - my understanding is the error tolerance limits have changed and have decreased than what they used to be.
Woman: I think we lost them.
(Erin Skelman): That is correct. It’s - they have changed.
Man: Can you elaborate?
(Erin Skillman): I cannot. I’m sorry.
(Eliyah Shin): Again, even though it has changed it is still case by case basis. I mean we just don’t look at the number of errors and just automatically determine that oh, this bank has this many errors on a percentage basis so therefore we should order that bank to resubmit. No.
We do consider all of the factors before we make such a determination. So yes, like the compliance management system that is in place at an institution and other things like the history of the compliance with the regulation, the reporting requirements, those are all taken into consideration, not just the threshold.
So - the error threshold - so it may have changed but again it is on a case by case basis. And I’m sorry, as (Erin) has indicated, we can’t elaborate further in terms of the figures for the error thresholds.
Coordinator: Our next question comes from (Tim Dom). Please check your mute button. Your line is open.
(Tim Dom): Hello? Actually you just answered my question. So...
(Eliyah Shin):Oh, okay.
(Tim Dom): it’s the same question. We’re all wondering about that error threshold because we keep hearing it’s 2% and I think that’s what a lot of us are - kind of have in the back of our mind. So that’s all.
(Eliyah Shin): Sure.
Coordinator: Our next question is from (Pam Buckley).
(Pam Buckley): So I have two questions. Can you hear me?
(Eliyah Shin): Yes.
(Pam Buckley): Okay. So the first question is relative to pre-approvals. It’s my understanding that denied pre-approvals must be reported on the HMDA LAR and I was seeking confirmation that pre-approvals that are approved but not accepted are not required to be reported but are optional. Is that correct?
(Erin Skillman): Yes.
(Eliyah Shin): Yes.
(Pam Buckley): Okay, great. Thank you. The second question has to do with a situation where individuals apply for a loan so the bank collects government monitoring information. But later the loan closes in the name of a trust or an LLC, you know, not a natural person.
So the question becomes what should banks do with that collected government monitoring information? Should that be reported on the LAR or should that be omitted from the LAR based on the final closing and the manner in which title will be held?
(Erin Skillman): So it was a closed in a non-natural person?
(Pam Buckley): Correct.
(Erin Skillman): Therefore I would assume that N/As would probably be more accurate, reflective of the actual transaction that closed and I would just ensure that procedures are in place to ensure that government monitoring information isn’t improperly collected.
(Pam Buckley): Okay. So the trouble spot would be that the, you know, initially and this is where it seems that...
(Erin Skelman): It changed. Yes.
(Pam Buckley): ...we need to ask more questions around, you know, the manner in which title will be held. But there are cases where it’s individuals applying initially for the loan but then ultimately, you know, the title is held in the name of the trust or the LLC.
So I would agree with you (Erin) that the thinking is that the government monitoring information, those fields should be reported as N/A and therefore banks I guess would then just need to somehow maybe again, document the loan file, that that was the case and therefore the GMI is not reported. Is that...
(Erin Skillman): That is my thought. Yes.
(Pam Buckley): Thank you.
Coordinator: Our next question comes from (Tracy Murphy).
(Tracy Murphy): Hi. I have a specific HMDA question. If you have a loan and it’s to purchase a piece of property with a home on it, that loan closes and it was to purchase a piece of property that was HMDA reportable.
Then a little down the road the customer decided to tear down the house but we gave - we were thinking of giving a loan with tearing down the house plus some other things to do with this project. Would you report because let’s say a small portion of proceeds were going to be used to tear down a house?
It’s not obviously improving the home. It’s demolishing it. Would that second loan, because it’s going to be used on that house or that property, be HMDA reportable at all? Or no?
(Erin Skillman): I think in a general sense, without having the entire scenario, if it’s a refinancing of a dwelling secured loan another new dwelling secured replacing that. But if there is a home improvement piece in there it could potentially be home improvement.
I would need to know a little bit more about that transaction.
(Tracy Murphy): Well the second loan - the first loan was to purchase the home. Then our customers decided to demolish the home plus needed working capital and some equipment, etc. So we were going to do a second loan with one - with a small portion of it to being demolish the house.
(Erin Skillman): Could you send an email to the New York Call at FDIC dot gov and we can further look at that to make sure we give an accurate response?
(Tracy Murphy): What’s the email address again?
(Erin Skillman): It’s NYCall, it’s N-Y-C-A-L-L-S at FDIC dot gov.
(Tracy Murphy): Thank you very much.
Coordinator: Our next question is from (Renee Washington).
(Renee Washington): Hello. We have a loan application where the borrower owns the property free and clear, recapturing equity from that property to refinance to investment properties. How would that be reported?
(Eliyah Shin): So if I understand you correctly, the property - the first loan that you mentioned is originated by your institution and proceeds - part of the proceeds is also financing on new property. Is that...
(Renee Washington): No.
(Eliyah Shin): ...pretty much what...
(Renee Washington): No. There’s...
(Eliyah Shin): Can...
(Renee Washington): There is one property that is owned...
(Eliyah Shin): Right.
(Renee Washington): ...free and clear so there’s no debt.
(Eliyah Shin): Okay.
(Renee Washington): But they’re recap - they’re pulling equity out of that property and taking those funds to pay off - refinance or payoff...
(Eliyah Shin): Right.
(Renee Washington): ...two other investment properties.
(Eliyah Shin): Two other investment properties. In that case there is a lien on the other investment properties to end - pretty much it is replacing that lien. Is that correct?
(Renee Washington): Yes.
(Eliyah Shin): Then (Erin) if you could help me, I believe in terms of the property that is being reported will be those investment properties.
(Erin Skillman): Right. Multi family or one to four family?
(Eliyah Shin): Yes.
(Erin Skelman): Is it a home purchase, home refinance or a home improvement? You have to look at the purpose as well as the collateral. I would actually defer you similar to the last caller to maybe send that particular scenario in and we can take a look at it.
(Renee Washington): Oh, okay. And send it into where again? I missed it.
(Erin Skillman): It’s okay. New York calls - N-Y-C-A-L-L-S at FDIC dot gov. So they’re very particular types of scenarios that we want to make sure we address appropriately.
And just so that you know, - that all of these questions, if you’re interested in the scenarios presented or have similar cases they will - we’re planning to compile the Q&As and post them in a Word document for everyone to view.
(Renee Washington): Okay. Thank you.
(Erin Skillman): You’re welcome.
Coordinator: The next question is from (Mike Sajak).
(Mike Sajak): Hello?
(Eliyah Shin): Hello?
(Mike Sajak): Hi. I’ve got a two part question actually. It’s in reference to the FHA - FHLBB City Grant. And typically we’ve been seeing those come in as a third lien position.
And my two part question is one, do you think that we would need to report those loans, HMDA reportable, even though there’s no repayment...
(Mike Sajak): ...unless they sell or re-fi within five years and it also is secured by a note. But typically it’s not disclosed, no disclosures need to go out for (RESTA) or anything.
And my second part of the question is if that grant comes behind a first and a second mortgage if they’re - it becomes now a third lien. And is there actually a code on the LAR for a third lien position? Because I thought there was only a one or a two.
(Erin Skillman): Okay. I’m not familiar with this scenario. (Elyiah) are you familiar at all with that?
(Eliyah Shin): Well to answer the question about the lien positions you’re referring to whether or not it has, you know, it’s a third lien for the (sale) code. I believe the wording - hold on one second please.
(Erin Skillman): The wording is generally subordinate?
(Eliyah Shin): Yes, it’s subordinate. It doesn’t say that the second lien but it’s the subordinate like the code number one for the lien status will be the first lien. But anything after that is subordinate lien.
(Erin Skillman): Right.
(Eliyah Shin): So yes, code two secured by subordinate lien. So yes, if it’s a third lien and if it’s a reportable yes, you could still report under LAR.
(Mike Sajak): Okay. So I could report those? You would recommend reporting the FHLDP city grants?
(Eliyah Shin): Now in terms of the other part of your question I’m afraid that I would have to ask you to kind of submit that question in writing because it is kind of a peculiar case that either (Erin) and I have dealt with. So if you could send that we’ll try to research it and get back to you on that one.
(Mike Sajak): Okay. I appreciate that.
Coordinator: Once again, if you have a question please press star 1.
(John Vogel): Maybe while we’re waiting for another telephone question we could read some of the email questions we’ve received?
(Erin Skillman): Well one of them that came was asking about the minimum error rate. We’ve addressed that. Let me see here. Another question came in. My question is a HELOC. As a choice, the financial institution can decide whether to report or not report. That is correct.
Does that same choice apply to business lines of credit? I guess I would ask for more information. Are they talking about HMDA? Are we talking about the small business LAR? So I would maybe request additional information be provided on that to make sure that I understand the question.
(Eliyah), do you see that one there?
(Eliyah Shin): Right. I mean...
(Erin Skillman): Yes.
(Eliyah Shin): ...it’s just a business line of credit. I think this question more is referring - in reference to the small business LAR, not the HMDA LAR.
(Erin Skillman): Right. And then we got a question yesterday. Let me go back to that one. It was regarding a loan - here we go - in which the original loan amount was let’s say $30,000 was the actual amount given. And it was approved. The loan process drags out because of the seller.
So the borrower requests that the commitment is extended and that the loan amount is reduced to $18,000 because of the price of the home was reduced. The loan is then reapproved at $18,000. Then the commitment expires. The application is closed approved not accepted.
The question was do you report the $30,000 that was initially applied for or the $18,000 on the LAR. Again, (Eliyah Shin) and I talked about this one. I actually have submitted this one down to Washington because it’s kind of quirky.
And my initial instinct was the $18,000 because it’s not, you know, HMDA getting it right speaks to a counteroffer going back to the initial amount.
But where this was still kind of a loan in process my thought was going to the $18,000 as the actual, you know, final amount that the borrower and the bank kind of agreed upon and the commitment was made for.
But to that individual banker and to everyone who’s going to look at the Q&A, we’ll address that as well once we get a final answer. (Eliyah Shin), I didn’t know if you wanted to address the comments that you had.
(Eliyah Shin): No. I think that pretty much sums it up for now with regards to that question.
Jim Keller: Hi. This is Jim Keller. I’m here in the New York regional office as part of the call. And we’re manning the email box and we see that there was a follow up to that previous question in regards to business lines of credit. If it’s a line of credit that is a HMDA reportable loan then that also would be optional.
Just as long - again, we’re talking about a HMDA reportable loan that meets the definition of a HMDA reportable loan. But if it’s a line that would also be the bank’s choice of whether or not to report.
Coordinator: We have another question from the phone. (Debra Lazat), your line is open.
(Debra Lazat): Hi. We just had an audit by the FDIC. They came and did the initial HMDA review. They’re coming back in September. And the auditors did make mention of upcoming HMDA changes. Are there going to be any changes to HMDA for 2011?
(Eliyah Shin): That’s - I think - I’m not too certain but I think the examiner may be referring to the effect of the Dodd-Frank Act and also what impact it will have in terms of HMDA reporting requirements. But I believe that determination has not been made with regards to reporting additional fields.
(Erin) do you have any information on that?
(Erin Skillman): That would be my guess as well as to what they were referencing. Obviously the HMDA hearings are up and running.
I haven’t been following them that closely but I know there has been plenty of discussion because there is the potential of additional data fields to capture additional items for underwriting. So we just don’t know when that will happen.
That’s my personal opinion but we don’t know for sure if and when. And of course, you know, from our perspective that our procedures, you know, will follow whatever changes are made.
(Debra Lazat): Okay. Thank you.
Coordinator: And we have a question from Ms. (Murphy).
(Tracy Murphy): I have a question about the last caller - not the last caller but a few callers before on the trust, the LLC, how to report that. And someone made a comment that you would capture the data even after, towards the closing part of that loan.
I thought HMDA data was supposed to be captured at application.
(Eliyah Shin): Are you referring to the government monitoring information that needs to be captured?
(Tracy Murphy): Yes.
(Eliyah Shin): Yes. Technically the way it - the regulation is written like in terms of the collecting government monitoring information it is yes, as you have indicated with doing the application space.
So you based upon that. But the previous scenario was that at the end when the loan closed the borrower has changed from a natural person to a trust.
So again, the government monitoring information again, you - if it were to be reported on the LAR it would be based on what was collected during the application stage.
(Tracy Murphy): So we would not go and change it to an LLC or N/A. We would keep the information that was captured at application if it was two borrowers even though they changed to hold that property in the name of a trust or an LLC.
At the time of closing we would keep the application data for the GMI that we took it application and not change it?
(Erin Skillman): Right. I think the question was in those scenarios what do you report then? If they originally applied as two individual persons but then at a later point in time - if you weren’t able to establish at application that would be the best practice, to find out how it’s going to close.
If not, you want to probably more accurately reflect the transaction taking place. So if the borrower or applicant is ultimately not a natural person we would say that you probably would want to be using the codes for not applicable.
(Tracy Murphy): So any time during the application if something changes, if we collected as a two person borrower and for some reason it changes to a trust or a third - well a third person probably wouldn’t apply. But if it changes to a trust then we need to change the information?
(Erin Skillman): Maybe. It’s case by case.
(Tracy Murphy): Well - okay, I don’t understand.
Woman: It has to be one or the other.
(Tracy Murphy): It has to be one or the other. We have to have some direction on that because this is a common occurrence.
(Erin Skillman): So I would recommend that be institution implement procedures at application to determine how the loan is going to close.
(Tracy Murphy): No. Okay. Maybe you’re misunderstanding. Mr. or Mrs. Smith come in and they say we’re going to borrow this - we’re going to buy this piece of property under the name of Mr. and Mrs. Smith. We go through. We take our application, etc. We process it that way.
Many a time through attorneys or whatever, they’re told to put a name a trust and decide to put a name a trust. They come back in and say okay, by the way, the property is going to be held in the name of ABC Realty Trust.
(Erin Skillman): Okay. I’m thinking we want to run this through the Q&A process and do the research. I don’t want to give any inaccurate information. So if you wouldn’t mind including that in your email please?
(Tracy Murphy): That would be great.
(Erin Skillman): And we can get a formal answer for you on that.
(Tracy Murphy): Thank you.
(Erin Skillman): You’re welcome.
Coordinator: Once again if you have a question please press star 1.
(Erin Skillman): During our call we received one email question. I’m just going to read it (out loud). We have a commercial loan with two properties, one a commercial office closing and a one to four family home. They are both going to be financed at the time of the loan.
Would this be reportable or not? The commercial building has the most value and the one to four was taken as an (unintelligible) precaution. We would address this on our Q&A because I think we need to do some more research on this particular question at this time.
Coordinator: And at this time we do not have any questions.
John Vogel: I don’t think there’s any email questions left either.
Jim Keller: Hi. This is Jim Keller again in the regional office. I can understand if you’re frustrated if you’re not getting specific answers to specific questions.
But I think you’ll appreciate that for us to give a quick answer in this type of forum and for people out there to take that as definitive it just doesn’t make sense.
So while we’re pretty sure we know these answers we really do want to handle it in the Q&A format so that we’re able to fully research it and the information that we provide is entirely accurate. And that way there’s no ambiguity or interpretation issues.
So I think, you know, we do want to apologize that we weren’t able to give more specific answers. But the reason why we’re not able to do that is just because of the technical types of questions that we’re getting. And obviously HMDA is a pretty technical regulation.
(Erin Skillman): And Jim I would actually add to that the question that (Eliyah) just read that came through the email, I would even ask the individual who submitted that to provide more information on that particular loan such as the purpose so that we can have all of the information to conduct the research.
Jim Keller: Right. Exactly. And some of these questions that are coming in may require some back and forth to get the specifics again to make sure that we’re giving, you know, entirely accurate answers. We don’t have any additional questions in the email box either.
Man: Any questions on the phone (Marla)?
Coordinator: No questions on the phone sir.
Man: Okay. I guess that will just about do it.
John Vogel: All right, well thank you very much. Again, this is John Vogel in New York and I appreciate your participation on today’s call. Thank you.
Coordinator:Thank you for participating in today’s conference. You may disconnect at this time.
|Last Updated email@example.com|