Insights
Market reports, forecast data, industry insights, and more from iEmergent.
iEmergent has expanded our technology and tools to help lenders use all types of data in their lending strategies, but our foundation is our proprietary forecasts. We provide forward-looking data that drills down to the neighborhood level and quantifies future mortgage opportunity in markets in the U.S.
In other words, we use data to show lenders where loans are going to be. We can see, in each neighborhood in the U.S., how many purchase loans will occur this year, next year, and the year after that.
Here, we’ll share insight into our forecasting approach, the fundamental concepts of our purchase forecast model, our process for validation, and our consistently high accuracy.
iEmergent forecasts mortgage opportunity for markets at various levels of geography—census tracts, counties, MSAs, regions, states—across the 50 U.S. states and the District of Columbia.
What is mortgage opportunity?
A forecast of mortgage loan volume in units and dollars that will occur over a specific period of time within a specific geography.
The purpose of our forecasts is to help organizations make better business decisions that lead to a successful, sustainable future. By knowing what’s next in a market, lenders can anticipate change and capture opportunity—as efficiently as possible. Amid economic uncertainty, shifting homebuyer behavior, and regulatory changes, leaders need more than expertise and experience to maintain and improve performance. They need reliable data that show what’s going to happen in their markets.
Our insights help mortgage lenders:
National mortgage opportunity forecasts are nice, but what can an individual MLO or lender do with that information? Nothing much actionable.
That’s why our forecasts are built from the bottom up: from the census tract level, which roll up into county, MSA, region, state, and national forecasts.
A lender in California shouldn’t have to—and we’d argue couldn’t effectively—use state or national forecasts to make strategic decisions for their local markets. Each and every census tract expands, contracts, and changes in unique ways, creating micro markets across the country. Understanding what’s expected for every market each year for the next five years is the only way to build strategic, localized lending approaches with confidence.
There are currently 84,414 census tracts in the U.S.
Geography isn’t the only way to narrow down forecasted lending activity. Borrowers are not a monolith, and your forecasts shouldn’t be either. In addition to market breakdowns, iEmergent forecasts can be segmented by:
iEmergent clients can view forecasted loans and dollars for, as an example, moderate-income Black borrowers down to the neighborhood (census tract) level. That’s powerful insight into the future that’s not possible with any other data or tool.
Understanding the nuances of each market gives lenders a data-backed path to building forward-looking lending strategies—an imperative in today’s lending environment.
The iEmergent forecasting method is a hybrid of traditional demand forecast models, and it has evolved significantly since 2004, when we issued our first forecast. From the beginning, our model did not attempt to explain why each mortgage market behaves as it does; instead, we focused on market outcomes to identify how many and what type of loans will be originated over the next one to five years.
A lot of variables go into our forecasts, but the following concepts are fundamental to our approach.
The behavior of mortgage markets is complex—there are hundreds of indicators, trends, patterns, and events that impact how and why markets behave as they do.
After analyzing millions of loan application records over decades, iEmergent recognized specific patterns of behavior that captured the complexity of these multiple factors. What emerged was the Purchase Mortgage Generation Rate (PMGR), or the rate at which an individual market produces purchase mortgages. Not only is the PMGR of each market unique, but it is predictable by our model. Thus, it is the primary driver of our forecasts. The PMGR inherently captures the impact of broad-scale economics, decades of history from the HMDA Loan Application Record, homebuyer and housing behavior patterns, and other prominent indicators.
Through the PMGR, we simplify the projection of loans and dollars through a single indicator.
Many traditional mortgage industry volume forecasts are calculated using top-down, optimal-utility methodologies. Assumptions are made based on broad supply-side market behaviors that follow Say’s Law of Markets: “Aggregate supply creates its own aggregate demand.”
In contrast, iEmergent applies a “demand-driven” approach that captures the changing household-buyer patterns that define housing patterns over time. Here’s why, as explained more than a decade ago by our founder, Dennis Hedlund:
"Houses can’t buy themselves. Low interest rates can’t shop for homes to buy. Available credit won’t spontaneously buy homes. Low housing prices don’t buy homes. Secondary markets by themselves don’t incent people to buy homes. Big inventories can’t write a check for the mortgage. Households buy homes. And if households don’t buy homes, then mortgages aren’t originated."
Households that could potentially finance a home in a given year constitute the homebuyer pool.
The homebuyer pool is the number of households that are ready, willing, and able to buy a home.
These potential buyers are the foundation of homeownership demand and the second pillar of the iEmergent forecasting methodology.
Per the U.S. Census Bureau, there are an estimated 132.2 million households in 2025. Each year, iEmergent partitions households into three groups:
Using probability theory, adjustments are made to each of these three groups as new households are formed, and households convert from one group to another (e.g., non-homeowners become homeowners, and homeowners without a mortgage convert to homeowners with a mortgage).
As of January 2025, the U.S. Census Bureau estimates an overall homeownership rate of 65.1%.
Most important to our forecast methodology, however, is determining the size of the homebuyer pools for upcoming years. Despite 132.2 million households in 2025, not all of them will be ready, willing, and able to purchase a home in the next twelve months. Therefore, we account for that portion of the pool by creating a fourth partition that removes those households that are least likely to originate a mortgage. Households that have recently purchased or refinanced a home are taken out of the pool, as are those households who are unemployed, are struggling with balance sheet/credit issues, or starting foreclosure.
These segments tell us about the demand side of mortgage origination by understanding who’s in and out of the homebuyer pool. In 2025, only 992 million households (75.0%) are part of the homebuyer pool.
Housing markets are complex ecosystems, and homeownership behaviors are constantly evolving, ebbing, and flowing. The supply-demand dichotomy will eventually establish new equilibriums—at different points in time for different communities. The iEmergent forecasting methodology is built on the reality that homeownership demand is a critical driver of mortgage opportunity.
The relationship between each market’s Homebuyer Pool and PMGR determines the final outcome of our forecasts: the number of purchase mortgage loans and dollars that will be originated over the next one to five years.
Telling the future isn’t easy. Thankfully, our data-driven models—refined year after year—give us insight into what’s ahead. We pride ourselves on maintaining well above the industry minimum accuracy standard of 70% for predictive analytics. Each year, we compare our forecasts to actuals and remain well above average for accuracy.
What are the “right” actuals? The annual Home Mortgage Disclosure Act (HMDA) data release is the most comprehensive reporting of mortgage loan data. While we are aware that HMDA omits some volume of mortgage loan data because some small volume loan originators don’t have to report their information, it represents the most relevant total of mortgage originations, and this total is our target when we forecast the mortgage market.
iEmergent is one of few institutions putting out consistent national mortgage purchase origination forecasts. How do our forecasts compare to the other key publicly available forecast sources, namely Fannie and the MBA?
First, we have to determine what “accurate” means. HMDA data is a complex set of information. Different “actual” totals can result depending on the parameters used to query for the total purchase volume originated. In fact, among iEmergent, Fannie, and MBA, none of us ends up with the same national total year to year.
The following chart shows each forecaster’s purchase origination total via line graph, with iEmergent’s deviation shown as a percent. Our total forecast is generally a bit lower than the others.
For this reason, our accuracy assessment uses each forecaster’s individual total as the reference point to calculate deviation from the actual number.
Another factor comes into play when comparing accuracy among forecasts, which are constantly being updated. Fannie and MBA update their forecasts on a monthly basis while we update our forecast quarterly. Thus, choosing when to take accuracy snapshots becomes an important consideration.
Two forecast timeframes seem particularly important. For lenders, an important strategic planning snapshot is in the third quarter of a year, forecasting for the following year. When we compare a forecast released in Q3/Q4 for the next year, we refer to this as the “next year’s forecast” accuracy analysis.
A second critical snapshot looks at the last estimate before actual data is released. We call this the “last estimate” accuracy. As expected, it is usually, though not always, closer to the actual number.
To illustrate the timing of these forecasts, the chart below shows a timeline for 2022 as an example.
Using each forecaster’s final origination totals as the actuals, here are the raw deviation results between forecast and actual for next year’s forecast. Negative deviation means the forecast was lower than the actual result; positive means the forecast was higher than the actual result.
iEmergent trends closer to actuals than Fannie and MBA for our “next year’s forecast,” with especially outstanding results for our 2019 and 2024 forecasts.
We can further analyze accuracy by using absolute percent error (APE), which turns negative numbers into positives and shows accuracy as a single percentage. Here are the results for both the “next year’s forecast” analysis and the “last estimate” analysis for each forecaster. Each chart also includes both 6-year and last 3-year mean absolute percent error (MAPE) calculations, which show accuracy over time. iEmergent has the best "next year's forecast" accuracy over this 6-year period (9.5% MAPE), and it has improved for the last 3 years (8.4%).
Note the change in scale between charts because of larger error rates on “next year’s forecast.”
In the "last estimate" category, Fannie was best over the full 6-year period (2.7% MAPE versus our 3.9%), but we've been best over the last 3 years (only 1.2% MAPE).
In 2021, which was an especially volatile year, all forecasters’ “next year’s forecasts” were 20%+ off, and everyone but Fannie was 10%+ off on “last estimate” forecasts.
As proof of our continued improvement in forecast methodology, iEmergent has a sub-2% error rate for the last 3 years on our “last estimate” forecasts, a remarkable result.
To this point, our analysis has focused on purchase origination forecasts. When we add in refi loans, error rates increase. That’s because refi originations are so volatile, even more so in recent years as mortgage interest rates have swung from all-time lows during the COVID recession to generational highs during the post-COVID recovery.
However, during each year, all forecasts adjust toward the ultimate actual results, so by the “last estimate” snapshot, forecasts are pretty good estimates of the actuals.
Note: Again, notice the difference in scale between the two charts.
Once again, after adding refi, the iEmergent forecasts have better accuracy than the Fannie and MBA forecasts over the 3-year and 6-year periods. Our “last estimate” forecast is usually within what should be considered an exceptional 2% error range.
National forecasts are important, but we pride ourselves on having the most detail for the future mortgage market segments. Such detail is of little value if it is not accurate, which is why we also analyze our forecast accuracy at various geographic levels and across segments.
Each year, we have validated our forecast results against the actual HMDA data for core-based statistical areas (CBSAs), counties, and the nation. The table below captures the volume accuracy percentage—((forecast dollars - actual dollars)/actual dollars)—for recent forecast years at the 18-month, 12-month, and 6-month time horizons for counties in the U.S.
Volume accuracy percentages for various levels of geography and time horizons.
The chart below shows our 2024 purchase loan accuracy for the Top 30 Metropolitan Statistical Areas (MSAs), using the average of our various time horizons and comparing them to the HMDA actuals.
In our validation studies, we also examine the Absolute Error (AE) by individual census tract. The chart below summarizes the census tract-level AE for 2024 purchase loans, comparing iEmergent’s 2024 forecast to the 2024 HMDA actuals.
In more than 70.1% of the 84,414 nationwide census tracts, the iEmergent forecast error was under 10 loans. At an AE of 15 loans, iEmergent’s census tract-level accuracy increases to 83.9% of all tracts.
We also forecast various segments, including race/ethnicity. Our accuracy rates for these breakdowns continually exceed industry expectations. For 2024, our Q4 forecast (made in December 2023) accuracy is at the national level for each borrower race/ethnicity group was:
Our overall national accuracy remains in line with—and often higher than—other major national forecasts, despite (or likely thanks to) our very different bottom-up approach. And our focus on market-level, segmented forecasts continues to help lenders take action to grow originations and meet sales goals.
As a forecast of the national purchase origination total, iEmergent’s forecast is:
Our “next year’s forecast” MAPE results, which show our near-term forecast performance, have averaged under 10% error at the 3-year and 6-year time periods, despite some extremely volatile years for the purchase origination market. For the more normal years of 2019 and 2024, our forecast MAPE rates were 0.8% and 0.3%, respectively—truly exceptional results.
Today’s mortgage lending market is challenging—lenders are having to do more with less, and competition is sky high. Knowing where loans will be in your markets—and which borrowers are expected to take out those loans—is like having a crystal ball.
When you understand your market—its past, present, and future—you can deploy strategies and resources with confidence, knowing they’re based on what’s next in that market.
Lenders use iEmergent’s data and tools to:
See how you can use iEmergent to meet your lending goals: Request a demo
Since our founding, iEmergent has been committed to producing and offering accurate forecasts of mortgage opportunity to organizations of all sizes and types. We help these lending institutions make decisions that improve their long-term performance and profitability. The forecasting approach, models, and methods we use were carefully chosen according to how well they help serve that purpose. Experience has taught us that the best models and methods are built on strong and proven fundamentals but are constantly being tested and modified to ensure ongoing efficacy.
Get the latest forecasts and analyses in your inbox.
See how iEmergent’s forecasts and tools come together in Mortgage MarketSmart to help lenders set and reach enterprise, regional, branch, and MLO goals. Schedule a Demo