Erdmann Housing Tracker · Housing & Cities
TIER 4 Tue, 21 Oct 2025 19:30:04 +0000
I have been outlining a standardized package I will be making available for dozens of metropolitan areas that is based on estimating a lot premium, which applies uniformly across all homes in each housing market as a result of systemic underbuilding in recent decades. ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ | | ---|---|--- | | | Forwarded this email? Subscribe here for more --- --- # Metro Area Housing Analysis Packages: Part 4 | | Kevin Erdmann --- | Oct 21 --- | --- --- | | | --- | | --- | | --- | | --- | | READ IN APP --- I have been outlining a standardized package I will be making available for dozens of metropolitan areas that is based on estimating a lot premium, which applies uniformly across all homes in each housing market as a result of systemic underbuilding in recent decades. This novel data point unleashes some basic quantitative insights that can be used to inform investment decisions across locations in housing markets. Upgrade to paid The package provides: Part 1: * An estimate of the **lot premium** embedded in home prices in most markets today, which will correct in most markets as more homes are constructed and rents moderate. * An estimate of the **neutral rate of construction** required to counter excess rent inflation in each housing market. Part 2: * Using the estimate of the lot premium in each market, we can estimate the portion of home prices that are temporary aberrations vs. permanent. * Permanent factors: Population growth, income growth, past general inflation * Temporary factors: Cyclical fluctuations (fast), lot premiums (slow) Part 3: * The simple process of isolating the lot premium caused by underbuilding can help to highlight which homes within a market are most affected by the premium and most at risk of changing lot values. * Comparing homes within markets, exposure to construction costs, locational amenities, and the lot premium can be matched to investor or builder risk preferences. * Comparing homes across markets, and comparing the expected lot premium with market land prices, can identify markets with better risk/reward opportunities. #### Forecasts Isolating the portion of home prices that are attributable to inadequate construction can help to forecast conditions into the future. The sample market I have been citing in these posts is an extreme case here. For the past decade, between 20,000 and 30,000 units have been completed annually. That's about 7 units per 1,000 residents. That's higher than the national average (about 4 units). Before 2008, it was common for fast-growing cities, including this city, to have construction rates per 1,000 residents in the teens. Austin has been the only city to return to that, averaging about 14 units per 1,000 residents over the past decade. Figure 1 compares a rate of construction continuing at the current pace and a rate of construction that would be associated with a reversal of lot premiums. This may seem extreme. The scale of the forecast might seem unreasonable. Of course, the end user can adjust their expectations according to their own expectations and demand forecasts. This package takes a naive approach to demand. These estimates of the potential for new construction in this market simply extend the cyclically neutral level of demand from the previous 5 years indefinitely into the future. | | ---|---|--- Figure 1 How can the estimate of potential construction demand be so high? Well, among the largest metropolitan areas, this market has seen the most rent inflation over the past decade, even after permitting an above-average number of new homes. After adjusting for general inflation, this market has averaged more than 3% excess rent inflation annually, _for a decade_. In Figure 2, if you treat the rise in 2021 and then the flattening after 2021 as a full cycle of cyclically elevated demand growth that then corrected back to normal, you can easily imagine a relatively straight line from about $1,400 in 2015 to about $2,100 in 2025. A 50% rise, _in inflation-adjusted terms_. Zillow's rent estimate includes some compositional drift as new, better homes are built, which adds a bit less than 1% annually to the trend. The BLS estimates rent inflation for this market, and the metro area CPI Shelter component has risen by 44% more than non-shelter consumer prices since 2014. | | ---|---|--- Figure 2 Our baseline expectation of the supply gap in this market should be as shocking as that level of excess rent inflation is. If Figure 1 didn't make you blush, I would have been lying to you. Figure 1 shows a potential doubling of production. Keep in mind, doubling would simply move production up to a level equal to Austin production today and equal to a number common across a number of sunbelt markets before 2008. According to the BEA, population growth in this market has ranged between 1% and 2% over the past decade. Using a demand elasticity of -0.5, 3.2% rent inflation implies that there has been persistent demand for 1.6% additional annual population growth, which was met with rising rents rather than new homes. Either net migration or household formation were nudged down as those rents increased. We might expect some decline in population growth to follow Trump administration immigration policies, etc. So, the baseline demand expectation might be lower than a naive continuation of past trends. On the other hand, if national supply is short 15 to 20 million units, the potential demand for housing into fast-growing markets may not be that sensitive to recent population and immigration shifts. In an amply supplied market with uninflated rents, annual population growth of 30,000 households versus 60,000 households will have an immediate and sharp effect on construction activity. But, if there are 150,000 households that have either delayed moving into the city or have delayed forming new households, construction activity over time will not be as sensitive to short term changes in population growth, and in fact, this has already been the case. In this case, the potential for new construction will remain, but the trend in future rents and lot premiums will bend a bit more or less, depending on short-term fluctuations in potential population growth. Figure 3 shows past and forecast rents (using Zillow's ZORI estimate for the metro area). The red and black forecast lines correspond to the red and black production range in Figure 1. (Figure 2 was in real 2025 dollars, Figure 3 is nominal.) | | ---|---|--- Figure 3 Figure 4 shows the expected nominal lot premiums over time, again with the black line and red line corresponding to the top and bottom range of production shown in Figure 1. | | ---|---|--- Figure 4 Using these naive estimates, even with the aggressive production trend, lot premiums wouldn't peak until 2031. Figure 5 shows estimated nominal home prices (referencing Zillow's ZHVI estimate for the metro area). As the lot premium trends toward correction at a 60,000 to 70,000 units per year production rate, nominal home prices would generally level off for a number of years. | | ---|---|--- Figure 5 For comparison, Figure 6 shows the estimated production potential and estimated lot premiums in a smaller, slower-growing market that happens to be closer to a sustainable construction rate than the sample market is. The forecasts for this market aren't so extreme. At curent production rates, lot premiums are forecast to rise in this market, but not as sharply as they have over the past decade, and only a small rise in construction would bend the lot premium trend down. | | ---|---|--- Figure 6 This model provides some surprising food for thought. The sample market I have been referencing here is a market that has recently been soft because of the whipsaw of post-Covid population flows. It is also a market that has accumulated a significant lot premium. I have argued that those lot premiums are unsustainable, and that buying land for development in those markets comes with substantial risk. Yet, this data suggests that in this particular market, aggressive land acquisition might be a profitable strategy for a few more years, and that the potential for construction of new homes is very high. This is a market that likely has some turmoil in land markets - builders bailing out on land options that are now worthless due to the recent market softness, cutting planned community development, etc. Local underwriting knowledge, informed by this data, could point to some surprising opportunities for capital allocation. My inclination would have been to suggest that builders keep land pipelines short in this market because each lot has a $150,000 premium that is bound to dissipate. But, maybe, in this odd case, it is still early enough in the correction that those are long-term concerns rather than immediate concerns. And, in the meantime, there is potential for building a lot of homes. Upgrade to paid Maybe this is a market that a bold contrarian could set up for unusual profits over the next few years. You're currently a free subscriber to Erdmann Housing Tracker. For the full experience, upgrade your subscription. Upgrade to paid --- | | | Like --- | | Comment --- | | Restack --- (C) 2025 Kevin Erdmann 548 Market Street PMB 72296, San Francisco, CA 94104 Unsubscribe