What Statisics are Useful and which Statistics are Misleading
by Mark Pruner
There are a lot of statistics in real estate and they can be manipulated to tell different stories. Gleaning the gold from the lead is not always easy.
Overall house sales by month
One graph you see quite a bit in my Greenwich Sentinel column is the number of single-family home sales by month. The graph I usually use has home sales for each of the last two years, the current year and the 10-year average. What is not included in, but you often see presented elsewhere are sales numbers that include all types of real estate; co-ops, condos and sometimes even multi-family and land. If you are the Town Clerk and you want to see if conveyance taxes are going up or down the whole market is fine, but it’s not so good if you want to know how each type of property market is doing.
Condos, land and multi-family all have different types of buyers with different demand curves. There is some overlap, but generally house buyers don’t look at condos and families looking to buy land and build a new house don’t look at multi-family investment properties. Lumping them together muddies the state of the house market, and due to the size of the house market size, the other markets stats are overwhelmed by the housing market. So, when you see a sales number check and see what types of property are included.
The other thing you often see are month over month comparisons. These also are not very useful as we have a major seasonal element to our sales as you can tell by the 10-year average line for single family homes in Greenwich. The odds are very good that nearly every year in your lifetime, May sales will be higher than April sales and that November sales will be lower than October sales. Saying that sales are up or down from the prior month generally doesn’t tell you very much, because of this seasonality. If the change is different than the 10-year would predict, then something likely is happening. A good example are our July sales this year. This year and last July sales were higher than June sales and not just by a little.
If 2020 is the third year with higher July sales than June sales, we may well have a new sales curve and this may be to our detriment. If sales are moving to later in the year as they did this year and last year, then pundits are likely to see the market as weaker than it is by assuming second half sales will be similar to the first half of the year.
It’s going to be harder to say just how well the market is likely to do for the whole year by the end of the first half of the year. In 2007 56% of our sales were in the half of the year, while in 2018 it was only 46%. A 10% difference doesn’t seem like a lot, but it means that sales as a percentage of the whole are down 20% in the first half of 2018 compared to the whole year percentage in 2007.
|As of 9/30/2019||Inventory||Contracts||Last Mo. Solds||Last Mo Solds+ Contracts||YTD Solds||YTD+ Contracts||Mos Supply||Mos w/ Contracts||Last Mo. Annlzd|
Pundits are going to be making dire prediction in early July unless they adjust the 1st half sales up to account for our shift to second half sales. This will be particularly so as our sales over $5 million shifted to later in the year several years ago.
Months of Supply
Months of supply is also another very useful way to look at the overall market and at particular segments within the market. The good thing about this metric is that it includes not only sales, but the current inventory. It sets out how long it would take to sell our present inventory based on the prior sales rate. For an easy example if we had sold one house a month so far this year and you have 9 house in inventory then you have 9 months of supply.
You can also look at months of supply in different ways; such as only actual closed sales, sale and contracts, or just last month’s sales annualized. If I am representing the seller, I like to see each of these number decline. For example, our September closed sales numbers show that from $600K – 800K we have 8.3 months of supply, not particularly good for an under a million-dollar price range in Greenwich. That however includes the poor first half and the good third quarter.
When you add in contracts and assume they will all close in 45 days you get 8.1 months of supply. A little lower, but the buyers aren’t as active as we’d like given the inventory we have on the market. The good news is that when you annualize the four September sales in this price category you are down to 6 months of supply so as of September this part of the market is looking up even though inventory is up 50% over last year.
Days on Market
Months of supply sometimes gets confused with days on market. If you total up the number of days each house was on the market and take the average, you get days on market for the whole market. If each house came on the market and sold on average in 45 days, then you have 45 days on market (DOM). The lower the days on market, the hotter the market, as buyers snatch up houses within weeks of coming on the market.
If you had 45 days on market for the average time a listing was on the market you would have a very hot market. Right now, we have 141 days on market for our sold properties and 169 days on market for unsold listings. This is generally pattern as the well-priced, nice house sell quicker than the over-priced houses that sit on the market.
Curiously, the average days on market initially goes up in a hot market as houses that have been on the market for months and years are finally finding buyers. This is a number where the difference between the average and the median can be quite large. All you need is for a couple of houses that have been on the market for 1,000+ days to sell and the average days on market will jump while the median will barely budge. (BTW: In the prior paragraph I didn’t tell you whether I was using the arithmetic average or the median for the 141 DOM sold and the 169 DOM inventory. The better number is the median and that what these number area. The median is as affected by the long tail of days on market. The average DOM for solds is 238 days on market and the DOM for our inventory is 262 days. We have some people that have listed their houses for a really long time.)
Days on market varies significantly by price range. The lower the price range the lower the days on market. We also see the same effect in months of supply. The rough rule of thumb is that months of supply under 6 months are a seller’s market, but your get over $5 million and sellers start feeling good when months of supply drops below 12 or ever 18 months. (Sellers haven’t been feeling good for a few years now in those price ranges.)
So, if you are thinking about listing your house or considering buying, what numbers should you look at? Here’s a quick 5-point check list:
- Overall are sales and inventory up or down?
- How are sales and inventory changing in your area and your price range?
- What is the months of supply for sales, what about with contracts and for the prior month annualized?
- What is the difference in median sales DOM and inventory DOM? Is it getting better or worse?
- How is your market doing on a year over year basis and against the 10-year average?
For the seller, months of supply and days on market are significant factors in how aggressively you have to price your house. When these numbers are high you want to be at the lower end of price per square foot and have better staging than your competition. For buyers, you can afford to be more aggressive in your negotiating, particularly where you have other options that are satisfactory.
But numbers are just numbers, when it comes down to one-on-one sales negotiation. When you are negotiating you want to know the needs of the other persons, there personalities and their stress tolerance. In negotiations these are often more important than how soft or weak the market is. Knowing the numbers and knowing how to negotiate in the Greenwich market is what gives you the best result, a sense of control and can even make buying or selling a house fun.