Tuesday, October 27, 2015

Housing inflection point – August closing Update


The Case Shiller 20 city year over year (YoY)for August is one tenth of a percent lower than projected at 5.1%.  August grew at an Annual rate of 1% after 3 months of negative 2 % growth.  It is one month closer to falling off the cliff in October at 4.5% continuing down to 1% by March 2016. This projection is based on annual growth of 2%.  How can this be?  Because, year over year analysis only works well in a stable market. At an inflection point, it is very misleading due to uneven growth, seasonality, and lags in the data.  Under the present conditions, the year over year comparison over states the strength of housing.  However it will understate housing strength in the first quarter of 2016. The Case Shiller data is accurate, the way it is presented is misleading.  

The following chart illustrates the Case Shiller year over year growth and the Case Shiller index both seasonally adjusted and unadjusted.  Seasonal adjustment in house prices varies only slightly year to year.  In May and November the seasonally is one so the seasonal adjusted and unadjusted are equal in those months.  Starting in June the market is more active and prices increase until November. Prices are lower from December to April.  Prices are highest in August, averaging 2.2% above normal and lowest in February, when they are 2.4% below normal.  The House purchases are negotiated 1 to 2 months before th closing.  So December and January negotiations produce the prices seen in February closing data.  The unadjusted prices are lower in winter higher in summer.  The seasonal adjustment removes these swings so trend can be seen in the chart. The test of a seasonal adjustment is the ability to remove noise and this seasonal adjustment works very well.


Prices began growing again in 2012.  Price growth in 2012 and 2013 occurred relatively evenly throughout the year. In 2014 the first quarter GDP was negative.  The first quarter of 2015 was weak but better than 2014.  A house is both shelter and an investment.  Buyers concerned about investment value drive the market in the short run.  If the economy is weak the purchasers will be more conservative about purchase price. 

While August YoY increase is 5.0%, growth from Oct 2014 to March is 5.2%, the other months were weaker or negative.  It would take a major surge in prices to keep the YoY index from falling. In the chart the growth in price is assumed to go from an annual decline of 2% to zero growth in the projection. Comparing the August to August YoY growth marked by “A” to the March to March 2016 growth marked by “M”, Illustrates the impact of uneven growth.

As Yogi Bara might say, “forecasting is hard, especially about the past”.  August closing prices were negotiated in June and July.  Confidence was higher in these months so price growth turned positive.  However, they should not reach the level seen last year. Growth peaked in 2013 declining in 2014 and 2015. Furthermore, growth in the last three years ending in April has been very uneven and the cities that grew so sharply are reaching nose bleed levels.  The California cites grew by 47% and Florida and Nevada cities grew by 40%.  These cites represent one third of the Case Shiller 20 City Index.  New York, Washington, and Boston which represents another third grew by 12%.  I would anticipate that the growth in the next year would range for 2 to 5 percent. The forecasted Case Shiller YoY index through March 2016 is shown in the table below.



.

Even under a 5% growth the index declines and in this period understates the strength of housing.  Housing is driven by buyer confidence. Such a drop will dent this confidence and cause price declines, particularly in markets that saw high growth.

Friday, October 23, 2015

Case Shiller collapse with October closings


The Case Shiller year over year beginning with August is projected at 5.2%, 5.1% falling to 4.5% in October. It falls to 1% by March 2016. This projection is based on annual growth of 2%.  How can this be?  Because, year over year analysis only works well in a stable market. In a changing market it is very misleading due to uneven growth, seasonality and lags in the data.  Under the present conditions, the year over year comparison over states the strength of housing.  The Case Shiller data is accurate, the way it is presented is misleading.

The following chart illustrates the Case Shiller year over year growth and the Case Shiller 20 city index both seasonally adjusted and unadjusted.  Seasonal adjustment in house prices varies only slightly year to year.  In May and November the seasonally is one so the seasonal adjusted and unadjusted are equal in those months.  Starting in June the market is more active and prices increase until November. Prices are lower from December to April.  Prices are highest in August, averaging 2.2% above normal and lowest in February, when they are 2.4% below normal.  The House purchases are negotiated 1 to 2 months before the closing.  So December and January negotiations produce the prices seen in February closing data.  The unadjusted prices are lower in winter higher in summer.  The seasonal adjustment removes these swings so trend can be seen in the chart. The test of a seasonal adjustment is the ability to remove noise and this seasonal adjustment works very well.


Price growth in 2012 and 2013 occurred relatively evenly throughout the year. In 2014 the first quarter GDP was negative. The first quarter of 2015 was poor but better than 2014.  A house is both shelter and an investment.  Buyers concerned about investment value drive the market in the short run.  If the economy is weak the purchasers will be more conservative about purchase price.  The July 2014 to July 2015 increase is 4.9%.  But growth from Oct 2014 to March is 5.2%, the other months were weaker or negative.  It would take a major surge in prices to keep the YoY index from falling. In the chart the growth in price is assumed to go from an annual decline of 2% to zero growth. Comparing the July to July YoY growth marked by “J” to the March to March 2016 growth marked by “M”, Illustrates the impact of uneven growth.

As Yogi Bara might say, “forecasting is hard, especially about the past”.  August closing prices were negotiated in June and July.  Confidence was higher in these months so prices should start to improve from the negative 2% seen in the last 3 months.  However, they should not reach the level seen last year. Growth peaked in 2013 declining in 2014 and 2015. Furthermore, growth in the last three years ending in April has been very uneven and the cities that grew so sharply are reaching nose bleed levels.  The California cites grew by 47% and Florida and Nevada  cities grew by 40%. These cites represent one third of the Case Shiller 20 City Index.  New York, Washington and Boston which represents another third grew by 12%.  I would anticipate that the growth in the next year would range for 2 to 5 percent. The foretasted Case Shiller YoY index through March 2016 is shown in the table below.
Even under a 5% growth the index declines and in this period understates the strength of housing.

Wednesday, September 30, 2015

Case Schiller is negative for the last three months



How can that be, when the New York Times headline is “U.S. Home Prices Rose 5% in July”?  Both are correct.  The Chart below shows that the index is 5% above July 2014.  It also illustrates the surge in seasonally adjusted price that started in early 2012 and the current decline.  The last month of growth was April 2015.  The declines in average annual  compound growth ranged from 2.0 to 2.4%.  Negative growth also occurred in May through August of 2014 in reaction to weakness in the first quarter.  A similar pattern occurred this year.  It is possible that growth will resume in September or October.  However this year’s economic uncertainty could make buyers more conservative.  The longer monthly prices are negative, the more conservative buyers will become.

 


This chart demonstrates the wild swings in price growth that make forecasting so inaccurate.  While most buyers are looking for shelter, expectations of gain drives swings like this, not job growth or demographics.

The following table illustrates the average annual growth in the three years from April of 2012 to April 2015 in comparison to the average annual  compound growth rate in the last three months for the twenty Case Shiller cites.  These cities are grouped in order of growth during the bubble starting with Los Angeles which experienced the highest growth .  The market has changed but the grouping is still meaningful.
The growth of Los Angeles was 13% -Prices in the three years grew by 45%.  These prices are affordable only with low interest rates.  Prices in the last three months grew by only 1%.  Prices of the three California cities were flat compared to 14% in the three years earlier. The resort cities show a similar pattern. These two groups represent one third of the index.  Washington, New York, and Boston represents another third.  These cities grew relatively slowly (4 to 6 percent)  in the high growth years, but declined by 3.5% in the last three months. The low growth in these cities is a major change from the expansion that ended in 2006.  New York and Boston have Finance and Washington has government none which is growing.  Chicago experienced the sharpest decline in the last three months-13%. The Prospect of enormous tax increases will do that.   



New construction is focused on the high end of rhe market which is sensitive to price growth appreciation. The next three months will set the pricing pattern for 2016.  

Wednesday, August 26, 2015

House Price growth slows to 2% to 4 %


Stepping back and looking at price growth over the last 12 months compared to the two comparable prior periods illustrates the declining trend.  This should not be a surprise; the Case Shiller 20 city index for June 2015 is at 179.9 which is 27% above the level 3 years earlier. Price growth declined an average of 33% in each of the last two periods. Projecting the same decline will result in a price growth of 3.2% which brings house prices to 30% above June 2012. Thus, this simple projection indicates growth of 2 to 4 percent can be expected of the next 12 months.  The weakness in this projection is that it does not incorporate an increase in mortgage rate or change in the economy but these should not have a dramatic effect on the next 12 months. Higher interest rates and a perception that house prices are fully or over valued could begin to have a negative effect on prices in late 2016.

                              Source: Case Shiller 20 city Index seasonally adjusted

While growth in employment and income affect prices, it cannot explain this surge.  House price growth first appeared in March of 2012 following action to make government mortgage rules more attractive to buyers in November 2011. The price growth peaked a year later. A big factor in price growth, particularly in 2012 and 2014, was the belief that housing was undervalued.    The growth in price has not been smooth but it has been declining steadily.

The following chart illustrates that this growth is not smooth. The Case Shiller price index released August 25 is for sales recorded in June and negotiated in May-a lag of almost four months.  By that time people forget what happened when the prices were agreed upon. If the economy is weak, buyers will be more cautious about offering too much so the price growth turns negative.  The winter of 2014 was cold and the economy was hurt leading to negative price growth in March through August 2014 closings. In 2015, negative economic growth in the first quarter led to negative house price growth in April through June recorded sales.

                            Source: Case Shiller 20 city Index seasonally adjusted

Some have speculated that the current stock market turbulence may make buyers more cautious therefore lowering price growth.  That may happen, but most of the third quarter closings have occurred or are under contract. Weakness in price from the market turbulence will not appear until the October closings which appear at Christmas. Stock market turbulence will not lower house prices unless the turbulence is prolonged or deep.  The next three months should average positive price growth at an annual growth 2 to 4% mirroring the improved economy since the spring.

The Table below compares the growth in the first and second quarters by city. Growth in the first quarter was at an annual rate of 12.3% which dropped to negative 0.8% in the second quarter. The cities are grouped by volatility. California price growth has typically been above average but these cities grew only by 0.1%. The only city that experienced more growth in the second quarter was Las Vegas.  Portland, Denver, Phoenix, Detroit, Los Angeles and Seattle were the only other cities growing faster than a 2.5% annual rate in the second quarter.
                                             Source: Case Shiller 20 city Index seasonally adjusted


Cities Such as San Francisco which were growing at incredible rates saw negative rates.  Others like Dallas that are usually stable grew at double digit rates corrected in the second quarterly. These results demonstrate the nationwide effect of changing consumer perceptions of the safety of a housing investment. Yes people buy houses because they need shelter and theythey love the house, but investment concerns are sufficient to drive house price swings.

Wednesday, April 1, 2015

House prices are Driven by Economic Perceptions


House prices are driven by the perception of the economy.  Disappointing economic news makes buyers negotiate harder for lower prices. This has  probably happened already, but deals negotiated in March will not be reported until the week before the fourth of July.  Prices in 2014 grew at 4.5% down from two thirds from 2013 despite a better economy, because of the euphoria in 2013 about the housing. recovery. Price growth in more expensive markets declined more than lower priced areas. Government moves to lower down payments and mortgage costs will support higher prices but prices in 2015 may not continue to grow at twice the growth in salaries seen last year.

The 20 City index released in March, increased by 4.6% year over year for January closings, up from 4.4% for December which can be seem with the green line in the following chart.  Standard and Poor’s likes the year over year comparison because it masks the variability in the monthly data. David Blitzer of S&P says “monthly data reveal slowing increases”. Untrue, the last 4 months have been above 10% annual growth with the last 2 months at 11.4%. Blitzer knows that April through September 2014 saw negative growth.  He also understands that if the current 11.4% rate continues next two month’s year over year numbers will be 4.6% and 4.4% - flat and declining.  To keep the index creditable, he has to ward off expectations of growth that can’t be achieved, without explaining the monthly variability. Case Shiller data is excellent, It revels the variability in the data which is key to understanding pricing, but which S&P believes confuses the customer. 


   
Will double digit growth continue? That’s unlikely.  In the short term the price growth is driven by economic growth and the consumer expectations that it generates. In March the Michigan index of Consumer Expectations dropped 22% and the estimates of first quarter GDP are declining- the third weak first quarter in a row.  February closing data will probably be good because the deals were negotiated in January. But later months closings will be impacted by weather and declining expectations. April closings should definitely be down as they were in 2013 and 2012. Alas, April classing data will not be available until the week before the fourth of July. 


House prices in the short and medium term are not driven by increases in employment and the economy. The data demonstrates that market is driven by the potential for price appreciation   House prices fell for three years, before picking up in 2009 with the house purchase tax credit and other government programs. When the tax credit ended the prices fell in 2010. The market reacted to restoration of lower FHA loan limits in November 2011 and rebounded, peaking in May 2012.  Prices declined in June but the euphoria about price increases led investors and buyers to pile in, creating a peak in February 2013.  The investors started to leave at that point and rate of price increase declined.

The table below demonstrates the drop in prices growth in 2014 compared to 2013 for each of the 20 cities in the Case Shiller index. The index is grouped by cities that move is similar patterns. It is in order of the price growth in the expansion that peaked in 2006.  The cities that are most volatile dropped the most. The California cites grew at 20.6% in 2013 but only 6.3% in 2014, While the cities that grew the least in the bubble the stable cities grew 8.8% in 2013 and 6.4% in 2014. The California cites growth declined by 70% while the stable cities growth declined by 27%.  


In 2015 I expect the trends will continue that growth will be lower than 2014. I also expect that in the higher priced cits, growth will decline more than the lower priced cities. House affordability is a bigger problem in Los Angles than Dallas.  However, Government actions can increase housing prices. The lower down payments and lower FHA premiums will boost prices.  The question is can these actions offset the trends in 2015. They can’t in the long run.

Tuesday, February 24, 2015

House Prices grew 10% last quarter

House price Upsides
The following chart illustrates the seasonally adjusted Case Shiller 20 city index for December closings the bulk of which were negotiated in November.  Prices fell sharply in the April closings negotiated in March because of the last year’s winter storms which lowered personal income.  Prices began to improve with the August closings. The three month moving average is now at 10.5%, well above the year over year growth of 4.5%.  A quarter of data at 10.5% is highly significant.

Should the next three months grow at an annual rate of 10% (0.8% month) the year over year average for the next three months will be: 4.5%, 4.3%, and 4.1%.  This decline is because the negative months are starting to impact the year over year calculation. Is this why David Blitzer says, “The housing recovery is faltering”?  It is for his favorite index, but not in the real world.

The following table compares the price growth of the third quarter closings to those in the fourth quarter. The 20 cities are grouped by cities that move in similar patterns. The order of the cities is based on the growth from the start of the bubble in 1997 to the peak in 2006. 



Case Shiller Surge
Annual  Price Growth



Percentage Point Change

3Q 2014
4Qt 2014
3rd to 4th Quarter
Los Angeles
0%
12%
12%
San Diego
-1%
6%
7%
San Francisco
-2%
18%
20%
TL California
-1%
12%
13%
Miami
4%
12%
7%
Tampa
2%
18%
16%
Las Vegas
3%
5%
1%
TL Resorts
4%
12%
8%
Washington
-5%
8%
13%
New York
-2%
6%
7%
Boston
-2%
8%
10%
Seattle
2%
10%
8%
TL Large Cities
-2%
7%
9%
Phoenix
1%
6%
5%
Minneapolis
-7%
7%
14%
Chicago
-9%
7%
16%
Portland
1%
10%
8%
Atlanta
2%
21%
19%
Detroit
-9%
10%
19%
TL Medium Priced Cities
-4%
10%
14%
Denver
5%
15%
10%
Cleveland
3%
6%
3%
Charlotte
4%
8%
4%
Dallas
7%
12%
5%
TL Stable
5%
11%
6%
Composite-20
-2%
10%
12%

The growth of the 20 cities went from -2% annual rates to 10% in the fourth quarter a difference of 12 percentage points. The salmon colored cities had the greatest improvement quarter to quarter. San Francisco improved to 20 points, not surprising with the high tech successes in this city.  But the next two cities are unexpected. Atlanta and Detroit at a 19 point difference, both of these cities are low priced and were hit hard by the recession so catch up is to be expected.  Chicago at 16 points is similar. Tampa at 16 points is a lower priced city than Miami.  Resort cites typically grow rapidly in a recovery and both Miami and Los Vegas  saw good growth in the third quarter so forth quarter growth is good. 
  
New York’s is an expensive city with slow growth so a 7 point swing is reasonable. Cleveland, Charlotte and Dallas are low priced cities with stable prices, much like most cities outside the Case Shiller Index. 

As usual the outlook for house price appreciation varies significantly by region.

Monday, February 2, 2015

House price upside in 2015


“Home Price Gains Continue to slow”, is the headline on Standard and Poor’s January 27th Case Shiller report.   David Blitzer chairman of S&P’s Index Committee stated: “With the spring buying season and spring training season still a month or two away, the housing recovery is barely at first base. Prospects for a home run in 2015 aren’t good”.  Typically vague, but data on page 4 of the report contradicts Blitzer.  Seasonally adjusted prices for the 20 cities in the last 2 months grew at 0.7% per month, an annual rate of 8.7%, double the 4.3% annual number Blitzer relies on.  Is it churlish to point out that the latest data is for the November closings negotiated in September?  September is a long way from the spring market. Blitzer says that seasonally adjusted data is not as reliable as unadjusted Case Shiller data. He does not explain why.  Could it be that if the seasonally adjusted number forces the question of what happened - a question that neither David nor S&P are comfortable addressing?  If you do not know what happened last year, how can you accurately forecast the next year. You can’t. That is why forecasters have gotten house prices wrong for the last 4 years.

What is good about the Case Shiller?  The data on the 20 cities is the most accurate available. Analyzed carefully it is a rich source of information to understand price trends. Using this data   I project those prices in the Case Shiller 20 cities should increase by 5% and possibility more in 2015. These cities are expensive and most are high riced relative to the where they were at the start of the housing bubble. Nationwide the prices should increase 7%. Some of the lower priced areas will see growth of 10%. Even within the 20 cities, price growth will vary sharply.

The chart below illustrates the seasonally adjusted month over month growth of the Case Shiller 20 city index, along with a three month moving average.  
.This year’s results are more confusing than ever.  With the data for April 2014 closings, the index dropped sharply, turning negative with the May closings. Most of the sales that closed in April were negotiated in February. The spring market saw sharply lower prices, but sales that closed in August (negotiated in June) began an improvement.

Personal Income along with the inherent volatility of the city drives the price changes. Personal income growth in fourth quarter 2013 was only 0.1% above the prior quarter.  For the year, it only averaged 2.0%. The GNP was negative in the first quarter of 2014. In a volatile city like Los Angeles prices are heavily driven by investment considerations, in a city like Dallas people are really buying a house.   Over a period of years this relationship correlates at above 90%. In a single year particularly 2014, 70% is as good as it gets. 

The fourth quarter’s low income  and first quarter negative GDP was the result of winter storms but it carped off a very low growth year caused by tax increases, and the sequester.  By 2014 fiscal policy was positive and the economy was strong with the first three quarters having income growth of 3.6%.  But the 2014 spring market showed the effects of the low income growth as buyers were very cautious on house prices.

The following chart illustrates the result of a correlation of income and the house prices in the beginning of 2014 as a percent of those at the beginning of the expansion adjusted for inflation (a measure of inherent volatility).  Los Angeles grew at 4.6% annually in the first 11 months. The correlation predicted a growth of 5.4% - 0.7% above the actual result. There are several outliers. Miami, Tampa and Las Vegas grew well above the projected rates. Resort communities tend to shrink rapidly in a recession and grow more rapidly in the recovery.  Growth in Phoenix and Minneapolis grew slower than projected.   Historically growth in both cities is choppy.


The big weakness in this approach is the delay in personal income data by city which will not be available until November of the following year.

The following chart illustrates the outlook for the twenty cities. These cities are grouped by cities in the order of the growth in the expansion from 1997 to 2006.  So the cities are grouped by cities that perform similarly to changes in income.


Personal Income in Los Angeles grew at 2.2% in 2013. It is very volatile with real prices 200% above the start of the housing bubble. The average home is $645,000. By comparison Dallas is 122% above the start of the bubble and the average price is 36% of Los Angeles. Prices are lower because it is easy to build new homes while it is difficult to build in Los Angeles.  The high prices make Los Angeles at risk for declines as businesses move to lower cost areas. Dallas and Denver are also at risk because of the concentration of Oil industry.

The resorts are growing rapidly as they do in a growing economy. Despite Miami’s price level of 146% of the start of the expansion it still has room to grow. Tampa and Las Vegas are lower risk. The only large city with consistently high growth is Seattle, with its high technology industry.  New York has seen prices grow slower than inflation. Washington prices have surged as lobbyists flocked in, but its growth has slowed with budget pressures.  Atlanta and Charlotte were hard hit by the rescission but the prices are low and the income is recovering.   

 The average annual growth house price in the first 11 months of last year was 3.7% while the growth in the last 2 months is 8.7%.  I would work off the 11 month numbers and move it up to approximately 5% for the Case Shiller cities.  Obviously there are large uncertainties. The data in the next few months may lead me revise the number, most likely upward. With more work it is possible to forecast individual cities with reasonable accuracy.  


The Case Shiller cities are expensive and well above the level at the start of the housing bubble. Lower priced areas will see greater appreciation. In the first three quarters of 2014 the FHFA nationwide expanded price index increased 6.2% vs. 3.4% for the twenty cities. Case Shiller has a National index but that only covers selected urban areas. The data is good but it is not representative of the nation.  The nationwide price increase for 2015 should be about 7% with some areas outside the Case Shiller growing at 10%.