Wednesday, February 18, 2009

JP's market analysis January 2009

One of the blog regulars, JP, planned to post the following analysis in the comments section but it deserves more room and attention. I’d like to thank JP for taking the time to share his thoughts with us.

JP's analysis...

As a mathematician, the analysis of the monthly RMLS report is always a pleasure. While it is my personal opinion that macro market conditions are very bad, I also look for bright spots in the market. Currently I do not see much-possibly the increase in pending sales, but that may be part of the normal trend.

Poor indicators include:
-The level of Inventory.
-Average prices are down.
-Median prices are down.
-Inventory in months has skyrocketed.
-Total Market volume (closed sales X average sales price) is very low--this suggests that liquidity is approaching zero.
Total time on market is up.

And my unscientific consumer sentiment measure is very low [=unfavorable]. Of course there will always be those who think the market is going up, and there will always be those who think the market is heading down. I don't know about your crystal ball, but mine is often foggy.

I invite questions and further observations about what the data might represent. There are often times when the data will not specifically include or exclude a possibility. Other times it is possible to exclude different possibilities from the data set. It is just as important to understand what the data does not say as it is to understand what it might suggest.

As always there is a tension between trend analysis and the Efficient Market Hypothesis (EMH). The EMH suggests that no matter how much past data you have, it’s impossible to predict a future event, and if you could predict these events, then it would already be priced into the market. This is a basic overview of the two theories—please do further research about this issue.

Also please do not take this as investment advice. This is not written to any one person’s individual circumstances. I am not suggesting that one go short or long. Personally, if I was going to invest in real estate, I’d look at the government backed Ginnie Mae, but your risk-reward may vary.

Let’s start by looking at pending sales and inventory in months, which is calculated by taking the active listings at the end of the month and dividing by the closed sales for the entire month.

Pending sales are going down, while inventory in months almost doubles every year.

Now let’s look at closed sales, average selling price, and total market volume (closed sales X average selling price).

Please note that the average selling price in January 2009 was essentially the same as January 2006, 36 months prior. I would guess that the prices are statistically the same ($500 difference on $297,000 = 0.17%).

REALTORs who have been in the industry for three years have experienced, on average, a pay cut of over 50%. Alternatively, if 50% of the REALTORs exited the industry, then on average the remaining ones would still be taking a pay cut.

At 732 sales are getting discrete enough that I will start considering whether or not an individual agent actually makes the same number of sales as in the past. If you take a constant field of 1,000 agent pairs, then on average each agent made 1.8, 2.0, 1.1, 0.7 sales during January 2006, January 2007, January 2008, January 2009 respectively. Thus given a constant field of 1,000 agents, some agents were clearly left without any sales during January 2009.

The law of demand suggests that as prices go down, quantity sold will go up, or, depending on elasticity, the quantity sold should at a minimum stay the same. In this market, we have both quantity sold going down and prices going down. Thus the demand curve has shifted to the left.

At the same time, the law of supply suggests that as prices go down, so should the quantity supplied. In this market, however, we are seeing an increase in supply as prices are going down. This is probably because of the length of the supply cycle. The supply cycle is probably 24 months or more. Thus once suppliers realize that prices are no longer attractive, it takes up to 24 months to stop the flow of new places onto the market. This added supply should suggest further price reductions, and so on until a new equilibrium point is reached.

Now let’s consider the sales of million dollar plus homes. As most of you know, I am tracking one in particular. My question is simple: How many million dollar homes (and I am going to use a number close $1,000,000) could sell given the RMLS statistics. I hope it is clear: Not many. There just isn’t much room in the total sales volume. Make whatever nice assumptions you wish, the bottom line is that not many million dollar plus homes sold.

PR suggested that the numbers need to be seasonally adjusted. While it is true that January is a low point on a historic basis, the numbers presented above concentrate on January from on year to the next, and January 2009 is very low in terms of quantity sold and total sales volume.

Using trend analysis, we can expect that sales will increase during the summer months, but given that the quantity of sales is at a historic low, it does not take much to suggest that the quantity of sales should increase. It should also be noted that at 732 sales, January 2009 was the worst month for many years. Again, this is in an environment of falling prices.

I am sure there is going to be some claim about the weather, so let me cover that too. The worst portion of the weather was in December 2008. Unlike December 2008, the majority of January 2009 was not very interesting on a climatologically basis. That being said, let’s review:

Possibly the drop in 2008 may be partially explained by weather. I am not sure that the entire drop in closed sales can be explained by the weather, but it might be difficult to dispute.

Possibly the drop in 2009 may be partially explained by weather, but keep in mind that those who really wanted to list in December 2008 could have listed in January 2009—possibly these sellers are delaying until February 2009? Even if the weather had something to do with this, a lower number of new listings reduces the inventory in months, all other things being equal, yet the inventory in months is over 18.
Average selling prices have been declining since mid 2007, given a reasonable statistical error band (no, I didn’t compute one, but if you actually run statistical computations and find a different result, please post it in the comments).