How to Survive the End of the Refi Boom

by John D. Beneventi 16. December 2020 04:21

How to Survive the End of the Refi Boom

Tips for Aligning your Business to the Changing Lending Environment

We have all seen the signs that the refi boom is coming to an end. On November 16th, 2020[i], the Mortgage Bankers Association predicts that 2021 loan originations would shift with purchase demand increasing by 11% and refinance decreasing by 49%. If this is true, there may be a net change in demand for mortgage of 39%. Let’s assume that the MBA will get this correct within a margin of +(-) 5%. We could use this 32-44% range as a starting point for analyzing what may happen to your company. Here are some tips when contemplating the potential impact.


Objective: create a simple analysis of the changes your company will experience given the predicted change in demand for rate and term refinance loans and run a model to see the resulting net profit(loss).


Steps and Resulting Action Plan

  1. Determine what a decrease in refi demand will mean for 2021 production
  2. Simulate changes to determine how this reduction in demand might impact originations using a range of low and high (volume and units)
  3. Establish an accurate break-even
  4. Examine Vulnerabilities
    1. Review refi/purchase distribution for company, branch, and loan officer to find vulnerabilities
    2. Evaluate production capacity and determine if any department will be over-staffed.
  5. Create an action plan with triggers (based on loan vol/units) to maintain profitability.

In the following example, we will walk through each step using dummy data, ACME Mortgage, a sample we have created in Telemetry BI . Last we will create an action plan based on what we have learned during the review.

Note that ACME has experienced a large gap between purchase and refinance. This is the area of risk as the refi will come down to normal levels after the rate and term portions end.

Next we want to get more information comparing the number of purchase / refi units, year over year. Note that we had another refi bubble in 2019 which started the fall and ended sometime in Q4 which could overstate the prior year as well.

What you should look for is a normal distribution of purchase loans and whether or not the company can be supported financially without relying on the refinance portion. Second I would add back the refi portion based on past performance and evaluate the potential range (low – high) for predicting 2021.

Based on the above you will need to pull the purchase and refinance numbers, average them, and create a summary table to establish a range of potential funded units / volume. If you are using a tool like Telemetry BI, this will be an easy task taking a few minutes. 

Summary of the analysis:

I have adjusted the projected 2021 figures to reflect the percentage of changes based on the MBA reporting. 

For those of you who have experienced the end of the last refi boom, you will remember how it ends, it ends abruptly like shutting off a valve. No more rate and term refinance. Period. So there is no need to ramp down the transition period.

You may want to further calibrate these numbers by pulling out 100% of your rate and term refinance units before applying the 49% reduction (MBA) number.  This will set your lower range to a much more conservative starting point.

So we have just completed our first step, establishing a fair, conservative, representative distribution of purchase and refinance with lower and upper range (bounds):

     2021 predicted units:

                                                LOW                      %              HIGH                   %_

     Purchase                              1,053                     87            1,053                    56          

     Refinance                                162                     13                813                   44_

     TOTAL                                  1,215                                     1,2115

2. Simulate the changes in a consolidated P&L model using the metrics established in the above step.

If you are using a tool like Telemetry BI, you will have a consolidated P&L broken down in BPS (where applicable) and dollars. The important issue here is to segregate the fixed and variable expenses so we can run breakeven calculations properly.

NOTE: you can never be too diligent in determining which general ledger accounts (“GLs”) are fixed or variable and make sure that you don’t mix two types of payments into a GL or your break-even will be skewed.


Do not use a “Salary” GL to also post overtime and bonus. Where Salary (base) is a fixed expense, and OT and Bonus are variable. It’s probably worth talking to your accounting department before performing this exercise or you may overstate your break-even and make decisions to cut when you may not need to.

Sample Consolidated P&L and Cost per Loan (from Telemetry BI):

3.Establish an Accurate Breakeven:

In Telemetry, we look at break-even in units.

Telemetry also has a table that extrapolates the volume required (based on average loan amount x units). This is a nice view as it gives you not only a sense of what it takes to breakeven but potential net income as you close larger amounts. You should use the header in the table below as a sample of the calculation needed to determine what your company break-even is (in # of units based on your avg loan amt for the period). 

Applying the BPS from the consolidated P&L (avg costs per loan) and the breakeven # of UNITS, we can start to examine the potential impact of our range of predicted closings for 2021.

This sample shows how we apply our loan metric range to our projection model for next year

This sample shows how we apply our loan metric range to our projection model for next year

Adjusting out the derivative values is extremely important. Derivative values, as you know, are NOT cash receipts but mark-to-market values of assets (like Interest Rate Lock Commitments “IRLC” and Originated Mortgage Servicing Rights (“OMSRs”)). By adjusting out the derivative values, you will be looking at the actual net income (profit) based on the originations alone. Assuming all closed loans are purchased at booked (accrued) values, this would be the resulting cash from the production.

NOTE:  This comes in handy when you are thinking about owner distributions and maintaining your equity positions for covenant purposes. You can imagine if you distribute the GAAP reported net income shown above, and you hit the low end of the range, you will reduce equity by almost 500K (annually). 

Use of a tool like Telemetry BI will compute these values for you to help you avoid the “distributed on GAAP trap”.


4. Examining Vulnerabilities

So, now that we have determined the range (bounds) for our 2021 and run a simple consolidated P&L model, we can begin to think about how we can execute at the upper range and make the predicted 4.3 million dollars. What are the risks?

1.Margin squeeze - due to competition (as refi demand slows, the market will be oversupplied and will most likely become extremely price competitive).

2.LO Fallout due to :

      i.Inability for agents to pivot to originating a higher percentage of purchase loans

      ii.Agent thinking the grass is greener at another company that has “good leads” or “better pricing”

3.High Fixed Expense due to Over Capacity – It is likely that during the boom you have had to make decisions to hire new fulfillment staff to keep up with the workload. Of course,if the overall business trails off by 46 to 26%, you may not need all of the production staff and the goal should be to be able to break-even at the low end of your predicted range.



To investigate LO vulnerabilities, we can focus on a couple of things (1) the mix of loans they have been originating, (2) the average amount of loans during a normal cycle, (3) their ranking amongst the entire pool of loan officers. Here is a sample of some views exemplifying these investigations:


EXAMPLE 1: Loan Officers

LO: Alvan Condliffe

In the following graphics you can see that Alvan may be susceptible to the change as his mix of refi to purchase is 67% refi | 33% purchase. Given his 12-month trend, it looks like his average monthly closings were between 3 and 4 loans per month. If we apply the range to his productivity, he could be severely impacted:

Ranked 22 out of 55

Loan Purpose Mix:67% refi / 33% Purchase

Average # of closings:5

Avg with adj : 3

ACTION: Review 12-month goals and methods for generating more purchase loan opportunities.

EXAMPLE 2: Fulfillment Personnel:

You can examine fulfillment personnel the same way using a ranking system and productivity graphs (# of files they can manage every monthly). You might find that your capacity to produce loans far outweighs the projected volume of business. You may want to cut back where you are overstaffed.


(This sample is ordered lowest to highest of two milestones assigned to the Processor (1) Application – as the processing staff at ACME help the LO with the loan file origination requirements and (2) Processing – which includes all aspects of file processing prior to submission.)

Using Telemetry, you can manage all roles in your business (e.g. U/W, funder, closer, shipper, loan officer).


Given all of the information, we will determine the action plan (executables) for staff to work on so that we can achieve our high-side range and avoid losing  money during the transition.



5. Create an Action Plan:

1. Goals (vol / units)

Low (modified) = 133 units which is our break-even at current levels of expense

High= 155 units 

2. Max fixed expense (in $) = $800k per month

3. Max variable expense (in bps) = 138 bps

4. Maintain a 100 bps net margin

5. Min distribution of refi / purchase for LOs (in volume and units) =  57% purchase / 43% refi

6. Min productivity per role (# of files / max touches) – 15 files / 4x touches

7. Any reductions?

1. Sales expense or Fulfillment expense

    If we eliminate one team, we can save x$ which would reduce our fixed expense by $ and lower our break-even by Y;

2. Fixed or Variable Expenses

3. Demand related (potential) changes – evaluate vendor contracts when business slows to see if incremental costs can be reduced.


If you follow these steps, you will gain a much better understanding of the vulnerabilities of your company based on past activity and focus on refinance loans. This should help guide your thinking when making important decisions during the transition.