RATA Associates Quarterly Newsletter
RATA Associates

Comply 365
Edition #1


2018 CFPB HMDA Changes

2018 CFPB HMDA Changes

RATA Associates has been through two major changes to the HMDA regulations, first in 1989 and then again in 2004. We are currently working on the CFPB's 2018 revision to HMDA, which includes the most substantial changes to date. Many institutions are nervous that vendors will not be able to implement the changes correctly or in a reasonable time frame. This article is designed to assuage your concerns as it pertains to RATA and the Comply software. RATA has always implemented changes to regulations faster than any other software vendor, and our clients have never experienced issues with the transitions.

Since the proposed changes were announced, we knew this was going to be a particularly arduous task for the institutions more than the vendors because most institutions have multiple data sources they need to be concerned about. The repurposing of some existing fields with added enumerated values and just the sheer number of new fields will make these changes challenging to say the least. We have found that a multi-phase deployment will serve all RATA clients the best and allow for ample testing of your processes along the way.

Phase I for RATA is to add all of the new fields to the system. This will affect both the data entry screens and the data import functionality. Because the 2016 data collection and submission are 100% unaffected by the changes, we will be branching our code path this year around September or October. The Comply 16 branch will contain all of the new analysis features described in some of the other Newsletter articles. Sometime in January, we will be releasing a Comply 16.1 version that will contain the 2018 field changes. Because clients will be in the midst of the 2016 submissions, we DO NOT recommend that you update your LIVE system to 16.1. Institutions that want to get a jump on testing the 2018 changes should do so in a test environment to minimize confusion at that key time. Version 16.1 will actually be able to submit the 2016 data, but again we don't recommend it. If minor changes are needed for the Comply 16 version, they will be made in the 16.0x code path.

We have already begun the design phase of our changes but will not start the actual coding of the 16.1 features until the 16 version is coded, tested, and released to reduce duplicate development efforts. Below are some of the sample screens we have designed for 16.1. The first four tabs are the submittable data fields as compared to one tab in the current Comply. That is a lot of new fields!

Once you install Comply 16.1 to your test environments, you will be able to test your data imports to make sure all of your new data is importing properly. At the same time, you can be creating your new import definitions so that you are not doing this at the last minute.

Phase II of the development will be the new 2018 edit checks, which again, judging by the number of fields, will be substantial in number and complexity. The edits and report specifications have not been released as of yet.

The release of the actual submission functionality will be broken down into three releases. For 2016 the submissions remain exactly the same as they have been, fixed length files. For 2017, even though the data itself will remain unchanged, the submissions will be required to be in the new pipe-delimited format. (The same file type that will be used for 2018 but with the fields and values from 2016.) Then for 2018 the new fields and values will be the pipe-delimited file type. Due to our agile software development practices, these submission changes will not been difficult for RATA. However, the rest of the industry may have problems getting the incremental submission format changes correct over the next three years. Rest assured, we have you covered. We will keep you up-to-date on our progress through our newsletter and with Start Page articles in Comply.

New Pre-Defined Comply Reports

New Pre-Defined Comply Reports

When Comply 16 is released in the fall of this year it will incorporate many new pre-defined reports to aid in the analysis of your HMDA, CRA and Locations data.

We have noticed over the past year that examiners are doing many new performance tests for CRA examinations. We have reviewed the many tables created by the FDIC, FRB and OCC regulators during examinations and combined similar aspects to create ten new performance evaluation reports. The reports will include the following:

  • Loan portfolio summaries
  • Market profiles
  • Branch and ATM distribution
  • Loan-to-deposit ratio analysis
  • Small business, small farm, and HMDA loans by census tract characteristics

To meet specific client requirements, we are also adding reports that will allow you to show census tracts in the proximity of your location data, such as branches and ATMs. These reports will show census tract income and minority concentrations within user-specified distances of any branch or ATM within the Locations component of Comply. These reports in concert with the ten new Examiner Audit Tables will allow you to get a handle on your lending patterns in preparation for upcoming examinations by your regulatory agency.

If you need other numbers that our "pre-defined" reports can't give you, have no fear. With Comply 16 you will also be able to purchase a new analysis component that will give you ultimate flexibility to design custom reports that can contain HMDA, CRA, Peer, Demographics, Deposits and more.

New Custom Analysis Reporting

New Custom Analysis Reporting

With exam preparation becoming more and more complex we have seen clients trying to break data down in new ways. Over the last year we have had many requests for special projects to create customized analysis reports.

We have brainstormed internally and come up with a new analysis component, which significantly expands the custom reporting features in Comply. The new analysis component will integrate all of the different analysis elements (HMDA, CRA, Peer HMDA, Peer CRA, Deposits, Locations and Demographics) and allow users to create custom pivot tables in minutes.

Our development team is currently working on creating a compressed peer file that clients will be able to load directly into their local Comply databases. By using this data as well as Microsoft SQL Server Analysis Services (SSAS), Comply users will be able to create customized interactive reports using Excel or other compatible OLAP client tools. Once Excel is connected to SSAS, a user can simply drag and drop Comply fields into their report and it will be generated automatically in real time.

Here are some examples of custom reports that were designed and created in just a couple of minutes using the new tool.

This will be an additional component, available for purchase by customers, who want the added reporting power. We plan to have the component ready along with the release of Comply 16 by mid-September in time for the release of the 2015 HMDA and CRA peer data. We will update you on the progress and show you more sample reports in the next newsletter. We will also have some videos available to illustrate how the analysis component works and to illustrate how easy it will be to design your new custom reports.

Fair Lending Automation

Fair Lending Automation

When we talk about Comply, we use the word "definition" quite a bit. This word can have a few meanings depending on where we are using it, but basically it means a template or a defined set of steps or operations. The more you take advantage of definitions the more powerful and efficient Comply becomes. Import definitions help speed up getting data into the system, export definitions let you get custom sets of data out of Comply, report definitions allow you to quickly create reports, list view definitions let you focus on the fields that are important to you, etc.

In Comply Fair Lending, Analysis Automation definitions allow users to automate repetitive or periodic tasks, perform data cleanup and partitioning, and automatically create new analysis elements. All of these processes can be designed by users and saved as definitions.

An Analysis Automation definition is a user-defined series of steps that can be executed on demand using the Analysis Automation wizard or automatically when a new workspace is created. Comply ships with a few sample definitions in order to help users understand how the tasks are designed and executed. To create, modify, or examine Analysis Automation definitions, open the Definition Manager from the Home ribbon.

Each step in the definition has a data source, a set of conditions, and a set of actions. While the data source for a step is typically all of the applications included in the current workspace, the user can also specify a filter, an application group, or a regression model as a data source.

When a step is executed, the conditions are evaluated. The user can specify whether any or all conditions must be met in order to execute the set of actions. For example, a set of conditions that will find missing or invalid credit scores might look like this:

If the step's Match Type is set to "Any", the step's set of actions will execute when any of the conditions evaluate to true. When the Match Type is set to "All", the actions will only execute when all of the conditions evaluate to true.

The step's actions will determine what operations get performed automatically by Comply. Some of the available operations include excluding applications from the workspace, creating application groups, setting field values, and processing regression models. In the example conditions above, we are looking for invalid credit scores based on custom criteria. If we want to add the applications with invalid credit scores to an application group, our sole action will be:

We can then save this step in an Analysis Automation definition called "Check Credit Scores". Each definition can include as many steps as needed. In order to execute the automation process using the new definition, launch the Analysis Automation Wizard and select the "Check Credit Scores" definition. After the wizard finishes the process, a report is created with a log of all the steps executed, including the number of applications or elements affected by each step.

In order to learn more about the types of conditions and actions that can be used, review the Comply documentation and explore the sample definitions that ship with the software. You will probably find that there are a number of ways that Analysis Automation can be used to both standardize your processes and save a lot of time and effort.

BISG Proxy

BISG Proxy – Just like the CFPB

On October 6, 2014 Comply became the first HMDA/CRA product to implement and emulate the CFPB's Bayesian Improved Surname Geocoding (BISG) proxy methodology with the release of version 14.0.

The CFPB's BISG proxy methodology combines the use of the surname of the applicant/co-applicant and the demographics of the property location to apply statistical probabilities to the race/ethnicity calculations. The ability to use area demographics improves the ability to accurately derive the race from the surname based upon the racial dispersion of the area in question.

Depending on the precision of the geocode assigned to the property address, Comply will use the demographics from the Census Block Group, Census Tract or ZIP Code. The higher the geocode precision, the smaller the demographic area and thus a more specific demographic. The method then uses the racial composition to add weight to the race derived from the surname alone to more accurately reflect the predicted race. Because of the extremely high precision levels that RATA's GeoPlus service returns, a very high percentage of the calculations use the Census Block Group demographics which is the most accurate.

Let us look at an example of how the combination of demographics can impact the calculated race using the surname of BUCKNER as an example. In and of itself, the surname BUCKNER is primarily assigned to white persons on the order of 69% and to black persons 26% of the time. Under these circumstances, one has a fairly good chance of the person being white, based on the surname alone. However, let's factor in some demographics from the St. Louis area. Using census block group demographics where the population of black persons is 79% influences the resulting calculation and pushes the surname to be considered a black person by virtue of the higher black population of the area.

The use of BISG proxy calculations is very useful in studying CRA lending to consumers where collection of race data is prohibited and is really the only way to offer insight into fair lending practices for these types of loans. The CFPB regularly uses this methodology when reviewing institutions for potential discriminatory practices and we highly recommend that you utilize this functionality in self-assessment exercises to know how your lending practices may reflect in future reviews by your supervisory agency.

ZOOM Updates

ZOOM Updates

If you're not already using it, the ZOOM Geocoder is a web-based tool that is designed for geocoding of hard-to-find addresses and also auditing of other processes to make sure they are producing correct results (click here to learn more). We're constantly reevaluating and improving our products and ZOOM is no exception. With the most recent update, we've added even more useful features and enhanced existing ones which include the following:

Substantial Minority Layer

In addition to the existing Census Tracts and Income Levels layers, you can now also display the Substantial Minority layer on the map. This will provide shading over the specified census tracts to designate those areas while still allowing you to see the income level colors at the same time.

 

 

Zip Codes Layer

Much like the Census Tracts, you now have the ability to display the Zip Codes layer over the map. This can be used in conjunction with the other layers as well.

 

Toggle Layer Labels

In some instances, especially when zooming out of an area, labels can get lumped together and make it difficult to see the areas you're looking for. To help with this, we've provided options under the Census Tracts and Zip Codes layers to turn the labels on and off.
cid:image007.png@01D188CF.2F41BC20 cid:image010.png@01D188CF.2F41BC20

Printing

Sometimes there's a need to print the maps to get a hard copy and is why we also added a Print button to the toolbar. This will resize the map to the appropriate size for printing onto a standard sheet of paper and to allow you to adjust the map, markers and layers to the exact view you'd like.

 

Full Screen

Although this was a pre-existing feature, it wasn't overly obvious that it was there. We've made the button and description a bit clearer now which allows you to enlarge the entire map to get a full screen view without the clutter of any other information.

 

 


Improved Performance

We've optimized multiple areas of the system to increase performance and reduce loading times. Despite the amount of data we're collecting and processing from multiple sources, we've managed to reduce loading times by as much as 8 seconds in certain instances.

We hope that these additions and improvements prove beneficial to you and your organization. If you haven't already, you can try ZOOM free of charge by going to: www.zoomgeocoder.com/trial/. As always, if you have any questions or need assistance using ZOOM, please get in touch with one of our support representatives and we'll be happy to help.

What's New with Mapping

What's New with Mapping

Over the years RATA Associates has been improving on how maps are created in Comply to allow our users to take full advantage of their data. The progress continues as we have implemented more mapping options.

New Geographic Area Layer Mapping Feature

In previous versions of mapping, users could only see the tract characteristics of the census tracts within their assessment areas. That was great for seeing what your area contained, but it was not good for seeing what the demographics were for the areas outside. In version 15, we added a new layer called Geographic Area, which expanded the mapping capabilities extensively. With this new layer users can show multiple assessment areas, see the demographics of both in and out of the area, select tracts on the fly to add or remove from assessment areas (then save), shade or hatch differently in or out and much more. We recommend you play around with this new layer to see what you can do! In the Map Properties option, you will see separate layers for Census Tracts and Geographic Areas.

Shading and Hatching

Many users have asked if there were a way to cross hatch tracts in their assessment area by a particular field (e.g. Minority Concentration), while at the same time being able to shade the tracts to show another data field (e.g. Tract Income). With the new cross hatching option, you can now do this. This feature can be used in both the Geographic Area layer and the Census Tracts layer to give you both effects. Users can now create a more in-depth map with flexibility and ease.

Borders can also be added to Geographic areas to isolate your selected area from the rest of the map.

System Map Definitions

Making maps can get a bit time consuming, so to help out those who don't have the time or patience to select colors and shading we included in our latest pre-defined mapping definitions. These definitions will help eliminate some of the guesswork when creating a new mapping definition and give you a good starting point for your custom definitions. Like all of our definitions you can modify and save them to suit your needs.

Tips

Tips

This section of the newsletter will list helpful tips for functionality contained in the Comply software.

Reporting Entity Properties (F6)

The Reporting Entity Properties contains counts and links to different subsets of data such as applications with edits, rate spreads over the threshold and non-geocodable addresses. You can right-click on a reporting entity in the Data Manager and select Properties or simply push F6 on your keyboard. On the Status tab you will see the counts and hyperlinks. Clicking on a hyperlink will filter you to that subset of data.

Copy Reporting Entity Forward

Instead of creating a new reporting entity from scratch in the new reporting year you can copy the previous reporting entity forward. This will retain settings such as transmittal information, options and geocode ID. Highlight the reporting entity under the current year then click the Copy button on the Data Manager ribbon and select Copy Forward to Next Year. If that option is grayed out, you will need an administrator to perform this action.

Humor

Comic

Sales & Support
(407) 831-RATA (7282)
support@rataassociates.com