Housing Bubble 2019?


The most frequent question I got in 2019,  (aside from the ubiquitous query if America is going into recession) is if U.S. home prices are in a bubble which will eventually lead to a significant collapse.  The short answer is no.  One must be careful with the use of that scary word “bubble.” small ebbs and flows in pricing occur regularly and should not be confused with an actual economic bubble formation. I know we have a bubble in everything is in a bubble talk lately, but even that skeptical theory is only done by a few people who literally believe life on the planet is only continuing because of the Fed’s QE in this cycle. So let’s take a look at the housing bubble of the past period and find out if we do have a housing bubble about to pop.

The features of an economic bubble popping, include a 35%-65%  minimum decline in prices over a short duration. Eventually, prices return back to their pre-bubble phase.
An excellent historical example of a stock bubble that can be used for comparison is South Sea stock prices in the early 1700s.

Issac Newton:

“I can calculate the movement of stars, but not the madness of men.”


Also, a real bubble is a disconnect from economic fundamentals that needs speculation to push prices higher.

Issac Newton’s investment in the South Sea underwent a massive spike due to speculation and then collapsed in a very aggressive fashion.  Now that’s a bubble!

For those anticipating a housing bubble 2.0, I ask you to consider the difference between real home price gains from 2002-2005 vs. 2012-2019 as charted below. We don’t have the rapid adjusted to inflation price gains in this cycle like we did during the housing bubble years of 2002-2005. Since mortgage rates are lower in this cycle than the previous one, demand will be good enough to prevent an epic collapse.

From Doug Short

June Real home price Log

June real home prices YoY
Real home prices are high enough to impact demand when mortgage rates get 4.5% or higher, but we will not lose 1 million to 1.5 million existing home sales in a calendar year from mortgage buyers in this cycle. The peak existing home sales print during the housing bubble was 7,260,000, which crashed down to below 4 million in 2008. We don’t have that booming mortgage demand in this cycle to warrant speculation buying that could lead to a collapse in national home prices anytime soon.

Unlike the housing bubble years, “when anyone with a pulse could get a mortgage,” and, perhaps more significantly, anyone could cash out on their homes as well, in this cycle quality is more important than quantity.  This cycle has the best loan profiles I have seen in my 23 years in finance.  The majority of homeowners in this cycle can own the debt, and the debt is very vanilla, i.e., no exotic debt structures or old school stated income loans.

This is another reason why mortgage demand isn’t bubble booming, we lend to capacity now. A significant factor to think about is that when the next recession happens, we won’t see the mass scale loan defaults creating distress sale prices like we did in the previous cycle. For sure we will see defaults and foreclosures, but the highest risk owners would be late cycle low down payment homeowners who lost their jobs and don’t have the equity to sell their homes without a short sale. We have created an excellent nested equity position for homeowners who have bought houses from 2010-2016. In short, the scale rate of distress default sales will be low compared to the previous bubble bust.

From Calculated Risk
From Freddie Mac’s must follow person  @lenkiefer 

2019 cash out bubble
June Home Equity Housing
From Doug Short

June EHS

Real home prices don’t look like a bubble, existing home sales don’t look like a bubble and now let me show you another piece of we aren’t in a bubble housing cycle. Housing Starts have had their weakest economic period ever, and so has new home sales.

From Doug Short:

June Starts + Permits adj

In fact, in my recent housing starts article, I showed how housing permits have been stalled for 29 months now.


New home sales have had a similar chart as above, the weakest new home sales cycle ever recorded in U.S. history even with the longest job expansion, longest economic expansion and mortgage rates really below 5% for the most part since early 2011.

From Doug Short:
June new home sales adj pop

I would really get concerned about housing affordability if mortgage rates got back to 5.875% and this has been my thesis since 2013.  For those who have followed my work over the years, they would know since 2014 all my prediction articles have stated that the 10-year yield would stay in the range of 1.60%-3%. This has been the case for the most part outside the Brexit period in 2016 which yields went below 1.60% briefly, and last year we traded quickly over 3% as higher oil prices drove inflation expectations higher than it should have.

Even for 2019, I was forecasting this range and the 10-year yield going to below 2% if world trade got weaker.

“For 2019, I am sticking to my call that the 10-year yield will channel between 1.60% to 3%.  If world trade gets weaker, we could see the 10-year yield with a 1% handle again.”


In every housing cycle since 1981,  mortgage rates have fallen 2% or more.   In this cycle, the low range has been 3.25 to 4.25% For a 2% fall in rates in the next cycle, we would need mortgage rates to be 1.25% – 2.25 %.  This would likely mean the 10-year treasury yields would fall below zero.  Unlikely much? The trend says we could see negative rates here in America in the next recession.

The best case against negative rates here in the U.S. is that our younger demographic growth will keep a lid on any significant deflationary push downward.

However, as you can see from the chart below, the trend is your friend!

From Doug Short:

June 28 Fed Funds Yields

June 28 Fed intervention
June 28th Long bonds

As you can see from the data above, we don’t really have any speculation demand going on in housing even with low rates, and the home price gains have happened with some time duration, unlike the real bubble years of the past where it heated up from 2002-2005.

Will home prices come down again, absolutely and noticeably in coastal areas. Will we see national home prices fall like saw in the previous in this cycle, absolutely not when the next recession hits. You can see below how much supply was created during the housing bubble burst, and it would require a lot of distress supply, a lot of demand weakness now.

We also have to factor in that housing tenure has doubled in this cycle as well. Housing tenuring double is a crucial aspect. People are staying in their homes longer and longer the amount of supply coming onto the market during the next recession would be limited to distress sales and the natural supply of homes staying on the market longer. This should prevent the monthly supply spike we saw from 2006-2011 from happening again during the next recession, which would make calling this cycle housing bubble 2.0 not a valid theory. Remember, the homeowners in this cycle from 2010-2019 look about the exact opposite of the homeowners we saw from 2002-2005.
From the NAR:

Housing Tenure NAR 2018

You have seen a model on what a housing bubble looks like. We need to see real massive price gains in a short time, speculation demand, total demand growth, and vast production of homes because people believe we simply have no homes to buy ( wink wink). Can we have another bubble in housing, of course, but hopefully what I showed above gives you an idea of what to look for and why this cycle doesn’t look like the previous one.

Last thing! 

Typically you don’t see a bubble in the very same sector back to back, remember that the next time people say housing bubble 2.0. Also, we are running into a better demographic patch for housing in the next decade, (Ages 25-31 are the biggest in America today) which means we have decent replacement buyers coming into the next decade.

Logan Mohtashami is a financial writer and blogger covering the U.S. economy with a specialization in the housing market. Logan Mohtashami is a senior loan officer at AMC Lending Group,  which has been providing mortgage services for California residents since 1987. Logan also tracks all economic data  daily on his own facebook page https://www.facebook.com/Logan.Mohtashami

28 thoughts

  1. All I know is no matter what is said in the media, ones I start seeing 2 to 3 families trying to buy a house we are in a big feeling bubble and that is not sustainable! The market will crash and I’ll wait as long as it takes to buy that house when the market crashes at the very lost and try to get a 10-year loan on it.

    1. This thesis has no logic in any of the data that has been presented in the United States. However, the anti-American bears, typically the anti-central bank people have been talking about crashes in America since 1913. However, we always seem to come back. This new everything is a bubble that us that these American bears have more in common with Russian Trolls than American citizens.

  2. Well. You are wrong. Just like you were in the last bubble that popped in 2005. When interest rates rise, everything will come down plummeting.

    1. Ah, the typical Anti American bearish response on websites. I don’t have a model but prices are a bubble and we will crash back to pre bubble trends in a bubble like fashion. We all wonder why the paranoid anti central bank people promote doom and gloom on websites, they have no economic discipline to push any models what so ever but a faint hope of the collapse of America.

      1. What about foreign holdings and occupancy rates for the same? Perhaps those trends are ready to suddenly show a huge glut of inventory when global investment retreats.

      2. Foreign buyers make up less than 280,000 of the average 5,800,000 -6,100,000 total homes bought on average in the last few years in America. This sector is still primarily driven by an 80% mortgage buyer market place for existing homes and over 90% for new homes. A better thesis is that cash buyers are still 20% of the existing home sales market, if they fall back to the historical norms of 10% of sales the mortgage demand make up would need to take the purchase application data to the 21st century over the 300 indexes to stay at par with sales trends now. That, to me, is a better question to ask than foreign buyers, which is the smallest % segment of the market place.

  3. From Luis Matos: “You are wrong. Just like you were in the last bubble that popped in 2005.”

    Logan, Luis correct?

    I wasn’t following you prior to the peak of the 2000-2006 housing market bubble.

    Did your model call the top? Did you forecast that U.S. housing prices would fall 35% (Case-Shiller)?

    1. Take the entire thesis of this article and reverse it and what do you get.

      1. Adjusted to inflation massive price increases from 2002-2005
      2. Significant speculation on housing not only on exotic debt structures to buy homes but cash out loans as well.
      3. Significant production of homes, the 2002-2005 housing starts vs. population actually broke the downtrend we were having for decades in a cycle
      4. Purchase application data, existing home sales data, new home sales data all were deviating from 1996-2001 trends
      5. Prime age labor force growth was going to peak in 2007 and go down

      Pretty much 2002-2005 is the exact opposite from 2012-2019. This is why I stress people need to have models to show why something is happening and the counter to the current thesis. Housing economics is an excellent example of this. Prices will fall at some point when inventory channels look different, but it won’t be anything like what we saw from 2006-2011. In fact, 20%-30% of sales for existing homes for years were bought by cash, very hard to get significant distress supply from cash buyers.

  4. If you read the pricing statements I have written over the years, it points to a monthly supply heading toward six months as the red pricing flag; This is the same thing I said on Bloomberg financial this year in that interview. I even showed how homeowners that bought homes last year would be down year over year in California if they bought early in 2018 in a bidding war. Like my models for recessions, it’s all a process so when monthly supply data gets over five months, and sales trends are negative that is a red flag just like the six red flags I have for recessionary data lines. I even did this with new home sales last year, monthly supply broke over 6.5 months, sales trends were negative, and HMI data fell noticeable but stated previous year not to call for the top in housing because sales and starts are too low. The difference between speculation demand buying and traditional regular buying was the big difference between this cycle and the past. The importance of having models to show data lines is much more useful than saying all price levels are bubbles because it gives people data to look forward too. This is why the YoY real home price charts from 2002-2005 vs. 2012-2019 are the actual evidence why just the pricing cycle in this expansion is much different. Even now adjusting to inflation we have some negative year over year prints on pricing as sales are down and inventory is up in some major metro cities on the Case Shiller index. However, we don’t see rapid price declines and major monthly supply data spikes for existing homes, we see longer days on market and pricing weakness. There is a difference between that and a major bubble crash in prices which really means a 35%-65% crash in less than 18 months but in real terms we go back to 1996 levels.

  5. Logan, I am well aware of the differences between this housing market upcycle and the last.

    I know what bubbles are – and how they differ from run-of-the-mill booms.

    I have personally used real estate models myself for over 20 straight years – and still do – so I know about models.

    Did your model call the peak of the housing bubble in 2006?

    Yes or no?

    1. If you read what I just said, it showed it in 2005, look back at the monthly supply data, purchase application data, housing starts, and the existing home sales data. Just back test the model to 2005 and you can see.

      Remember, numbers are the closest thing we have to the handwriting of God. It can’t lie, take the model itself and back test it to 2005. Annual monthly supply spiked in 2006 to 6.5 months but 2005 is where purchase application data, housing starts, existing home sales data and monthly supply data were flashing the red flags. However, go back and look at the data for yourself and you can see what I am talking about it. This is why models matter a lot

  6. So your model DID call the 2006 housing market peak, correct?

    Yes or no?

    I’ve tested my real estate model all the back to 1982 to evaluate it’s accuracy.

    1. All you have to do is back test the model and it showed the red flag in 2005, 2006 was too late to issue first waive red flags. This is the third time I have said this and you haven’t done it yet. I don’t believe you have a similar model to mine.

    1. Most people don’t have economic models or adjusted to variable model factors to work with in each new cycle. It simply is too much work and discipline to tie economic model factors to all data points. However, for housing, mine was presented in this article and it takes 3 minutes to see that 2005 showed red flags. Like all my work recently it shows why people with models are tied to rolling data points not investment theories.

  7. My real estate timing model is based 100% on five key housing market indicators – and 0% on theories.

    I developed the model in the early 1990’s – and I haven’t changed it in 30 years.

    Like I do, you should consider posting a chart showing how well your model has worked over the years. Proof is everything.

  8. A real estate timing model that you can believe in is no different than a stock market timing model that you can believe in.

    There’s one chart with buy and sell signals on it.

    No words and lengthy explanations are needed – nor are multiple charts.

    Either the model works well – or it doesn’t – and a single chart tells all.

    Proof is everything.

    1. This is why economics and investment websites are two different worlds. I care about economics, math, facts and data and variable economic model changes within a cycle and you’re pushing an investment. I look at unemployment claims, LEI, currency changes, truck sales and you’re talking about timing the stock market. We are from two different worlds.

  9. A model worth following either works or it doesn’t.

    Kinda of like a pregnancy test that shows a woman if she is – or is not – pregnant.

    Multiple charts are not necessary – nor are any explanatory words.

  10. No, we’re both in the same exact world – the world of real estate timing models.

    Nor am I “pushing anything” other than good information. That’s what I sell.

    Logan, you’re a mortgage broker, correct?

    And yes, I care about economics too – in fact I have a degree in it (UCLA – 1969) – but that has nothing whatsoever to do with whether a real estate timing model works or it doesn’t.

    1. See you have a degree in economics and don’t have a economic model. Nothing wrong with that but your world is different than mine. You keep on talking about 2006 and my model shows 2005 as the red flag for housing. This is the difference between people who track all economic data everyday from people timing real estate investments, our worlds aren’t the same nor should it be.

Comments are closed.