We Need More Homes!
1. What would you buy if you had a million dollars?
2. According to Zillow you could buy 2 homes in California ($512,000 average) or 6 homes in Texas ($172,000).
3. What do TCAC 9% award winners do with $1 million of government money? We build 2.3 units per million.
4. We build at an average cost of $423,000 per unit, raising less than $30,000 in conventional financing.
5. Besides spending $143,000 per unit in public funding we awarded $296,000 credits per unit (up 25% from our worst year ever—2016).
6. Counting credits alone, California gets $1 billion each year to develop affordable housing. In 2016 we built 4,508 low income units at rate of 4.5 per million. In 2017 we built 3,676 units at a rate of 3.6 per million.
7. This embarrassing reality isn’t helping our housing crisis.
8. Over the past 13 years I have submitted more than 40 applications to CTCAC. I constantly study the 9% program and regularly publish insights about it.
9. I have been appointed to a number of CTCAC working groups, including the advisory committee for the California Affordable Housing Cost Study, which resolved in 2014:
"To take advantage of these opportunities to lower the cost of developing multi-family affordable housing in California, additional incentives for producing more units at a lower cost could be incorporated into existing state policies.”
10. Yet since that study was released we have produced fewer units every year. Meanwhile, our competitive scores have gone up as our public funding has gone up.
11. Being rewarded for “producing less units with more resources” is asinine. But that’s our scoring system.
12. Our industry is regularly scrutinized by “cost per unit” and yet we remain embarrassingly inefficient.
13. We need more housing and we need a better system to build it.
14. I have been challenged by Mark Stivers to propose a better system and invited by TCAC committee members to educate them on the issue.
15. With feedback from Mark and others in the industry, I have developed a proposal to modernize the 9% tiebreaker, which will receive state-wide attention in 2018.
16. My goal here today is to share that proposal with you and request your support.
17. In short, I propose we award 9% credits to the projects that produce the most units per million credits.
18. Whoever develops the most low-income units per million credits requested wins.
19. This “return on investment” measurement will motivate developers to maximize units and minimize unnecessary costs. They will take advantage of tools like density, economies of scale, and efficient design to score well.
20. In this system there are two fundamental ways to improve your score: increase units (numerator) and decrease credits requested (denominator)—build more or request less.
21. The typical behaviors that will result from this system:
a. Developers will try to maximize density to produce more units
b. Developers will try to cut unnecessary costs
c. Developers will try to get the best rates and pricing
d. In short, affordable developers will act more like market rate developers
22. The benefits of this system include:
a. It clearly communicates what it wants—units
b. It will harness developer creativity
c. It doesn’t prescribe any one financing plan
d. It is understandable and defensible
e. It will produce what we need—more homes
23. Having proposed efficiency based scoring systems for years, I’ve heard an objection or two.
24. Stakeholder objections come in two basic flavors:
a. I do XYZ which increases my costs, and I’m disadvantaged for it (e.g. I build infill developments)
b. I do ABC making my units more impactful, yet I receive no advantage (e.g. I build close to transit)
25. The good news is that objections such as these can be researched, discussed, and (if deemed good policy) addressed with adjustment modules in the system.
26. Consider the objection, “infill developments will be disadvantaged.” Less densely populated areas require less intense construction methods, therefore cost less, therefore need less credits, and therefore will score better.
27. Research, using statistical analysis, proves that building in (a) highly populated areas, (b) high median income counties, and (c) high rents districts, cost more to develop housing in. Using regression analysis the effects of each factor can be measured and those findings used to craft adjustment modules that neutralize the effect.
28. If policy makers deem it appropriate, the system can include an adjustment module that neutralizes, incentivizes, or dis-incentivizes any given project characteristic (in a simple calculable manner).
29. Adjustment modules will either (a) add or subtract from the denominator, simulating the justifiable need for more or less resources or (b) multiply the numerator by more or less than 100%, simulating the greater or lesser public benefit per unit.
30. The beauty in this scoring system is, after legitimate policy concerns have been adjusted for, developers will be motivated to stretch our scarce resources—to better serve our communities—to build more homes.
31. We need more homes! Please assist me in educating TCAC and the industry how to do so thoughtfully.