RPICPI User Group

last person joined: 3 months ago 

To foster co-operation and the exchange of information between the ONS, its advisory bodies and other users. Please see 'Aims of the RPI/CPI User Group' in the library for further details.

2 % breach and Housing. Comparing the CPI-H to the US index

  • 1.  2 % breach and Housing. Comparing the CPI-H to the US index

    Posted 23-03-2017 09:19

    The American CPI-U breached the  2 % inflation level in December.


    Housing was a decisive part of this 2.1 % monthly breach figure. The documentation with the release shows that if  'Shelter' was not part of the 'basket' the monthly figure would only be a compliant 1.3 %. (Most of the Shelter component is a representation of the rental value of owner-occupied housing)


    US house prices had peaked last June but some fall back was probably underway by December. Although the most recent CPI-U figure has gone further up to 2.7 %,  Shelter is no longer the driver. (Without Shelter the CPI-U would still now be 2.3 % anyway)


    Links to my sources below.


    This CPI-U is not a house price index, but it is clearly responsive to house price inflation as a significant part of consumer inflation. Such housing inflation was a major prompt for the interest rate rises 2016-2017 just as it was in 2006 - as illustrated in my old 2012 blog in the footer of this message.


    The fact that in the UK the CPI and the CPI-H both breached the 2 % this week with identical 2.3 % figures could not be clearer for how our UK house price inflation is not significant in the CPI-H.   


    The UK needs an index that can do for house price inflation what the US indices do. The CPI-H does not cut it.


    It does not work like Rental Equivalence does in the USA.



    US CPI data for December


    https://www.bls.gov/cpi/cpid1612.pdf


    National Association of Realtors house price affordability figures for last year


    https://www.nar.realtor/sites/default/files/reports/2017/hai-3-17/hai-01-2017-housing-affordability-index-2017-03-17.pdf



    PRICE MOVEMENTS DECEMBER 2016 The Consumer Price Index for All Urban Consumers (CPI-U) increased 0.3 percent in December on a seasonally adjusted basis, the

    The figure without shelter can be seen on P. 6




  • 2.  RE: 2 % breach and Housing. Comparing the CPI-H to the US index

    Posted 04-05-2017 17:19
    Hi Andrew

    I agree with you that the concept of Rental Equivalence works better as an inflation measure in the US than in the UK. One of our problems is the way that the numbers are "smoothed" I think. But with the caveat that it has its critics in the United States as well what do you think that they have done better than us?

    Regards

    Shaun Richards


  • 3.  RE: 2 % breach and Housing. Comparing the CPI-H to the US index

    Posted 05-05-2017 09:37

    Hi Shaun

    The most important reason why RE works for the US, is that the Fed has always acted to stop a house price bubble setting in.

    Hence rents become the main driver for house prices.

    The Fed is just better than the Bank of England. In the 20 years since the Bank became independent we have seen a house price explosion that can be seen from space. And a bank wreck that can also still be seen from space.

    Yes, I hear a lot of criticisms of the US systems too. To which I say they should come over here and see what we have got, then they would get it more in proportion.

    On specifically statistical issues, they use the concept of RE in 2 specifically different ways in the CPI indices and the PCE indices. And I have only ever been aware of problems in the PCE use, but they have been addressed.

    Did you see my email from last year on what I called the miss-translation of RE by the ONS ?

    ------------------------------
    AL

    how the US CPI actually worked to contain housing inflation

    http://www.localisewestmidlands.org.uk/2012/housing-reflation-and-housing-inflation-alternative-inflation/
    ------------------------------



  • 4.  RE: 2 % breach and Housing. Comparing the CPI-H to the US index

    Posted 08-05-2017 15:42

    Dear Shaun and Andrew:

    If we are speaking about a consumer price series being used as an inflation yardstick by a central bank, we should remember that the US Fed switched to the personal consumption expenditure price index as its main measure in February 2000. At no time during the US housing boom and bust was the CPI-U its inflation yardstick. If it had been it would have been of no help. By coincidence, February 2000 was the last month that the national Case-Shiller house price index had an annual inflation rate under 8.0% until June 2006. It peaked at 14.5% in August 2005. Over the same March 2000 to June 2006 interval the CPI-U equivalent rent index ranged between 1.9% (January 2004) and 4.5% (December 2001 to April 2002). Note that its inflation rate peaked over this interval just at a time when housing price inflation was accelerating. The equivalent rent series was insensitive both to the housing boom and the housing bust, so it would not, anymore than any other equivalent rent series, be an appropriate indicator for central bank use.

    Lately, house price inflation has picked up in the US. The Case-Shiller national HPI inflation rate, which was 4.2% in March 2015, was at 5.8% in February 2017, with Zillow Real Estate Research forecasting a 5.9% inflation rate for March 2017. Over the same period, the owner's equivalent rent index has gone from 2.7% to 3.5%, pulling up the CPI-U inflation rate. It would be a mistake to make too much of this. No-one denies that rental markets and housing markets are linked or that they can both tighten at the same time, as now seems to be the case in the US. However an equivalent rent series is in no way a protector against a housing boom and bust, as the performance of the US series in the previous decade makes very clear.

    That being said, the ONS obviously needs an equivalent rent series at least for its HFCE deflator, and it might well be useful to look at the US methodologies for things that might be usefully adopted. I am not sure, for example, if the ONS estimates have an adjustment for ageing bias, like the rent and equivalent rent series for the US.

    Best regards,

    Andrew




  • 5.  RE: 2 % breach and Housing. Comparing the CPI-H to the US index

    Posted 08-05-2017 17:09

    I agree with very little of what Andrew Baldwin says about the US CPI.

    The US Fed did say in 2000 that it would use the PCE index as its yardstick.

    However if you look at the US Fed minutes in 2006, when the Fed pushed up interest rates and halted the house price problem in the US, they were clearly following both the PCE and CPI indices. And the shelter issues were playing a role in the CPI as I argued in my blog of 2012. (And as they were last year)

    Further, the CPI was the index clearly signalling a breach, in part because there were problems with the PCE that were requiring significant revisions.

    The Fed is not legally committed to any particular index, unlike the UK.

    As for using the Schiller figures, I prefer the National Association of Realtor figures which I referred to in my earlier post because they offer a link to incomes. I think this is essential and enlightening when you are looking for significant changes that could have a bearing on weights.



    ------------------------------
    AL

    how the US CPI actually worked to contain housing inflation

    http://www.localisewestmidlands.org.uk/2012/housing-reflation-and-housing-inflation-alternative-inflation/
    ------------------------------



  • 6.  RE: 2 % breach and Housing. Comparing the CPI-H to the US index

    Posted 20-09-2017 20:27


    The headline UK inflation rate was 2.7 % last month on the basis of the latest headline inflation index. It would be 2.9 % if they were still using the CPI index that was the main focus until March.

    There is reported to be some thought being given to interest rates edging up at some very vague point.

    After all a 2 % figure has been seen as the critical level in the advanced world for decades.

    Contrast this with what is happening in the US.

    The main inflation index (CPI-U) breached 2% last December, and interest rates were edged up. Now that inflation is 1.9 %.

    Housing and house prices were a main driver last December. Without shelter being in the then US figures inflation would have only been 1.3 %.

    Looking at the National Association of Realtor (NAR) figures, you can see how mortgage rates bobbed up after the Fed interest rate was raised. But in the Spring lenders managed to let mortgage rates slip back. And then house prices edged back up.

    https://www.nar.realtor/topics/housing-affordability-index

    Shelter, most of which is a representation of owner- occupation, is about 0.6 % of the current  1.9 % CPI-U figure.

    These NAR figures illustrate how the US inflation figures and interest rate adjustments keep house prices nearer to 3 times rather than 4 times average household incomes.

    The UK inflation indices do not work in this way, and were not designed to do so. And as the US Fed strives to get back to normal interest rates, and monetary policy, the UK is just not in the same world.

    And this is a problem for the UK that will probably dwarf most of the worst possible scenarios for any Brexit failure.



    ------------------------------
    AL

    how the US CPI actually worked to contain housing inflation

    http://www.localisewestmidlands.org.uk/2012/housing-reflation-and-housing-inflation-alternative-inflation/
    ------------------------------