Last time housing was at this level unemployment was 6%

Unemployment was 5.8% in July 2008 and 6.1% in August 2008.  Housing starts were 923,000 in July 2008 and 844,000 in August 2008.

Housing starts are now at 872,000.  And unemployment is 7.8%.  If you prefer other numbers like employment to population, or U-6, then things are much worse.

It’s not housing!  It’s NGDP, and it always has been.

With more NGDP we could be at 6% unemployment, even if we were just building 900,000 houses per month.



39 Responses to “Last time housing was at this level unemployment was 6%”

  1. Gravatar of Tommy Dorsett Tommy Dorsett
    17. October 2012 at 10:23



    Scott, agreed, BUT: if recovering housing helps to take pressure off V (remember velocity peaked in 2006 when home prices started falling) AND open-ended QE boosts M, we should get more NGDP. Even if its 5% pa instead of 4% pa, that would be meaningful. In short, positive shocks may help the Fed ease, just has negative shocks caused them to over-tighten. Bernanke has implied a higher tolerance for inflation if there is evidence of slack. Thus, they likely won’t lean against positive shocks and are actually leaning into them as inflation breakevens were higher when QE3 went off relative to QE2 and QE1.

  2. Gravatar of luis H Arroyo luis H Arroyo
    17. October 2012 at 10:50

    Well, I see some relation between homes and emploment, here:

  3. Gravatar of luis H Arroyo luis H Arroyo
    17. October 2012 at 10:58

    Or in this one, perhaps?

  4. Gravatar of Randy W Randy W
    17. October 2012 at 11:25

    Scott – agree with the point of the argument, but those are annualized numbers. So it’s it’s more like a twelfth of that number per month. Your post makes it sound like we started 900,000 homes in August.

  5. Gravatar of marcus nunes marcus nunes
    17. October 2012 at 11:35

    Very true. But Bernanke clings to the inflation target, and that´s a ‘spoiler’:

  6. Gravatar of Doug M Doug M
    17. October 2012 at 14:12

    And the economy was the picture of health in August of 2008.

    Once again, what is your point!

  7. Gravatar of Willitts Willitts
    17. October 2012 at 14:43

    Oi vey. Housing is rising fast, in percentage terms, from the lowest level in the past three decades. Real residential investment is at 1991 levels (which doesn’t even compensate for population growth). Although the contribution to GDP has been positive, the RI share has been small and fell sharply last quarter. Real GDP growth has slowed for the past two quarters.

    I know that you want to say, “If they listened to me….” but there is no evidence that looser monetary policy would induce more investment or consumption. We are risking.inflation with no associated benefits (not that I’m worried about inflation).

    Cheap money often ends up where you least want it. This is a time for risk taking, but everyone who wants to take on risky ventures has done so. Now we are just encouraging wanton risk taking.

  8. Gravatar of Steve Steve
    17. October 2012 at 15:38

    But we have structural unemployment!!!

    Too many oil workers in Lousiana and too few LEED certifiers in California. Too many teachers in Wisconsin, and too few roofers in Nevada. Soon to be too many pipefitters in North Dakota, and too few doctors in Texas.

    P.S., Being sacarstic if it isn’t obvious. People don’t know the difference between structural and frictional.

    The pattern here is that businesses piss and whine to the Fed whenever they experience margin compression, and they have stooges like Fisher, Bullard, Plosser, and Lacker to dissent over the “structural” problems facing their businesses.

    Now it’s the homebuilders facing margin deterioration. They don’t want to raise wages to lure back workers, but the are having trouble raising prices due to appraisal problems. Not to mention the great hyperinflation in lumber prices, probably because all the paper money printing is absorbing the lumber supply!

  9. Gravatar of ssumner ssumner
    17. October 2012 at 15:43

    Tommy, I hope you are right.

    Luis, Your graph shows housing plunging two years before employment.

    Randy, But the 2008 numbers are equally non-seasonalized–hence the comparison is fair.

  10. Gravatar of Tony Hansen Tony Hansen
    17. October 2012 at 15:48

    Scott, I am a long time reader of your blog & work, but first time commenter. I think you’re failing to recognise that 12 months prior to August 2008 Housing starts were 1,330,000, they declined 36.5% year on year to that point. Basically construction companies (and building product companies & the whole housing supply chain) were finishing a large backlog of building & then laying employees off. The reverse is true now (with the 45% YOY rise), there is a large an increasing backlog of new housing to be built and construction companies (and their supply chains) will need to add meaningful numbers of new personnel. If housing starts plateaued for 6 months at current levels, unemployment would most likely continue falling meaningfully. I agree that NGDP is more important, but housing really changes the face of unemployment very meaningfully – Tony

  11. Gravatar of Suvy Suvy
    17. October 2012 at 16:32

    I don’t think it’s NGDP that’s the problem. Aggregate demand in an economy is the spending in an economy. You can spend money on goods and services and assets. So it’s not just NGDP, it’s a lot of different things. I still don’t understand how people can say that debt doesn’t matter and assets don’t matter.

    I’m actually in the process of reading Ken Rogoff’s book and when I look through all of the data in that book; I still can’t understand how anyone can say debt and assets don’t matter.

    One way I look at it is by supposing that there is a period of leveraging up to buy assets under a period of low volatility(2002-2007). Of course, volatility is not constant and it’s bound to increase. When it does increase and asset prices fall, the entire economy has more liabilities than assets. Therefore, on a micro level, everyone is trying to deleverage all at once, which reduces the amount of money circulating in an economy.

    Saying that AD=NGDP and that debt doesn’t matter just seems blatantly wrong. Money can be spent on both goods and services and assets. Not only that, but the empirical evidence for endogenous money seems extremely clear(the ECB wrote a paper not too long ago on the fact that the money supply being endogenous).

    Again, I’m for targeting NGDP because it makes far more sense to target NGDP than inflation, but to say that NGDP solely determines AD seems not only wrong, but extremely misguided.

  12. Gravatar of Major_Freedom Major_Freedom
    17. October 2012 at 19:10

    NGDP doesn’t drive employment. NGDP is in competition with all other non-NGDP expenditures, including wage payments.

    “Demand for commodities is not demand for labor.” – John Stuart Mill.

    Spending out of wage payments is what (in part) finances NGDP. Employers and others who make productive expenditures did not expect a sudden drop in spending on final output. It was precisely the decline in productive expenditures that explains the drop in NGDP, not the other way around.

    If NGDP falls, then we have to find out why productive expenditures (such as wage payments) fell PRIOR, since those expenditures logically and temporally precede spending on final output.

    People do not suddenly and capriciously decrease their spending on final output. The market monetarist story that people expected a fall in NGDP, and that is what reduced their productive expenditures, as if this spending would not have fallen if only NGDP kept growing, is a weak rebuttal. It is weak because no investor or seller cares about NGDP. If an individual seller or employer expects their own revenues to rise by X% next year, then they won’t care if NGDP drops by half, or doubles.

    The solution to the recession puzzle resides in the area of savings and capital; in productive expenditures and asset valuations. It does not reside in final spending, which can theoretically be composed entirely consumption spending tomorrow, in which case there would be zero demand for labor, zero demand for capital goods, and a devastating depression. Market monetarists would not be able to explain why there is a depression even though NGDP spending is rising X%. This is definitely an unlikely scenario, but the point of it is to show the logic, which is clearly flawed, behind NGDP targeting theory.

    It is wrong to claim that if only the government printed and spent $200 billion more (or whatever) on weapons of war, that unemployment over the whole country, in all lines of industries, would have been 6%.


    NGDP targeting in the US will also make US firms less productive. If you can understand why should the central bank print and spend money to ensure that a single US bank always receives 5% annual revenue growth, would eventually end up weakening the quality of that firm’s operations, due to a reduction of market force discipline, so too should you understand that should the central bank print and spend money to ensure that 2, 3, 4 or all US firms together always receive 5% annual revenue growth, would eventually weaken the quality of that country’s operations. This is the long term effect of NGDP targeting that market monetarists don’t consider in their myopic zeal to remove the symptoms, but not the cause, for why NGDP fell 2008.

  13. Gravatar of Greg Ransom Greg Ransom
    17. October 2012 at 19:56

    Thanks, Tony, for calling Scott out on his continuing campaign of disinformation on “housing starts” and construction employment, etc.

    Note well: According to the Census Bureau, “it may take four months to establish an underlying trend for building permit authorizations, five months for total starts and six months for total completions”, so investors should look more closely at the forming patterns to see through often-volatile month to month results.

    Also, note well, the “housing starts” statistics provides “No differentiation between size and quality of homes being initiated, only the nominal amount.”

    And here has been a dramatic fluctuation in the size and quality of homes built across the recent boom and bust cycle.

    Read more:

  14. Gravatar of Morgan Warstler Morgan Warstler
    17. October 2012 at 20:12

    I have a dog in this fight, so I’ll just suggest people read this:

    Happy to have a slow footed Och out, but also 100% SURE this has been DRAMATICALLY affecting home prices, construction and starts.

    Every single public sale, in every distressed market, at every single court house in the market, the same 6-10 guys who work that court house, have ear pieces connected to cell phones and $500K in cashiers checks everyday bidding against each other. Nobody else wins anything.

    There are very few “flippers” – some but not many. And IF the structural stuff goes south, only tech, only hyper-efficient operations will be able to withstand a softening of rents.

  15. Gravatar of Bill Ellis Bill Ellis
    17. October 2012 at 20:27

    Talk about cherry picking.

  16. Gravatar of Bill Ellis Bill Ellis
    17. October 2012 at 20:30

    Ya know what the housing market needs ? Ya know what will create demand for all of those empty houses ?

    A big auction on the internet.


  17. Gravatar of Greg Ransom Greg Ransom
    17. October 2012 at 20:53

    Here is a chart of “housing starts” and “housing completions” across the last 4 decades:

    Here’s a chart of YoY percentage change in “construction spending” over the last 20 years:

    Here’s a chart of construction employment for the last 45 years:

    Here’s a chart showing how construction, mining & timber were by far the biggest boom-bust jobs sector when it comes to employment and unemployment over the last decade:

    Here is the construction employment collapse by specialty:

    Here is the construction unemployment rate compared to the overall unemployment rate:

    In California, over 42% of all construction jobs were lost in the bust:

    Wages have dropped almost 20% in the construction sector:

    Here is a chart of authorized building permits:

    Here is a chart of Case-Shiller price declines by city:

    Here is a chart of the boom and bust in commercial and industrial loans:

    A substantial percentage of the long-term unemployed come out of the construction and related industries. I haven’t yet found a graph depicting the statistics and estimates.

  18. Gravatar of Greg Ransom Greg Ransom
    17. October 2012 at 21:10

    Can Scott tell us the time to completion numbers and percentages for different kinds of housing construction?

    Can Scott tell us how much construction labor takes place AFTER a house is “completed”? It’s actually rather a lot.

    Can Scott tell us the geographic and economic relations between new housing construction and new retail and business construction? In my new housing development, new housing construction was matched by new commercial, business and retail construction, which followed *after* the housing construction.

    Scott’s simplistic ‘construction/employment’ story is simplistic to the point of being far more false and far moe misleading than it is true or informative.

  19. Gravatar of Saturos Saturos
    17. October 2012 at 21:22

    Don’t laugh Bill, Morgan’s idea here is a good one:

  20. Gravatar of Ritwik Ritwik
    17. October 2012 at 21:28

    And last time NGDP was at this level, unemployment was?

    Oh that’s right. NGDP was never at this level. It’s at a record high.

  21. Gravatar of Saturos Saturos
    17. October 2012 at 21:36

    Ritwik – wrong level. You should be looking at the ratio of NGDP to hourly wages.

  22. Gravatar of Ritwik Ritwik
    18. October 2012 at 01:36


    Exactly. And we should be looking at the ratio of housing starts to hourly wages.

  23. Gravatar of ssumner ssumner
    18. October 2012 at 05:14

    Tony, earlier I looked at housing completions in a post, and got virtually identical results. So that explanation won’t work.

    Greg, I know you already saw my response to Tony in many other posts, so I assume you are being intentionally dishonest here in agreeing with him. You know full well I got roughly the same results with housing completions.

    And I won’t bother looking at all your graphs. If they show something other that “housing is cyclical” I might be interested. Not otherwise.

    Ritwik, Have you never heard of trend lines?

  24. Gravatar of Ritwik Ritwik
    18. October 2012 at 05:24

    And what’s the baseline trend growth for housing Scott? 0%?

  25. Gravatar of Greg Ransom Greg Ransom
    18. October 2012 at 09:19

    They don’t.

    “And I won’t bother looking at all your graphs. If they show something other that “housing is cyclical” I might be interested.”

    And, no, I haven’t read your earlier replies to Tony.

    A housing start does not equal a housing completion, and the larger houses and multi-unit housing take more than a year to complete.

  26. Gravatar of Greg Ransom Greg Ransom
    18. October 2012 at 09:33

    Scott, you simply won’t deal with the fact that your simplistic comments about “housing” and “construction” aren’t even a start at being persuasive to anyone who who’s looked at the data, understands the issues involved, and has thought extensively about the place of housing in the *wider* network of misdirected resources and labor and distorted prices and values across the whole system of relative prices, credit, and assets, and their changing liquidity and risk status.

    That’s the bottom line.

    You present radically incomplete and radically misleading and radically simplistic “data” that persuades no-one who knows the topic.

    That’s a problem — you can’t honestly and successfully marginalize an explanatory enterprise using such inapposite, unresponsive, misleading, and even essentially bogus arguments and “data”.

    Housing and housing construction and housing construction employment are part of the *network* of the economy, intertwined with other industries, eg timber, home decor, real estate, retail,construction materials, etc.

    It’s simply unpersuasive – bordering on dishonest – to pretend otherwise.

  27. Gravatar of ssumner ssumner
    19. October 2012 at 05:41

    Greg, Are you saying you haven’t read my posts with data on housing completions? If you haven’t read them, how were you able to comment on them? And if you have read them, why are you lying now?

    Answer the question or else I’m through responding to your comments.

    Ritwik, Trend growth in housing is roughly zero, why do you ask?

  28. Gravatar of Greg Ransom Greg Ransom
    19. October 2012 at 07:22

    Scott, I’ve read your posts, I haven’t read every comment and every reply to every comment in your comments sections.

    You referred explicitly to “replies to Tony” which I assumed were direct named replies to Tony in your comments section. I don’t recall reading any of these, although I may have.

    If you have explicitly mentioned “Tony” and replied to “Tony” in a post, I don’t recall ever seeing it. Its possible I forgot reading or simply missed that post.

    I’m not in the habit of lying or bullshitting people.

    Scott writes,

    “Greg, Are you saying you haven’t read my posts with data on housing completions? If you haven’t read them, how were you able to comment on them?”

  29. Gravatar of Doug M Doug M
    19. October 2012 at 07:32


    I am with you, Scott is picking the numbers that suit him and looking at housing its decline and its impact on the economy in the most superficial way.

    I have only been reading your blog for a couple of months. You say that you addressed some of these criticisms in earlier posts, but not since I have been reading. Perhaps you should either recap you thoughts, or link to your older posts.

  30. Gravatar of ssumner ssumner
    19. October 2012 at 09:55

    Doug, Here’s the post Greg claims he never read:

    Notice that he left 7 comments. Pretty strange to leave 7 comments at a post you never read, wouldn’t you say so? Do you still think Greg is telling the truth?

  31. Gravatar of ssumner ssumner
    19. October 2012 at 09:58

    Doug, I repeatedly swat down all the arguments, and they come back here repeating the same nonsense, hoping to impress new and impressionable commenters like you.

  32. Gravatar of Greg Ransom Greg Ransom
    19. October 2012 at 13:25

    Scott, help me out. I don’t see the name “Tony” anywhere. Maybe I need to look closer, but who has the time?

    I don’t know “Tony” from a tree in the forest (no offense to Tony, I just don’t remember the name on this blog.)

    These charges of “lying” seem completely unhinged from reality.

    I don’t recall a conversation between you and someone named “Tony”.

    This does not make me a liar. It’s simply the honest fact.

    To suggest otherwise is insane.

  33. Gravatar of ssumner ssumner
    19. October 2012 at 14:19

    Greg, Are you saying a different Greg Ransom wrote this:

    “Thanks, Tony, for calling Scott out on his continuing campaign of disinformation on “housing starts” and construction employment, etc.

    Note well: According to the Census Bureau, “it may take four months to establish an underlying trend for building permit authorizations, five months for total starts and six months for total completions”, so investors should look more closely at the forming patterns to see through often-volatile month to month results.”

    This is really getting amusing.

  34. Gravatar of Greg Ransom Greg Ransom
    19. October 2012 at 15:00

    Scott, All I can figure out is that we are misunderstanding each other in some absurd way.

    This is the first thing I can remember reading from or about “Tony” — a comment in this very post, not any earlier post or comment, especially from you, Scott.

    You have directly suggested otherwise, Scott, for reasons I can’t fathom. Or have misunderstood your plain words.


    Your charges of “lying” are unjustified and ungrounded and unsupported.

  35. Gravatar of Greg Ransom Greg Ransom
    19. October 2012 at 15:03

    To be clear, I I don’t see the name “Tony” in the link you provide, which is suppose to show I am lying when I say I don’t remember any previous post or reply by you, Scott, to remarks by Tony on housing.

    “Scott, help me out. I don’t see the name “Tony” anywhere. Maybe I need to look closer, but who has the time?”

  36. Gravatar of Greg Ransom Greg Ransom
    19. October 2012 at 23:35

    Scott, you choose to falsely accuse me of being “liar” rather than address the uncomfortable facts for your unsupported and untenable position.

    Address any of these:

    1. Housing size and quality changed across the boom and bust.

    2. Many jobs with home builders are *not* classified as ‘construction’ jobs, and employment in those jobs take place on a different time table than the start and completion of a pn individual house. Eg my neighbor in 2010 lost his computer programming job with a major California home builder.

    3. Extensive construction employment and home improvement employment takes place after official house ‘completion’, for we’ll more than a year after completion.

    4. New houses and new developments and new home purchases are just the start of a train of further economic activities, including construction activities, eg gas stations, grocery stores, retail stores, dry clearness, restaraunts, etc.

    5. House construction booms and busts are directly tied to booms and busts in a whole train of related industries and sectors, eg forestry, transportation, construction equipment, and an endless number of others sectors and industries,

    I don’t expect you to seriously engage any argument or fact or data series or consideration on this matter.

    I await instead your newest and most creative effort to smear me as a liar.

  37. Gravatar of ssumner ssumner
    20. October 2012 at 06:24

    Greg, I don’t know how you can be so dense:

    1. You agree with Tony’s complaint that I never bother to consider housing completions, just starts.

    2. You leave 7 comments at a post where I show that housing completions give the same results as starts.

    3. So you know the housing completion complaint is FALSE.

    4. But you still cheer Tony on, even though you know he’s providing a false argument. I have considered completions, and show that you get the same result. So why do you tell Tony he is right? Do you just enjoy throwing mud?

  38. Gravatar of Greg Ransom Greg Ransom
    20. October 2012 at 07:59

    You assume wrong, Scott — ask a 2nd greater about the implications.

    Tony pointed to the 1,330,000 number and the 37% decline.

    And when are you going to address the role of he housing securities and housing value crash and its role in the collapse of shadow money and the crash in liquidity and the increase in money cash balances demand and the crisis in the mismatch between short term debt and long term debt brought on by thiis?

    It’s punting, punting, punting, strawmen, strawmen, ignoring of the causal mechanism of the rival you attmept to belittle and head in the sand all of the time on all of these issues.

    This does not instill confidence.

    Multiple causation is the rule and the nature of the game when dealing with complex phenomena — you constantly imply that this is an either / or game of exclusive explanatory alternative, when the rival you belittle *includes* you own mechanism as a functional part.

    But out of the other side of your mouth, you acknowlege a role for a distorted network of relative prices and for a role for malinvestment in housing and transportation and related sectors. The problem seems to be that you are incapable of dealing with how your own explanatory mechanism is bound up with the replace price, malinvestment, asset crash, shadow money crash, etc.

    Or is that purposeful?

  39. Gravatar of RebelEconomist RebelEconomist
    21. October 2012 at 00:58

    This is an absurd argument! The present downturn has much deeper roots than housing or the reaction of monetary policy in 2008. The fundamental problem was that risky asset prices – some of which were housing related, from houses themselves to CDOs including mortgages – were allowed over a long time to get out of line, because the authorities dared not do anything that would end the party, especially when the influence of global competition was beginning to make the background more difficult. Remember “irrational exuberance”? That was in 1996! Extending the boom into housing – ie spreading it to even more people – made the eventual bust bigger. Housing was merely the last straw. From then on, no ordinary monetary policy (ie short of an inflationary reset of asset prices) was going to do anything other than shape the bust.

Leave a Reply