Appraisers' Free Forum

Edd, the Census Bureau is getting pretty granular. They'll have counties/areas down to a 20,000 population online by 2010; they've got it down to (I think) 50,000 now. You can disaggregate the reported data to arrive at some of the numbers.

E.g.:

100,000 population; 2.5 people per household; therefore, there are 40,000 households.

Banks allow, say, 30% of your income to be consumed by a mortgage payment (Roger? How close am I?)

Last sale of subject was $150,000; typical LTV is 90%. Amount financed, $135,000; payment $800± at 6%.
So, to make the cut, a borrower has to make at least $2700/month ($32,000/year).

Now look at the Census figures. How many households in your MSA have incomes above $32,000/year? Knock off the ones who wouldn't consider living in a house that cheap. (Your call)

The number you get will give you a pretty good demand number for SFR's that cost between, say $120-180K. But that sort of thing should be reported as a range anyway. (Pay attention to the error range reported, too.)

After all, they are supply and demand *estimates*, and shouldn't be treated as exact numbers.

Thanks Jim. Good stuff that I am going to use. But what the heck do I do with this OFHEO HPI? It seems to be used as a stand alone authority with resect to market value trends and yet OFHEO writes disclaimers that say it should not be used that way. What gives? How can I use the HPI in the analysis of market trends where I am?


EDIT": Where do you get your financing info from? There is an etrenched and official belief in Colorado that that info is a secret. You know the old confidential thing.

Edd, you are one of the heavy hitters.

I cannot answer your question directly, but don't back down. The index information is useful, but only in the context of your own analysis. The alternative is like giving up the appraisal process to an AVM.

I see three possibilities. Your analysis could confirm the index numbers. Your analysis could refute the index numbers. Or, your analysis could be inconclusive.

I the third instance, I would tend to go with the index. However, in the second instance, my analysis is better than the index because it is more topical and more specific to the appraisal problem at hand. In that case I would report and weight the index, but rely more heavily on my own data. Then, if someone wants to dispute that, I would give them the same line I do if they want to dispute my value: "You paid for my opinon, here it is. I will only change my opinion if there is new data that indicates a change is necessary. And, BTW, analysis of new data may involve a new appraisal fee unless it is something we can agree that I should have found to start with."

Perhaps that line will get BOA or OFHEO to release some actual data to you. If not, you are not responsible for what you cannot know. When I am confronted with a bully (and most bankers and most government employees are) my response is always the same: "prove it."

You're welcome, Edd. I just finished AI's new course titled General Appraiser Market Analysis and HBU in Chicago. Go take it when it gets near you. I promise you that you'll learn some really good techniques for these topics. Disaggregation is just one of them, and the instructors offer about two pages of data sources. (This course is for commercial stuff, but the techniques can be applied to residential). You can apply the disaggregation technique to just about any of these data sources.

And I wouldn't try disaggregating an index unless I knew exactly how that index was calculated. For example, if all I have is a parking index for a commercial buildings and their gross rentable area, and I know that the index is calculated by dividing the number of parking spaces available by the number of square feet of gross leasable area (GLA for four-figure fees), then I can get the total number of available parking places by dividing the total GLA in the study area by the parking index. (I have no idea why one would want to know that; I'm just using it as an example. Could be a feasibility study: Is there marginal demand for a new parking garage?)

Edd wrote:Where do you get your financing info from? There is an etrenched and official belief in Colorado that that info is a secret. You know the old confidential thing.

Call lenders. Ask them how much income can be allocated to a house payment? Ask whether that number is PITI, or just principal and interest. Hell, tell them you're a real estate agent, and want to pre-qualify your prospective buyers. You may get a different number from different lenders; but maybe Fannie and Freddie have guidelines for it.

Can you see how demand may have changed--instantly--in July 07? Demand + ability to participate is one of the elements of value. (i.e., DUST). If lenders suddenly went to a 28% requirement from a 35% requirement, demand changed! Without a doubt. Supply didn't change. How many markets went from increasing/balanced to oversupplied in a single week?

Take that course. It is extremely valuable.

There is no doubt this liquidity crises is a demand side problem with a dusting of overzealous supply here and there. And as for the sub prime crises, the frickin' thing was inevitable and therefore foreseeable. It is no crises. It is more like haphazardly planned event. But, the people who order and pay a pittance for fast, cheap and Fannie form stuffed appraisals saw a chance to make a quick buck at somebody else's expense.

Jim Plante wrote:

Edd wrote:Where do you get your financing info from? There is an etrenched and official belief in Colorado that that info is a secret. You know the old confidential thing.

Call lenders. Ask them how much income can be allocated to a house payment? Ask whether that number is PITI, or just principal and interest. Hell, tell them you're a real estate agent, and want to pre-qualify your prospective buyers. You may get a different number from different lenders; but maybe Fannie and Freddie have guidelines for it.

Can you see how demand may have changed--instantly--in July 07? Demand + ability to participate is one of the elements of value. (i.e., DUST). If lenders suddenly went to a 28% requirement from a 35% requirement, demand changed! Without a doubt. Supply didn't change. How many markets went from increasing/balanced to oversupplied in a single week?

Take that course. It is extremely valuable.

OK. So you use some sort of pro-forma LTV?

Edd Gillespie wrote:

Jim Plante wrote:

Edd wrote:Where do you get your financing info from? There is an etrenched and official belief in Colorado that that info is a secret. You know the old confidential thing.

Call lenders. Ask them how much income can be allocated to a house payment? Ask whether that number is PITI, or just principal and interest. Hell, tell them you're a real estate agent, and want to pre-qualify your prospective buyers. You may get a different number from different lenders; but maybe Fannie and Freddie have guidelines for it.

Can you see how demand may have changed--instantly--in July 07? Demand + ability to participate is one of the elements of value. (i.e., DUST). If lenders suddenly went to a 28% requirement from a 35% requirement, demand changed! Without a doubt. Supply didn't change. How many markets went from increasing/balanced to oversupplied in a single week?

Take that course. It is extremely valuable.

OK. So you use some sort of pro-forma LTV?

Edd, you're going to have to learn enough about lending policies of your local lenders to be able to segment the borrowers and generalize these lending policies, and it's a local job. If, for example, you're doing SFR's in the 50-75K price range, a bigger percentage of potential borrowers will be unlendable because of poor credit than will those buying in the 200-225K range. E.g.: $150K house, two buyers. One has a 700 FICO, the other has a 780. The one with 700 will have a lower LTV requirement; he may be able to finance only 80% instead of 90%. His interest rate may be higher; 6.25 instead of 6%. But his payment is only $740±, $30K/yr ±, so he can still get the loan, and stays in the demand calculation. (Assumes he can come up with the $30K down payment somewhere.) (Or find Skippy to overappraise the house by $30K). And he needs less than $30K/year to stay under our theoretical 30% of income limit for housing budget.

Jim, that is good stuff. But, one caveat... census data is usually quite old. In an area that is growing or shrinking at even a moderate rate, it would be out of date pretty quickly. Do you have any particular way to deal with that issue?

Steve,
Better go have a look at the Census' web site. They now have American Community Survey (ACS), which is updated annually. As i said earlier, the Census Bureau and HUD are getting more and more granular in their survey areas.

With annual updates of certain set areas, one can put a fairly fine point on a mathematical forecast. One comprehensive survey that HUD does uses the same properties over and over, once every year. These analyze the sale/resale of the same property (when it happens), every time.

I just discovered this latter survey, and don't have a good handle on it yet. But HUD will even give you the survey questions and the individual responses (with any identifying information removed). So unlike many government sources, this one has the possibility of being used for a fundamental (as opposed to inferred) analysis.

Start browsing the Census site in the American Community Survey section. Sooner or later it'll lead you to the HUD surveys. (They share information.)

One more thing, to point out the obvious: if you're in a rapidly changing market, meta-analysis and inferred analysis are worthless. You'll have to do fundamental analysis (at least level C), and do most of it yourself.