Saturday, January 13, 2007

Bottom Spotting

With the new year upon us, many "interested parties" are calling for a bottom in the Massachusetts housing market. I think their pronouncements are very premature. Keep in mind these points:

1. To real estate professionals, the bottom in a market occurs when SALES bottom, not when prices bottom. Their livelihood depends on turnover (commissions), not prices.

2. Affordability in MA is near record lows (see recent post). There is very little room for upside price movement.

3. Catch-22. Dropping prices will increase affordability and potentially increase the pool of buyers. But, stabilization in prices is needed to motivate and reassure buyers. This process will continue to play out with declining prices and sales.

4. Credit standards are tightening. An unprecendented number of mortgage holders will face rate adjustments and payment shock in the coming year.

5. The building cycle is still bringing new product to market as demand continues to dissipate.

6. Markets take a LONG time to turn. Think about how long the anticipated peak in the market took to come to fruition.
7. Psychology has only started to turn negative. At a market bottom, the psychology of fear will dominate greed. Sellers will capitulate.

8. Supply and Demand. "Months Inventory," the number of months it would take to sell all existing inventory at the current sales pace, continues to increase. This means more houses are coming on the market than are selling. By this metric, the market has not stabilized, it continues to weaken. Here is months inventory for the last 4 years:

Last November there was ~8 months of inventory. This November, ~12 months. Think the trend is decelerating? Here is the year over year change in months inventory over the last three years, along with a six month moving average.


How important is this metric to prices? Plotted below is the historical relationship between Months Inventory (averaged over each year) and Appreciation Rates (averaged over each year) for the past 14 years.

Looking for an early sign that prices will firm? Wait for "Months Inventory" to show a meaningful and sustainable year over year decline. Until then, all pronouncements of a bottom in PRICES are fantasy.

6 comments:

TheLievense said...

Big thanks for all of these awesome graphs like usual. Makes the whole market so much simpler to follow. Keep up the good work!

notsofastlouie said...

Warren Group DECEMBER condo numbers are out for specific towns.

Total condo sales for the Urban Core Market for December 2006 INCREASED to 500 from 438 for the same month last year. It's hard to determine an overall median price, but the median price was HIGHER in towns representing 72% of the total volume of sales.

See the numbers for yourself below - they are posted by town.

Boston (defined by the Warren Group as Back Bay, Beacon Hill, North End, South End Downtown)
#Sales 191 (12/06) vs. 194 (12/05)
Prices $579,000 (12/06) vs. $513,750 (12/05)

Dorchester:
#Sales 116 (12/06) vs. 46 (12/05)
Prices $278,500 (12/06) vs. $261,250 (12/05)

Cambridge:
#Sales 89 (12/06) vs. 87 (12/05)
Prices $389,000 (12/06) vs. $430,000 (12/05)

Somerville:
#Sales 52 (12/06) vs. 58 (12/05)
Prices $352,000 (12/06) vs. $299,000 (12/05)

Brookline:
#Sales 52 (12/06) vs. 53 (12/05)
Prices $411,000 (12/06) vs. $440,000 (12/05)

bostonbubble said...

notsofastlouie,

Sales are only up in your "urban core market" because of the large spike in Dorchester, which also happens to be the cheapest of the group and by a large margin at that. You previously did not include Dorchester in your "urban core market" stats - the earliest reference I found to where you defined the "urban core market" listed only central Boston, Brookline, Cambridge, and Somerville (see http://masshousemarket.blogspot.com/2006/11/october-sales-prices-and-inventory.html ). If you exclude Dorchester now, sales are down from 392 to 384.

notsofastlouie said...

Bubblebuster: Thanks for the comment. Actually I've been including Dorchester in my definition of Urban Core Market for a while (see my posts under "Bottom Spotting," "November Market Wrap" and "Affordability is the Problem").

In my post that you reference ("October Market Wrap"), I did not include Dorchester when I argued that condo sales in the urban core market were flat vs. the previous year. But if you include Dorchester for that month, condo sales would have actually been HIGHER in October vs. the same month in 05.

My definition of Urban Core Market grew to include Dorchester when I looked it up and saw how significant the condo market is there (2006 condo sales increased to 862 from 644 in 2005).

My point is that all the theories on this site, which are based on state-wide stats, do not explain the robustness in condo sales in the urban market.

Folks: I am not a realtor or a representative of MAR. Far from believing metrics to be irrelevant, I try to present additional demand metrics (number of sales, income, job growth) and supply metrics (construction, town-specific inventory) not offered on this blog and then offer my own theory to explain the data. You don't have to agree with me, but I think that additional data and debate are not bad things.

If I didn't think that there were worthy interlocutors on this blog I wouldn't waste my time on it. Let's keep the discussion friendly and free of ad hominem argument.

Dr Housing Bubble said...

CNN has a cover story today stating that December saw a large spike in foreclosures. The data you present again goes to show how all this is very predictable. In addition the article goes on to point out that the large jump was mostly in the sub-prime market; and considering that 25% of 2005 and 2006 originations is in the sub-prime market is there any doubt as to why this is happening?

In addition, many of these loans will reset in 2007 so this is simply the beginning.

http://drhousingbubble.blogspot.com

bostonbubble said...

notsofastlouie,

In the spirit of constructive criticism, I'd like to suggest that your method for selecting areas to include in your "urban core market" inherently biases the numbers. You added Dorchester because the sales were increasing so much. So naturally, your total will show better than average sales because that was the criteria used to select the areas - there will always be *some* area that performs better than average.