The Cromford Report – Daily Observations June 2017

    June 25 – When we want to get a overview of the state of the market, we like to use Days of Inventory as a guide and compare today`s number to that on the same date in years past.

    Including active listings in UCB and CCBS status, we currently have 90 days of inventory across all areas & types within the ARMLS database. This compares with 110 on the same date in both 2016 and 2015 and 133 in 2014. The market is definitely more favorable to sellers when this number is lower. Thus we conclude that we have a much hotter market overall than in 2014 through 2016. However in June 2013 there were only 83 days of inventory so that was a more frenetic market and appreciation was running at a faster rate. 2012 was hotter still at 78 days.

    We are within the normal range and those who think we are in another bubble should remember that in June 2005 we had only 37 days of inventory. I would suggest that fears of a bubble should be confined to times when the days of inventory measure drops below 60, as it did between July 2004 and September 2005. Having dropped significantly below 60, the rise above 60 in late September 2005 was a clear sign of the bubble bursting.

    These days we are gently dropping back to levels not seen since 2013 which indicates a very strong, but perfectly normal housing market.

    June 24 – Apologies for the scarcity of observations over the last 5 days. This was caused by a major power outage at my office at the same time as I relocated my family and pets to the UK. We escaped the 119 degree weather only to find the hottest June day in England in 40 years. However the English weather has returned to its cool norm now and power has been restored at my office in Mesa.

    I am writing this from a little village called Woodhouse Eaves in England where we are installed in an old vicarage built in 1837, next to a church and graveyard. This is considered relatively modern in England as most churches are far older. By contrast in 1837 the whole of Arizona was still part of Mexico. The difference in the concept of age is one of the most obvious cultural differences between the UK and the USA.

    Building techniques are also very different. In the Old Vicarage, the walls are of local stone and are about 2 feet thick. The roof is made of local Swithland slate which is sought after for its strength and longevity. Such materials tend to last several centuries before substantial maintenance is required.

    June 18 – Phoenix has been the initial location for two of the new so-called iBuyer operations, Opendoor and OfferPad. There is another start-up (Knock) with a similar business model, but they are initially focused on Atlanta.

    As of the end of March the 12-month total sales by these iBuyers in Greater Phoenix has reached 1,839. This is 1.5% of the total housing market of 121,379 units, a market which incidentally has grown by over 10% in the last 12 months. During the last 12 months the iBuyer share has split 21% / 79% in favor of Opendoor, though the OfferPad share has increased to 26% during the first 5 months of 2017 and 36% in May 2017. It seems that the gap between Opendoor and OfferPad is closing. Opendoor’s first sale was in November 2014, 14 months prior to OfferPad’s first sale, so they enjoyed first-mover advantage for quite some time.

    It also appears that the iBuyer market share has peaked. In February 2017, they reached a combined 2.37% of units closed, and this has fallen in the following 3 months to 1.93%, 1.53% and 0.95%.

    It is not altogether clear why they should be losing market share having gained it so quickly in 2016. Perhaps they are finding it more difficult to acquire homes in their target price range, which is towards the low end of the market where supply has been quickly disappearing. There are a number of other possible reasons, but rather than speculate I will try to find out and post further in this section of the Cromford Report if I am able to shed any light on the subject.

    June 17 – The monthly median sales price for all areas & types is on a tear at the moment, as you can see from this chart. From $230,000 in mid April it has raced to $245,000 in just 8 weeks. That is an increase of 6.5% in just 2 months.

    This is useful data if you are trying to persuade someone to make a move before prices get any higher. However April to June is the period each year when most of the progress is made. Last year we had $228,000 at the middle of June and we finished 2016 with $227,500. In 2015 we saw $214,900 in mid June and finished the year at $215,000

    So it would not be unusual if this year behaves like 2016 and 2015 and we may still be close to $245.000 when we reach the end of December. If this is NOT the case and the median sales price continues to make further progress, that would suggest that 2017 has greater price momentum than usual. The next few weeks will be especially interesting to watch for that reason.

    June 16 – Black Knight Financial Services continues to publish its Mortgage Monitor report, but it probably does not get as much attention as it used to when foreclosures were at their peak. In the USA as a whole, 4.9% of first position home loans were non-current at the end of April 2017, with 0.85% in some stage of foreclosure and 4.1% having missed at least one payment. At the peak of the foreclosure wave, 13.5% of first position home loans were non-current with 3.3% in foreclosure and 10.2% having missed at least one payment. The improvement in delinquency rates over the past 12 months was 8.7%

    Arizona has lower delinquency than the USA as a whole. We have 3.2% of first position home loans non-current with just 0.3% in foreclosure and 2.9% having missed at least one payment. This represents an 8% improvement over April 2016. Most states have improved over the past 12 months, but there are a handful of exceptions. These include North Dakota, South Dakota, Nebraska, Louisiana, Wyoming and Alaska. You will probably notice that the common thread among these states in a high dependence on the fossil fuel energy business – oil gas and coal.

    For the first time, Black Knight has included Puerto Rico and the US Virgin Islands among its list of states. Not looking good. Puerto Rico has far higher home loan delinquency than perennial top-ranking Mississippi. 17.3% of Puerto Rican home loans are non-current and 5.6% are in foreclosure. This reflects the dire state of the Puerto Rican economy. The US Virgin Islands (not to be confused with the British Virgin Islands) are not looking too brilliant either with 9.2% of loans non-current and 4.4% in foreclosure. It is just over a century since the USA purchased the islands from Denmark. Now that was a piece of real estate with a long time on market. It took 64 years between Denmark deciding to sell (following the abolition of slavery in 1848) and the USA agreeing to pay a sum of $25 million. The islands total 150 square miles so the cost per acre in 1916 was only $260.

    June 15 – It is time for our weekly look at the Cromford® Market Index for the single-family markets in the 17 largest cities (by dollar volume)

    We have 11 cities showing improving conditions for sellers and 6 deteriorating. Last week we predicted that there would be more green dots and fewer red one, and we are relieved to be proven correct.

    Weaker conditions are concentrated in the Southeast Valley – Chandler down 8%, Gilbert down 5% and especially Tempe – down 17%. Mesa and Queen Creek are the lone holdouts and are still improving.

    Paradise Valley had a good spring but is fading now that the triple digit temperatures are taking hold. It is the only one of the 17 cities not in the seller’s market zone over 110.

    Among the big improvers are Fountain Hills, Cave Creek, Buckeye and Glendale, with Scottsdale and Avondale not far behind.

    The West Valley has regained its mojo with all its cities showing improvement and its star player Avondale now breaking through 200 again.

    Phoenix itself is almost unchanged compared with last month.

    June 14 – The single-family luxury home market is starting to split into 2 segments – under $1.5 million which is looking much stronger than last year – and over $1.5 million which continues to be plagued by abundant supply and lower growth in demand.

    Price Range Active June 2015 Active June 2016 Active June 2017 2 Year Change Current Days of Inventory
    $500K-$600K 1,130 1,374 1,261 +12% 176
    $600K-$800K 1,215 1,419 1,441 +19% 239
    $800K-$1M 688 800 783 +14% 295
    $1M-$1.5M 680 751 738 +9% 358
    $1.5M-$2M 381 436 460 +21% 600
    $2M-$3M 377 369 391 +4% 728
    Over $3M 272 302 289 +6% 879
    All the above 4,743 5,451 5,363 +13% 274

    Overall, we have 13% more supply than 2 years ago, so there is still plenty of choice for buyers. However, the annual sales rate for homes over $500,000 has increased from 5,264 to 7,132, a rise of 35%. So here we can see that the growth in demand is faster than the growth in supply and after a weak period since peaking in mid-summer 2015, prices in most luxury areas are on the rise again, particularly for homes under $1.5 million. It is above $1.5 million where the market starts to change for the worse.

    We can see that inventory is over 600 days from homes priced above $1.5 million, and here annual sales have risen from 547 to 596, an increase of only 9%, while supply is up by from 1,030 to 1,140, a rise of 11%. Now we see a problem. When supply increases at a faster rate than demand, sellers are at a disadvantage. Pricing will have a hard time making substantial upward progress until this condition changes.

    June 13 – On the thirteenth of each month we publish the small city snapshots for the following:

    • Carefree
    • Coolidge
    • Desert Hills
    • Eloy
    • Florence
    • New River
    • Rio Verde
    • Tonopah
    • Waddell
    • Wickenburg
    • Wittmann
    • Youngtown

    Two of these barely fit our small city category these days. Florence has grown in size to justify being promoted to one of our secondary cities, while Desert Hills is really now just a suburb of Phoenix and is slowly disappearing into oblivion from a statistical point of view.

    Being small these don’t get a lot of attention, so I though we should review the more interesting things happening to some of the single-family markets in these small cities.

    • Carefree – 376 days of inventory, -5.3% appreciation in annual $/SF, -2.9% annual increase in annual median, contract ratio 16.1 – expensive and over-supplied with falling sales rate and unusually low listing success rate (46%). One of the most difficult places to be a seller, hence prices are falling
    • Coolidge – 55 days of inventory, 8.2% appreciation in annual $/SF, 9.0% annual increase in annual median, contract ratio 107.4 – very affordable and appreciating quickly – only 1.2 months of supply – underrated as investment opportunity
    • Eloy – 139 days of inventory, 7.0% appreciation in annual $/SF, 42.9% annual increase in annual median, contract ratio 36.1 – a relatively cool market, rapidly moving upmarket due to the growth of Robson Ranch
    • Florence – 93 days of inventory, 10.5% appreciation in annual $/SF, 10.9% annual increase in annual median, contract ratio 73.2 – a hot, affordable and expanding market with rapid appreciation
    • Rio Verde – 295 days of inventory, 8.8% appreciation in annual $/SF, 1.3% annual increase in annual median, contract ratio 18.7 – excess supply and little growth in sales means stable pricing
    • Tonopah – 105 days in inventory, -4.0% appreciation in annual $/SF, 4.3% annual increase in annual median, contract ratio 25.0 – a cool rural market, with volatile numbers due to low sales rate (45 a year)
    • Wickenburg – 275 days of inventory, 7.7% appreciation in annual $/SF, 3.1% annual increase in annual median, contract ratio 19.5 – a very cool and stable rural market, lower success rate than average (60%)
    • Wittmann – 157 days of inventory, -0.3% appreciation in annual $/SF, 9.8% annual increase in annual median, contract ratio 35.7 – a relatively cool and stable market
    • Youngtown – 21 days of inventory, 12.5% appreciation in annual $/SF, -1.1% annual increase in annual median, contract ratio 260.0 – only 0.4 months of supply, tiny area with extremely low supply and low prices

    June 12 – Today I would like to make a broad comparison of the different regions within the Greater Phoenix area and compare how they have changed over the last year and since last month. The two key measures I am going to use are days of inventory and quarterly average price per sq. ft. We will focus on the single-family market since the condo/townhouse market is very small in a couple of these broader areas.

    Region Days of Inventory June 2016 Days of Inventory May 2017 Days of Inventory June 2017 Annual Change Quarterly Average $/SF May 2016 Quarterly Average $/SF May 2017 Annual Change
    Phoenix & North Valley 93 87 83 -10.7% $147.21 $153.95 +4.6%
    Northeast Valley 225 213 191 -15.1% $235.44 $249.03 +5.8%
    Southeast Valley 92 73 73 -20.7% $137.98 $144.98 +5.1%
    West Valley 81 80 74 -8.6% $114.96 $123.56 +7.5%
    Pinal County 119 88 82 -31.1% $95.67 $102.04 +6.7%

    Note that Pinal County has shown the largest decline in days of inventory and hence the biggest swing in favor of sellers. The West Valley has seen the smallest swing, but it is still in favor of sellers and it was already a strong seller’s market in June 2016.

    The Southeast Valley has seen the second biggest swing in favor of sellers, but note that it is the only area that did not improve for sellers between May 2017 and June 2017.

    Days of inventory is a leading indicator of the market whereas quarterly average $/SF is very much a trailing indicator.

    The Northeast Valley has improved its pricing noticeably over the last 2 months, but we caution that the “third quarter effect”, where pricing weakens between June and September, is more pronounced for the luxury market than for other price ranges.

    June 10 – We seem to have reached the minimum level of new foreclosures in Maricopa County. For the past 11 months, the 90-day average for the number of new Notices of Trustee Sale has been at 20 per day (about 29 per working day). This is the lowest level since 2000 but there is currently no sign of it going any lower. I guess we could call this the “background level” of new foreclosure notices that will be filed even at the best of times. You can see this flat-lining in the chart here.

    We not not reached a minimum in terms of completed foreclosures (Trustee Deeds). The current 90 day average across Maricopa County is 6 per day, down from 9 per day this time last year and still on a downward trend. We still have a long way to go to reach the minimum of 1 per day that we saw in 2006. This is because at the peak of the bubble, investors would ensure that any home that entered the foreclosure process would get plenty of purchase offers before it actually went to trustee sale. In those days negative equity was yet to be discovered as “a thing”.

    June 9 – Let us take another look at the Cromford® Market Index for the single-family markets in the 17 largest cities by dollar volume:

    Here we see a slight majority of cities improving for sellers, with strong gains for Fountain Hills, Buckeye, Peoria, Glendale and Scottsdale.

    8 out of the 17 show deteriorating conditions, though there are no cities with CMI below 100. Tempe remains on a weakening trend, while Paradise Valley and Chandler are faltering.

    Geographically speaking, all of the East Valley cities lost some ground over the past month, while the Northeast Valley is looking better than it has for some time (except for Paradise Valley). The West Valley is regaining more of its mojo especially in the big cities of Glendale and Peoria.

    The overall picture is still heavily biased in favor of sellers due to the weak supply situation and over the next 3 months that supply is going to get tighter still. With lenders’ underwriting standards continuing to get looser, it is our bet that the supply will drop a little faster than demand, which always fades during the third quarter. Nothing is certain in this world, but we expect to see a few more green dots and fewer red ones in the weeks ahead.

    June 8 – The Cromford® Market Index seems to be gaining at bit of its mojo back, having risen from 146.5 to 147.2 in the last 7 days. This is because the supply is now falling at a faster pace and we would expect it to continue to fall until the end of August. The Demand Index, meanwhile, has stopped its mild decline from the 106 level and is holding steady at 103 or thereabouts.

    The consequent outlook is a mild improvement for sellers over the next month thanks to fewer competing sellers. Buyers are expected to thin out over the next few months also, but to a lesser extent.

    So overall, the picture remains positive with no storm clouds on the horizon just yet. Appreciation will no doubt be interrupted by the third quarter effect, however, almost certainly resuming in the fourth quarter.

    June 7 – Not every listing closed through ARMLS has actually been marketed on ARMLS. We are seeing more and more listings that are added retrospectively, after the deed has recorded. This may be done for various reasons, such as:

    • the listing agent wants to create a comp for future use by appraisers or other agents
    • the listing agent wants to claim credit for the successful transaction in regional or national statistic tables collected by the National Association of REALTORS® or others
    • the listing agent wants to give the selling agent credit for the sale

    As this trend grows, the connection between listings under contract (i.e. pending, UCB or CCBS) and listings closed starts to break down. We get sales counts growing without a corresponding growth in listings under contract because these listings added after the fact appear to have never been in escrow. Many of you will have noticed that closed listings are up significantly from last year (11% year to date), but listings under contract are slightly lower (down 4% as of today).

    I think there are multiple effects at play here:

    • listings spend less time in escrow because deals are getting completed more quickly in a busy market
    • lenders are able to get loans approved more quickly because underwriting standards are loosening
    • more listings are closed as soon as they are created, never hitting the market

    There are at least 2 ways we can spot the last of these.

    • closed listings on the database that are completely missing from our daily log of active, UCB. TOM, CCBS or pending status listings
    • listings with 0 cumulative days on market or 0 agent days on market

    The first category above is up 91% from last year. The second category is up 30%. The second category is about 7 times more common than the first in 2017.

    There used to be a strong relationship between the under contract count and the monthly sales count the following month. The growing weakness in that relationship makes market trends a little more tricky to spot. At the national level we are seeing statements that demand is weakening because the pending listing counts are below expectations. We think this may be misguided. The way the market works now is generating more pocket listings and therefore artificially tamping down the pending listing count. Of course the pending listing count is already much lower than normal because so many listings under contract are being placed in UCB rather than pending status. Our estimate is that 67% of the UCB listings are not really available for sale, but would have been in pending status if not for the existence of Zillow.

    We need to adjust our expectations of the under contract count and reset the relationship between that count and the expected sales count the following month.

    June 6 – The preliminary numbers for Maricopa County recorded deeds are now available for May. Total sales volume was 11,274, the highest total since June 2006 and up 14% over May 2016. This includes single-family, condos and townhomes.

    The overall monthly median sales price hit $250,000 for the first time since November 2007. The median for new homes was $325,324 which is unspectacular but up 1.3% from May 2016. The median sales price for new homes is being held back by higher volumes of entry level homes. Prices for similar homes are increasing far more than 1.3%. The median sales price for re-sales was $239.000, up from $225,000 last year.

    New home unit volume increased by 26% from last year whereas re-sales volumes were up by 12%.

    June 5 – We have now added a new Tableau chart that shows the seasonal nature of new listings in Greater Phoenix. You can find it here.

    The biggest month for new listings is January (the slowest month for closed sales). The slowest month for new listings is December.

    For homes over $1 million, October is also a very busy month few new listings. July is very quiet, almost as slow as December.

    June 4 – We have added a new Tableau chart that shows the seasonal nature of sales in Greater Phoenix. You can find it here.

    It includes filters for Dwelling Type, Transaction Type, Price Range, County, City and ZIP Code. You can also adjust the years that are included (the default is 2001 through 2016. Be warned that if you include the current year it will distort the picture since not all months will be included.

    May is the overall peak month for closed sales with 9.827%. However June is close behind and May and April quickly follow. By far the weakest month is January with only 6.174% of closed sales.

    When we look exclusively at condo / townhouse sales, April is the peak month and May drops into second place. For mobile homes, March is the favorite month with April in second place.

    For homes over $1 million, June is the biggest month and the drop between June and July is huge. This helps explain why we see average price per sq. ft. fall every year during the third quarter. September is the weakest month for sales of homes over $1 million.

    June 3 – The S&P/Case-Shiller® Home Price Index® report for March came out on Tuesday and showed the following monthly changes for the 20 cities they focus on:

    1. Seattle 2.57%
    2. Charlotte 1.48%
    3. Minneapolis 1.41%
    4. Boston 1.36%
    5. Detroit 1.31%
    6. Denver 1.29%
    7. Dallas 1.23%
    8. Chicago 1.15%
    9. San Francisco 1.05%
    10. San Diego 1.02%
    11. Portland 1.02%
    12. Los Angeles 0.96%
    13. Atlanta 0.92%
    14. Washington DC 0.80%
    15. Las Vegas 0.78%
    16. New York 0.76%
    17. Phoenix 0.59%
    18. Miami 0.26%
    19. Cleveland -0.13%
    20. Tampa -1.01%

    Food for thought for those who think Phoenix is appreciating too fast. Not only did we place only 17th out of 20, we were well below the national average of 0.81%.

    The monthly drop in Tampa is unusual. I suggest we keep an eye on that one.

    The more important 1 year change table looks like this:

    1. Seattle 12.3%
    2. Portland 9.2%
    3. Dallas 8.6%
    4. Denver 8.4%
    5. Boston 7.7%
    6. Detroit 7.0%
    7. Minneapolis 6.8%
    8. Charlotte 6.7%
    9. San Diego 6.5%
    10. Las Vegas 6.4%
    11. Miami 6.0%
    12. Phoenix 5.6%
    13. Atlanta 5.5%
    14. Los Angeles 5.4%
    15. San Francisco 5.1%
    16. Tampa 5.1%
    17. Chicago 5.1%
    18. Cleveland 4.4%
    19. Washington DC 4.2%
    20. New York 4.1%

    Here we see Phoenix firmly in the middle of the pack and just below the national average of 5.7%.

    I see no sign of the collapse in the San Francisco market that some have been forecasting. Seattle, Portland, Dallas and Denver remain the front runners, as they have for quite some time.

    June 2 – Below is the regular weekly table where we show how the Cromford® Market Index has changed for the single-family markets in the 17 largest cities by dollar volume.

    We remain in a seller’s market overall with 8 cities showing improvement for sellers and 9 showing deterioration.

    The last five months have seen the smallest overall change in the Cromford® Market Index since we started measuring it. However there are a few bigger moves in the specific cities.

    The largest percentage change is in Tempe which has seen a sudden increase in active listings over the past 6 weeks. Although it is still a seller’s market in Tempe, it has slipped from 4th place in this table down to 11th since April 20. It is not yet clear why this increase in supply has happened since most parts of the valley have seen a decline in active listings. Active listings (excluding UCB and CCBS) are up significantly in all four Tempe ZIP codes (85281, 85282, 85282 and 85284). In particular, 85281 has the highest number of active single-family listings at the start of a month since April 2014. In the neighboring areas of Ahwatukee in Phoenix 85044 and 85048 we also see a rising supply trend. This is particularly surprising since there is little new single-family construction going on in these areas.

    Fountain Hills is the biggest percentage riser, benefiting from both a fall in supply and rising demand.

    Most other cities are little changed with the next largest moves by Peoria (up 5%) and Chandler (down 5%).

    In the Northwest, Paradise Valley has faded a bit after a good spring season while Scottsdale is showing some improvement with good sales volumes and a fall in supply.

    The Southeast Valley has lost steam with the CMI for Queen Creek, Gilbert, Chandler and Tempe all down. Mesa is the only large city in the area showing a small advance.

    Outside the large cities, we have to mention Arizona City which is currently showing a CMI of 279.7. Supply has crashed in Arizona City with only 30 single-family homes available without a contract. This is the lowest we have ever recorded. There were 91 as recently as December and the previous record low in 2005 was 32. The annual sales rate is now 247, its highest level since 2014. The annual average $/SF is currently rising by 17.2% a year and the monthly median sales price has reached $130,000, very low by most standards, but up more than 29% from $100,500 a year ago. Its peak monthly median sales price during the bubble years was $169,000 in April 2006.

    June 1 – Opendoor and OfferPad are the two most prominent iBuyers of homes in the USA and both of them started operation in Phoenix. They have now been joined by Knock in Atlanta. Opendoor now operates in Dallas and in Las Vegas too (though to a very limited extent in the latter). They first started buying homes in Phoenix in August 2014 and I think now is a good to time to study their home buying and home selling in more detail. We are doing this primarily using recorded deeds rather than ARMLS data, since many of their purchases take place outside of the MLS. However the vast majority of their sales occur through the MLS.

    First here is a chart showing the unit volume of purchases across Maricopa and Pinal Counties:


    © 2017 Cromford Associates LLC

    The data used to create the Cromford® Report is obtained from public records and obtained under license from the Arizona Regional Multiple Listing Service, Inc (ARMLS). Cromford Associates LLC and ARMLS expressly disclaim and make no representations or warranties of any kind, whether express, implied or statutory, as to the accuracy of the data used or the merchantability or fitness for any particular purpose.


    Thank you for the fabulous job you did in helping us find and purchase a home in Scottsdale.  I have been involved in the legal and personal end of real estate for more than 35 years, and you are by far the most conscientious, helpful and competent real estate agent I have ever run across.  In short, you’re the best. – Craig and Charlotte H. (clients of Vicki Kaplan)

    Contact Us Now

    Any questions, comments, or feedback