Testing Temperatures 1

How the DWP got it so wrong


IDS and the DWP used temperatures for the ‘Temperature Link' policy from dataset which were published in 2003 by the Climatic Research Unit (CRU) of the University of East Anglia.  The dataset was researched by Dr. T. D. Mitchell, who had moved to Norwich to carry out the research for a Ph.D on the development of climate scenarios for subsequent use by researchers investigating the impacts of climate change.


The research took weather data from weather stations across the World covering the period 1961-1990.  It is a recognised resource for assessing climate change.


The figures contained in the dataset were never designed to be used in the way they have been used by the DWP.  They are 'fixed' absolute values, and bear no relation to annual or regional variations, or for that matter fluctuations in weather patterns year on year, or to climate change, the very subject the research was carried out to provide a baseline for measuring!


It is evident IDS and the DWP chose to ignore Dr. Mitchell’s published advice, in which he said:

“However, a number of caveats should be noted:

• The climate models shown here do not explicitly recognise the smaller countries within their grids.

• Since climate changes may depend on elevation, not just latitude and longitude, these models (which only represent elevation on scales of hundreds of kilometres) may misrepresent climate changes in smaller countries at high elevations.

• Where a country includes a number of different climatic regions (i.e. regions in which climate varies coherently), it is possible that the models may represent the changes accurately in each region, but that the average change over the country as a whole may not be physically meaningful.”


Ignoring the third of those caveats above, provides some of the proof that the DWP were so desperate to find something, one might say almost anything, which would allow the DWP to fulfil the policy Iain Duncan Smith was so determined to pursue.

In December 2012 the Met Office produced a Report for the DWP, in which the Met Office said:

‘The Department commissioned the Met Office to undertake this work following initial feasibility studies for the UK and France.’


Exactly one year later, in response to a Freedom of Information request which was trying to discover what had been done, the Met Office told us this:

‘DWP interpreted the data from this report to inform its policy for future eligibility to Winter Fuel Payments.’


We believe this statement, a statement which perhaps should not have been made by the Met Office, actually began to expose the duplicity of IDS and the DWP, and adds further proof to the use of 'fixed' absolute values of average winter temperatures as if they were the everyday norm for all countries throughout Europe.


It is important to note that the Met Office Report to the DWP has never been published by the DWP, and has never been placed in the House of Commons Library.  We got a copy one year after it was delivered to the DWP, in response to a Freedom of Information request.  The DWP have recently claimed that because they gave that copy in December 2013, that is tantamount to it being in the public domain.  Such a claim is rubbish  -   it was deliberately withheld from MPs, and there is no mention of it in the Explanatory Memorandum which accompanier the Statutory Instrument which changed UK Regulations.


The Met Office Report admits the document is based on “mean temperature estimates” (P2) and interpolation.  It admits - "... the approach is not without limitations” (P2), but it does not explain the impact of these limitations on the validity and reliability of its conclusion how, for example, French winter temperatures became marginally warmer than UK winter temperatures. The Report’s “bilinear interpolation method takes into account latitude, longitude and altitude” (P2).  It is mathematically impossible to use bilinear interpolation to apply three variables to estimate a value from 4 sets of data.  On P3, the report states that its bilinear interpolation calculations also take account of distances between the datasets and the centre of the territory for which a temperature is being calculated.


The Met Office Report, dated 12 December 2012, was deeply flawed and highly misleading.  It was used to mislead and deceive Parliament.  The Report was not produced independently of Government. It is neither impartial nor objective.  We quote from the Statutory Instrument 3270 Explanatory Memorandum: “DWP worked closely with the Met Office to define requirements and commission a report that shows the average winter temperature data for each EEA country and Switzerland.  DWP agreed that the Met Office would provide average (mean) temperatures covering the winter months (November-March)” (Para 7.4).


If the Met Office Report had been based on accurate data rather than on estimates, if its methodology was scientifically rigorous, and if the 'temperature link' had been truly impartial and objective, then it might have been possible to deny the right to export the WFP under EU Law.  And yet, the DWP, 10 Downing Street, the Office of the then Chairman of the Conservative Party, together with a number of Conservative MPs, all wrote that the decision to exclude certain countries was based upon ‘as robust a methodology as possible’!


New facts have emerged recently, which are relevant to the Statutory Instrument and its supporting Explanatory Memorandum, and which provide further evidence that the Secretary of State and the DWP deliberately set out to mislead Parliament:


a.   First - the SW of England is not the warmest part of the UK, as stated in the Explanatory Memorandum.  The Isles of Scilly are.  (5.6°C for SW England, compared to 8.7°C for the Scilly Isles)


b.   Second - almost all the temperatures used by the Met Office in their Report to the DWP dated 12 December 2012 were estimates, often extended by interpolations of estimates, since many figures were simply not available.  The estimation process is called spatial interpolation. In the CRU CL 2.0 data, the ratio of estimates to real data is 50 to 1.  In other words, 2% of the data comprised real temperature records and 98% of the temperatures were estimates.


c.   Third - and without question, the most serious discrepancy, is that the temperature figures used by the Met Office had a built-in margin of error of ± 2.0°C, which makes them meaningless, when you attempt to establish comparisons in the way the DWP have interpreted the datasets provided to them.


The ‘new’ average winter temperature which the DWP declared for France needs close examination.  The figure of 7.0°C as a temperature DOES NOT EXIST other than in the DWP.  For the DWP to constantly repeat: ‘the hotter French Overseas Departments had to be included in the calculation of the average winter temperature’ compounds the deliberate untruths told.


There is no reason whatsoever for the inclusion of those temperatures.  There is no objective evidence anywhere which can support such a decision.  For example, examination of the website of Météo-France, the official French Meteorological Agency, shows temperatures for France, meaning Metropolitan France and Corsica, and for Outre Mer, meaning France Overseas, which accords with the Constitution of the Fifth Republic of France.


In a Freedom of Information response, dated 13 December 2016, the DWP has finally admitted that it DID NOT consult either the French Government, nor Météo-France, the French National Meteorology Agency.  We are entitled to ask  -   why not?  Surely, if one Government is interfering in the internal affairs of another Government, it does require some common courtesy to be employed!


Iain Duncan Smith misled and deceived Parliament in several ways:  for example, he did not explain to Parliament that estimated winter temperatures are a very poor way to compare heating costs.  The Met Office uses a much better estimate of the impact of temperature variation on the cost of heating a home; called the Heating Degrees Days (HDD) formula.  He deceived Parliament by failing to explain the unreliable nature of the temperature link, and by failing to use the much more appropriate HDD formula.


“Ministers are concerned about paying Winter Fuel Payments in countries where the weather is generally warmer than the UK. Therefore, DWP has considered ways of ensuring that the payment is better targeted on people who live in countries with a cold climate” (SI 3270 Explanatory Memorandum, Para 4.2).


In claiming France is a ‘warmer’ country, Iain Duncan Smith failed properly to explain to Parliament, that he could only justify that claim by including French Tropical territories  -  four Outermost Regions of the European Union, Guadeloupe, Martinique, French Guiana, and La Réunion, in the Metropolitan France estimated winter temperature, and by extending the definition of winter to a wholly implausible 5 months  -  90 days to 151 days.


In the calculation to inflate the average winter temperature of Metropolitan France, the tiny islands of Guadeloupe, Martinique, and La Réunion, are just too small, and would have made no ‘discernible difference’ on the average of 7.0°C declared for France.  Collectively these three territories possess a land area equivalent to a mere eight-tenths of one percent of Metropolitan France.


However, when it comes to French Guiana, we find that the position is totally different.


In its Report to the DWP dated 12 December 2012, the Met Office published two temperatures for French Guiana.  One from the capital Cayenne, the other from a commune in the west of French Guiana, St Laurent du Maroni.


Staggeringly, the two areas listed represented less than 6% of the total land area of French Guiana.  The temperatures recorded from the two areas - 25.8°C for Cayenne, and 25.6°C for St Laurent du Maroni, were used as a critical part of the so-called ‘robust methodology’ evidence which identified average winter temperatures of that part of the South American Continent.


87% of French Guiana is forested.  The vast majority impenetrable Amazonian rain forest.  No temperatures at all were available from 94% of French Guiana  -  therefore how can any temperature used with such devastating results, be described as 'robust'?


The land area of French Guiana is equivalent to more than 15% of the land area of Metropolitan France.  France métropolitaine or la Métropole, is the part of France that is in Europe.  It can also be described as mainland France plus the island of Corsica.  Metropolitan France and Overseas France together form the French Republic.


Metropolitan France covers an area of 551,695 km² (213,011 sq. miles), whereas French Guiana covers a land area of 83,534 km2 (32,253 sq. miles) - or 15.14% of Metropolitan France.  A size which is almost the same as the new French Region of Nouvelle Aquitaine (see the map below).  Or, looked at another way, the land area of French Guiana is equivalent to more than one-third of the land area of the United Kingdom!






















The temperature used for French Guiana is at best, an estimate, and at worst, an interpolated guess.  It only relates to a tiny fraction of less than 6% of the land area of French Guiana.  The total land area of French Guiana alone thus exerted by far the greatest influence on the formula used to recalculate the average winter temperature of Metropolitan France, which inflated France from 4.9°C to 7.0°C, and deprived the 29,895 claimants in France from receiving the Winter Fuel Payment.


When you run the formula, which, according the Director of the CRU, would have been used to recalculate the average winter temperature for Metropolitan France, using only the data from French Guiana, you end up with a new temperature of 6.5°C - 1.9°C.  That is higher than the baseline for SW England, and thereby still sufficient to categorise France as ‘hot’!  Did the DWP advise Ministers, or for that matter, Parliament, of that result?  It did not!  Clear evidence that Parliament was misled.


We can safely conclude, that a South American Amazonian rain forest, with definitely dodgy estimated temperatures, located more than 7,000 kilometres across the Atlantic from Paris, was used as the critical part of the so-called ‘robust methodology’ used by the DWP to convince MPs to support the Statutory Instrument when it was introduced.  There is little point in blaming the Met Office, for they only obeyed the instruction given to them by their fee-paying client!


In the Statutory Instrument 2014 No.3270 Explanatory Memorandum, Iain Duncan Smith advised Parliament that average winter temperatures in some EEA countries are so much warmer than SW England; UK Pensioners living in these countries have no need of help to pay their heating bills, and can be denied the WFP without compromising their health and the purpose of the WFP.  In making these claims, with respect to France, he deceived Parliament.


Again in the SI 3270 Explanatory Memorandum Para 7.4 Iain Duncan Smith admitted that “climate data on which to base a country-level analysis was not available”. He then stated “DWP agreed that the Met Office would provide average (mean) temperatures covering the winter months (November-March)”.


Iain Duncan Smith misled Parliament by not explaining that the Met Office Report substituted estimated temperatures for the unavailable temperatures. Neither did he explain that the small number of actual temperatures available was augmented by a very large number of estimated temperatures. He did not explain that the Met Office stated in its report that the average temperatures are estimates.

We cannot accept that the action taken as a result of the 'Temperature Link' policy in any way meets the  EU requirements under Article 7 of Regulation 883/2004, the Coordination of Social Security Systems.


We believe that the so-called 'Temperature Link' policy is quite simply not proportionate to the legitimate objective of the national provisions.


Those who have been denied the WFP starting from the winter of 2015/2016, because they live in the six countries and one overseas territory declared 'hot', will receive this 'old-age risk benefit' automatically, if they return to live in the UK, because it is a 'universal' benefit, available to all who are eligible.  Or, they could make a fresh claim if they moved house to live in the Irish Republic or in Italy.


See the UK Government plans for Cold Weather

Click Here

Made with Adobe Muse