Printer Friendly

Debunking some production myths.

A new study of costs, profitability and productivity in the production side of the business has produced some eye-opening findings. One such finding shows that even with commissions factored in, the personnel cost per full-time employee in retail is a mere 6 percent higher than personnel costs for wholesale.

All industries have some commonly accepted principle around which most executives make crucial decisions for their companies. These principles may hold true in some cases and at some points in time. They also may be considered common sense. But in many cases, they need to be reevaluated or tested on a regular basis.

Frequently, changes in the economic or legislative environment, advancements in technology, new entrants into the industry, and/or mergers and acquisitions involving major competitors may affect the industry in such a way that these principles no longer hold true. Many of these principles - some can be better labeled as myths - exist in mortgage originations, and some reflect the current reality of the mortgage production industry.

In KPMG Peat Marwick's Mortgage Origination Performance Study, the commonly prevailing principles of mortgage production were scrutinized to determine their current validity. The study data, in combination with recent industry knowledge arising from various diagnostic analyses, reengineering projects, due diligence reviews and conversations and meetings with mortgage banking executives across the country, form the basis of the conclusions outlined in this article regarding the realities of mortgage originations.

The study

This year's Mortgage Origination Performance Study covered a variety of mortgage operations, diverse in geography, production channels, ownership, product mix and management philosophy. On average, these participants originated approximately $1.3 billion in 1991, with a mix of 60 percent retail and 40 percent nonretail. The average loan balance was approximately $130,000, with the greatest concentration of product coming from California, Florida, New York and New Jersey. For about 50 percent of these organizations, the parent is a bank, thrift or other financial institution. The remaining 50 percent is a mix of independent mortgage companies and subsidiaries of nonfinancial institutions. The participants have similar characteristics to the national group of mortgage companies in our data base.

Cost, productivity and profitability statistics were collected, validated and analyzed for each participant and in total, in order to ensure that the study looked at the complete picture. To further ensure comparability, all statistics were fully loaded, (i.e., inclusive of all functions and full-time equivalents [FTE] that support origination operations). All data were reviewed separately for retail versus nonretail channels. For our purposes, non-retail was defined as any loan where the institution did not have direct contact with the borrower. Cost data were broken down by each functional area - origination, processing, underwriting, closing, warehousing, postclosing, shipping and delivery, and secondary marketing - to the extent possible, as well as by major cost areas - personnel (including all salaries, overtime, benefits, commissions, bonuses and temporary staff), other direct costs (including outside services, postage, travel, occupancy, promotions and advertising), indirect costs (including data processing, depreciation, accounting/finance and overhead) and interest expense.

Personnel and productivity data were analyzed in total and by functional area with the relationship between cost/compensation and productivity in mind. The revenue and profitability data focused on the source of revenue, while considering the impact of servicing-release fees, favorable funding arrangements and management strategy.

Retail versus nonretail:

which is less costly?

A key principle of mortgage production revolves around retail versus nonretail channels of production. Most executives believe - for some valid reasons - that wholesale is less costly than retail production on a per-loan basis. The main reason for this additional cost is thought to be the commission that must be paid to the loan officer. Further, it is generally accepted that owing to this additional cost on the retail side, profitability is greater for wholesale producers than their retail counterparts. For quite a few years, in fact, many production operations considered getting out of the retail business in order to concentrate on their wholesale and/or correspondent operations, hoping to lower costs and raise profitability.

Our study confirmed the first part of this theory - that nonretail is indeed less costly than retail - approximately one-third less costly on a per-loan, total-cost basis. The fully loaded total retail cost of production was approximately $3,000 per loan, while the nonretail totaled approximately $2,000 per loan.

As expected, the difference was primarily in the personnel area, with the combined differences in other direct, indirect and interest accounting for only 20 percent of the total variance. The personnel costs, which include salaries and commissions, totaled approximately $1,700 for retail, and approximately $800 for nonretail. On a direct cost basis, the cost per retail loan was $2,300, compared with $1,200 for nonretail.

Further, the retail versus nonretail variance in the cost of the origination function - defined as those functions and activities associated with a loan officer, the sales process, the application process and pre-qualification - was significant. It was almost three times more per loan for retail than nonretail. However, the variation between personnel cost per FTE for retail versus nonretail was minimal, with retail being only 6 percent higher. The retail personnel cost per FTE averaged $55,000, while the nonretail averaged $52,000. This was surprising considering the frequently heard commentary regarding the burden of loan officer commissions on retail cost.

As a continuation of the controversy regarding commissions, we examined the different compensation practices among retail originators. We tried to answer questions such as whether retail originators that are commission based pay more personnel costs per FTE, and whether those FTEs were generally more productive. The theory is that an organization that compensates its loan officers more on commission and less on base salary will pay more per FTE, but those same FTEs will be motivated to produce more than their higher base salary counterparts. Our research revealed that organizations that paid proportionately more variable compensation dollars (i.e., less fixed salary and more commissions) did not pay significantly more per FTE than their peers. Variable compensation dollars, as a percentage of total personnel cost, ranged from a low of about 16 percent to a high of 55 percent.

The link between commission structure and productivity, however, was evident in the origination function - those that paid higher commission dollars generally experienced at or above-average productivity in this area. The average annual production per origination FTE was approximately $9 million, or $750,000 per month, but ranged among the participant firms from approximately $4 million to $20 million annually per origination FTE.

Relative to the profitability of retail versus nonretail operations, we examined the profit margins of both, excluding interest income and interest expense. For the retail originations, we found a near break-even situation. Organizations that sold loans "servicing released" generally had positive profit margins, as expected. Analyzing the nonretail channels, the participants in the study experienced a net loss of between 15 and 65 basis points. Although expenses were definitely higher for the retail operations, the additional revenues more than made up for the additional expenses, netting an average benefit of approximately 35 basis points. This additional revenue was largely due to higher origination and settlement fee income, including fees collected for credit reports, appraisals, surveys, assumptions, inspections and document preparation. These findings are contrary to the commonly accepted principle that retail operations are generally less profitable than nonretail operations.

The realities surrounding

high-balance and special


Other theories that we examined related to the origination of high-balance and jumbo loans. Generally, high-balance loans are thought to be higher cost. Based on the retail and nonretail operations studied, this was found to be true on a per-loan basis, although not necessarily true when cost was calculated in basis points. In our experience, however, there is no national average or trend on this issue.

The next question examined productivity. Is productivity highest for originators of high-balance loans? Our analyses revealed that the inverse relationship held true (i.e., originators with below-average loan balances had the higher productivity, based on loans per FTE). For retail production, overall productivity averaged 32 loans per FTE per year, while originators with below average loan balances generally had 34 to 35 loans per FTE per year. To some extent, this may be due to the fact that larger, nonconforming loans require more documentation and assurance in processing and underwriting. Further, a large proportion of jumbo loans are made to self-employed, high-equity individuals and generally need to be underwritten more carefully and put through additional levels of approval. Based on these findings, we hypothesized that the cost and productivity particular to the underwriting area would be negatively affected by high-balance and special loans. As expected, those originators with higher average loan balances, and thus a larger proportion of jumbo loans, experienced lower productivity and higher cost per loan in the underwriting function. In total, productivity in the underwriting function averaged approximately 540 loans per underwriting FTE per year, while those with higher average loan balances generally registered between 250 and 300 loans per underwriting FTE per year.

An extension of this same issue begs the question, "Is productivity lowest for originators of adjustable-rate mortgages (ARM) and other special loans, as well as jumbo loans?" As expected, participants with the highest percentages of convertible and nonconvertible ARMs had average or below-average productivity, based on loans per FTE. This is primarily due to the additional documentation required when originating an ARM loan. This finding is further supported by our national experience, for both retail and nonretail originators.

When looking at high-balance loans, the expectation exists that there is a relationship between average loan size and fee and/or total revenue. If fees and other revenue were based on a percentage of the principal balance, this would appear true. However, for retail originators, this hypothesis was not consistently confirmed. On the other hand, there was a clear relationship for nonretail originators - those with higher average loan balances did report higher than average fee income per loan.

Processing and underwriting:

the keys to total


Much of the discussion and analysis regarding mortgage originations is centered on the processing and underwriting functions; it is generally believed that with proper control in these two crucial areas, the entire operation would be on track. Certainly, this has some common-sense appeal, as poor processing and underwriting will certainly cause problems further along in the process. These problems can surface in the closing and postclosing areas, as documentation is found to be missing or inadequate. Faulty processing and underwriting can certainly cause difficulties in the secondary marketing area if loans are found to be unsellable or ultimately must be repurchased. So, shouldn't controlling costs and enhancing productivity in processing and underwriting be the key to overall low costs and high productivity? However, evaluation of the data in these areas yielded mixed results. For retail originations, those organizations that exhibited lower than average costs in the processing and underwriting functions consistently exhibited overall lower costs; however, the expected relationship did not hold true for productivity (i.e., those retail organizations with high productivity in processing and underwriting did not consistently experience high overall productivity). Only with nonretail originators was the expected relationship found - above-average performance in processing and underwriting was directly and positively related to overall above-average performance for both cost and productivity.

While discussing the areas of processing and underwriting, the question of staffing ratios usually surfaces. What is the ideal ratio of loan officers per processor and loan officers per underwriter for balancing high quality and productivity with low costs? Generally, retail participants with more loan officers per processor had lower direct costs. Those organizations with centralized processing and underwriting had significantly higher ratios of loan officers per processor and loan officers per underwriter, as well as higher than average productivity in these areas.

Although many factors should be considered in making these staffing-level decisions, it appears that two to two and one-half loan officers per processor and six to nine loan officers per underwriter are the average industrywide ratios. The centralization or decentralization of these operations, along with product type, closure rate, origination productivity and staff experience are among the factors that will influence these ratios. For example, an organization with centralized processing and underwriting had a ratio of almost eight loan officers per processor and more than fifteen loan officers per underwriter; however, this same organization had an annual production per loan officer figure that was well bellow Another example is an organization experiencing higher than average fallout rates - such as 60 percent versus the industry norm of 40 percent to 45 percent. For this organization, it was not surprising that the ratio of loan officers per processor was considerably less than average.

Economies of scale: do they


A widespread theory in the mortgage production industry is that economies of scale do exist, and although it's frequently debated, it is generally accepted that bigger is better. We examined this theory in regard to the effect of levels of production on productivity, cost and profitability. Based on this year's study, no clear relationship was found between levels of production and cost for retail originations. Productivity generally followed direct cost order; high productivity equaled low cost, but was not necessarily linked to higher volumes. Generally ally, those organizations that were higher than average in both production volume and cost were independent mortgage companies. This supports the theory that independent organizations have higher fixed costs, as they must perform many functions (data processing, accounting/finance and so forth) without the support of a parent organization. However, retail originators with high volumes can usually be more pro due to the fact that certain volume levels are necessary to support a retail structure. In most cases, nonretail participants with higher than average volumes generally exhibited lower than aver costs. When examining the link between volume levels and profitability, no conclusive relationship existed for either retail or nonretail originators.

The volume of production in 1991 far exceeded volumes in prior years, primarily owing to refinancings caused the drop in interest rates. This drastic increase in volume may have camouflaged evidence supporting the existence of economies of scale, as production managers scrambled to react to surge in volume. In some ways, this rapid increase in volume caused inefficiencies - heavy reliance on expensive overtime and inexperienced temporary staff to compensate for the increase workload. In addition, commission structures generally were not adjusted for volume increases; more costs accumulated as loan officers and account executives hit new highs in volume.

This wave of refinancings continue with force through the first six month of 1992; 1992 volumes for this period were 35 percent to 40 percent higher than for the first six months of 1991. Assuming that production will return to more normal levels at some point in future months, production managers be able to manage their growth more efficiently and realize some economies of scale. With this eventual drop in refinancings, they will face new challenges regarding staffing levels, productivity standards and quotas, product differentiation and other issues.

But regardless of where production volumes may be going, the tool of benchmarking - knowing how you measure up to your competition - will greatly add to any production operation's ability to compete successfully by providing comparable statistics, allowing an organization to identify its strengths and weaknesses and identifying potential areas for improvement and performance enhancement.
COPYRIGHT 1992 Mortgage Bankers Association of America
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1992 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:mortgage production industry
Author:Oliver, Geoffrey A.; Reed, Regina J.
Publication:Mortgage Banking
Date:Dec 1, 1992
Previous Article:Another December surprise.
Next Article:Steal this idea.

Related Articles
Myths revisited.
Netting wholesale opportunities.
Selling protection.
Demystifying the business.
Myths and misconceptions.
Eyeing economic reality.
These Honored Dead: How the Story of Gettysburg Shaped American Memory.
Debunking eMortgage myths: the full adoption of eMortgages is inevitable. Current myths about the obstacles to that adoption won't make the outcome...
Debunking 9/11 debunking; an answer to Popular mechanics and other defenders of the official conspiracy theory.

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters