Home > RWER > RWER issue 55: Merijn Knibbe

RWER issue 55: Merijn Knibbe

Why did Dutch economists get it so wrong?
Merijn Knibbe   [Wageningen University, Netherlands]

Abstract
               
As late as February 2010, at the time when it already had exploded, Dutch economists denied the existence of a ‘housing bubble’ in the Netherlands. The reasons for this denial seem to be an unwarranted trust in formalized economic models as well as econometric estimates, the neglect of basic historical, comparative and statistical information as well as a curious lack of knowledge about ‘sate of the art’ ideas and models, let alone ‘heterodox’ ideas. This paper examines the failure of Dutch economists as a means of answering two transnational and ultimately theoretical questions:

  1. Is it possible to develop a kind of analysis which enables us to identify housing bubbles in an earlier phase, and
  2. Do economists when looking for housing bubbles look at the right variables and in the right way?

 1. Introduction
 Is there a housing bubble when, as happened in the Netherlands between 1986 and 2007(data from Özdemir and De Ward, 2005 and Centraal Bureau voor de Statistiek (CBS), http://statline.cbs.nl/statweb/?LA=nl; http://www.woningmarktcijfers.nl):

  • loan-to-value ratio’s rise to unprecedented heights;
  • loan-to-income ratio’s rise to unprecedented heights;
  • mortgage debt rises from about 100 billion euro in 1993 to over 600 billion euro in 2009 – and continuous to increase up to January 2010. November 2010 saw the first drop in decades;
  • real house prices rise about 150% in 21 years (1986-2007);
  • a fast increasing share of new mortgages consists of ‘interest-only’ mortgages or even ‘top mortgages’ of up to 125%  of house value and even higher;
  • 2009 mortgage debt per household is the highest in the world;
  • 2009 housing costs are the highest ever and the highest in Europe;
  • real house prices fall 8% in two years and continue to fall (October 2010);
  • the number of transactions on the housing market falls about 40% compared with 2006 and continues to fall. October 2010: was minus 12% compared with 2009;
  • cities like Amsterdam and Eindhoven run into major problems as they can’t sell land-with-a-building-permits anymore – and have to introduce draconian cuts in their infrastructure budgets;
  • ‘Theoretical selling time’ (number of houses for sale divided by average sales per month) increases to 48 months for more expensive houses and to 24 months for median priced houses;
  • construction output falls 20% in a year;

            Meanwhile nothing of the kind happens in neighboring Germany? According to Dutch economists there is and was, in spite of all these disturbing signs, no housing bubble in the Netherlands.

You may read the whole paper at: http://www.paecon.net/PAEReview/issue55/Knibbe55.pdf

  1. December 22, 2010 at 11:10 am

    44 comments (as of December 22) on this paper have been posted at http://sargasso.nl/archief/2010/12/18/waarom-zaten-nederlandse-economen-er-zo-naast/ All but one of them is in Dutch.

  2. Vast Goed
    March 23, 2011 at 9:38 am

    Leuk artikel (helaas eerder gemist op Sargasso).

    Een mooie aanvulling op je artikel is het model van de werking van vastgoedmarkten van Wheaton & diPasqaule. Zie voor uitleg:

    Click to access Vol203.33_44.pdf

    En voor een toepassing:
    http://ideas.repec.org/p/hkm/wpaper/212008.html

  3. Merijn Knibbe
    March 23, 2011 at 12:33 pm

    Thank you.

    This seem to be interesting articles, and I will read them.

    They, however, confirm my criticism:

    1. Business economics teaches us that, when analysing the financial situation of a company, one should not just look at
    (a)the profit and loss account and prices (basically: the budget line) but also at (b)liquidity (do streams of incoming money match streams of outgoing money) and
    (c)the balance sheet.

    There is nothing special about that, it’s part and parcel of accounting 101 – but (neo-classical) economists just do not do it.

    2. And indeed, these articles only look at income and prices.

    3. In the long run, however, liquidity and balance sheets are important, too. When analysing balance sheets we should also not only look at equity but also at variables like the absolute magnitude of debt and debt/income ratios.

    4. A sudden awareness of the fragility of balance sheets might cause liquidity problems for households wanting to buy a house, this is by the way exactly which at this moment is happening in the Netherlands (just read the headlines of the Dutch newspapers!). And indeed, that is our very own ‘Minsky moment’ (though it does develop much slower than in the USA).

    Minsky is often portrayed as a heterodox thinker, but in fact he isn’t. His ideas are completely consistent with ordinary business economics and his advice is in fact pretty conservative: do not borrow too much. Not even when rising incomes and prices seem to promise that real estate prices will go up for ever.

    The Dutch government recently (yesterday, in fact) limited mortgages in the Netherlands to 110% of the value of the house, down from an (unofficial) 140%. In Switzerland, so I’m told, this is 60%.

    I do not blame the articles for using a partial model (all models are partial). But economists do have to become aware that an article is not complete without a discussion of the results of a model which includes a meaningfull investigation of the relation of the model with the real world – in these cases at least an investigation of balance sheet and liquidity consequences of increasing incomes and increasing real prices of houses.

  1. No trackbacks yet.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: