“…if on appeal…it appears to the tribunal that the decision should be varied in a particular manner, the decision shall be varied in that manner, but otherwise shall stand good.”

16. We were referred to the following cases:

Mrs Daphne Gutteridge – SpC 534 [2006] STC (SCD) 315

Mrs McGough – Social Security Commissioner – CP/3593/2006 R(P) 1/08

Thomas Joseph Beamish v HMRC SC 3009 2009

Philip Langley Rose v HMRC SpC 574 [2007] STC (SCD) 129

Background facts

History of decision / appeal

17. On 31 May 2011 HMRC issued a decision under s8 SSC(ToF)A 1999 setting out the contributions the appellant had paid and the period in which he had paid them as follows:

Contribution Year	Class 2 (self-employed)	Class 3 (non-employed)
1969-1970	NIL	NIL
1970-1971	NIL	NIL
1971-1972	9	1
1972-1973	52	NIL
1973-1974	52	NIL
1974-1975	57	NIL

Tax year	Class 1 (employed earner)	Class 2 (self-employed)	Class 3 Non-employed person
1975-76		53
1976-77		52
1979-1980		11
1980-81		26
1997-1998	£389.56
1998-1999	£253.12
2004-2005	£433.45
2005-2006	£230.15

18. The decision did not show any entries preceding 1969-70 and while it set out the following years in the above table (1977-78 to 1978-79, 1981-82 to 1996-97, 1999-00 to 2003-4 and 2006-07 to 2008-09) it did not show any contributions for these years.

19. The appellant appealed against the decision on 13 June 2011 on the basis that:

(1) there were years outside of the decision (1962 to 1969) where there were contributions and/or credits which needed to be taken into account.

(2) For the years included in the decision where there were less than full contributions and/or credits, there were further contributions and/or credits that had not been taken into account.

20. The contributions record is of significance to the appellant because it affects the amount of retirement pension he is entitled to.

21. The appellant was born on 16 June 1944 in Sweden and grew up there. He moved to London in the early 1960s to pursue a career as a photographer and to break into the music business. It is not disputed that the appellant was in the UK prior to 1969 but what is in dispute is whether any national insurance contributions were paid in that period. The appellant reached pensionable age on 16 June 2009.

22. On the one hand we had oral evidence from the appellant and other evidence in the form of published material on the appellant from which we able to make certain findings of fact (see [37] to [51] below). On the other, we had documentary evidence presented to us by HMRC of the appellant’s contribution record held by HMRC, the procedures in place for ensuring the accuracy of such records and the particular searches that had been made by HMRC for further contributions that may have been paid.

National insurance stamps and payments

23. The Tribunal was given the following background from HMRC in relation to the system for registering for national insurance, and for paying and recording contributions. While the appellant did not dispute that this background was informative as to the procedures which were in place and ought to have been followed it was disputed whether these procedures had in fact been applied to the appellant and his contribution record.

24. Registration for National Insurance for a person aged 18 or over was made by a person attending an office of the Department where they were given a form CF8.

25. From the information entered onto the CF8 a permanent record of National Insurance, known as form RF1, was set up for the person. Each RF1 contained a unique National Insurance number which was transcribed to a contribution card.

26. The card was given to the applicant and contribution stamps were affixed to the card each week.

27. If the person was employed they gave the card to the employer who purchased a National Insurance stamp from the Post Office and affixed it to the card. The cost of the employee’s share of the stamp was deducted from their wages and the employer paid their share of the stamp.

28. If the person was self-employed or non-employed they were responsible for the cost of and the purchase of the stamp and for the safe keeping of their card. They were also responsible for exchanging an expired card for a new card each year.

29. Once an expired card was received at the local office of the Department it was checked and then forwarded to Records Branch in Newcastle upon Tyne. The number of stamps were counted and recorded on the person’s RF1. From 6 April 1975 the contributions were counted and recorded on the person’s computer based record.

30. From 1975 it was possible to choose to pay either self-employed rate contributions by direct debit from a bank account or by stamp card.

31. From 6 April 1993 the payment of self-employed rate contributions by stamp card ceased. It was possible to pay by direct debit or by quarterly billing.

32. When a person registered for National Insurance he had to complete form CF8 to apply for a National Insurance card. This asks if the person has had British National Insurance before and asks for the date of arrival in Great Britain before and details of previous periods. It also asked for full details of any previous social insurance record.

33. Form CF8 was destroyed after 6 years in accordance with the department’s record management policy (on the basis that a permanent record was noted to show all National Insurance information.)

The appellant’s national insurance record

Form RF1

34. We had in the bundles produced to us a copy of the appellant’s permanent national insurance record (Form RF1). We were also able at the hearing to examine the original document. The amounts and periods of contributions recorded in the decision under appeal correspond to the contributions on this record. The additional notations on the form are set out below. It should be noted that it is a matter of dispute whether the record correctly reflects matters relating to the appellant and in particular the appellant’s contributions which were paid. The record carried the following notations:

(1) The appellant’s record was prepared by Hendon local office on 11 August 1971.

(2) The appellant entered into the National Insurance scheme on 21 July 1969. This is disputed by the appellant who claims he entered the scheme in 1961-62.

(3) “FTE from 21.7.69 to 1971”. It was explained to us by Mr Greenshields that FTE was the abbreviation for the contributor being in full-time education. This is disputed by the appellant who says that although he was on journalism course this was not full-time and he did the course much earlier than the stated period.

(4) “Bankruptcy” from 5 April 1971 to 13 June 1971. The appellant acknowledged that this accorded with his recollection. He recalled the bankruptcy had arisen because a person for whom he had acted as guarantor while working at Olga Records (UK) Limited had defaulted on a debt.

(5) Above the notation for bankruptcy two dates were written in (21.7.69 and 4.7.66) and crossed out. Mr Greenshields said he thought that this might have indicated information about the start of the bankruptcy that would have been relevant for benefit claim purposes.

(6) “CF169” for the period 21 June 1971 to 5 March 1972. Mr Greenshields explained this meant the department had agreed it would not pursue arrears of contributions for this period.

(7) For the years 1971-72 to 1974-75 the contributions had been paid late.

Computer based record

35. The contributions paid according the computer records we were shown reflected the contributions set out in the decision notice. The record showed the appellant elected to pay his contributions as a self-employed earner by the stamp card method up to 6 April 1981. The record also indicated late paid contributions in various years and the years for which “no card notices” and contribution statements were sent out.

The appellant’s work and education history in the UK

36. We found the appellant to be a credible witness who assisted the Tribunal in giving very open and full answers. Where he was not able to recollect a matter he was frank about this. We had no doubts that he was honest in putting forward his recollections. But, we must nevertheless take into account that there are inevitably going to be some concerns of accuracy in making recollections going back so far in time. We were able to make the following findings from the evidence put forward by the appellant which as well as his witness statement and his answers to questions put to him by the Respondents and the Tribunal included certain written biographical materials.

37. The appellant was born on 16 June 1944 in Sweden. In the mid-1950s, as a teenager, he was a rock and roll performer who then moved on to managing and promoting music performers in Sweden. He was intent on moving to London to further his career in the music business. After some abortive attempts he was given to leave to enter the UK for a period of time to get himself established.

38. We accept the appellant’s evidence that in the years 1963-64 to 1964-65 he found employment as a commis waiter at the Piccadilly Hotel in London and that this was a full-time position. In making this finding, in addition to the lucidity of the appellant’s recollection of the working environment, the types of customers he served, and the reason why he had been employed there (there was a Scandinavian restaurant on the premises) we were assisted by the fact that the appellant’s evidence was consistent with a reference to that employment in a book published on the appellant’s life and to an excerpt from sleeve notes to a CD which referred to him working as a waiter.

39. Although both these documentary sources derive from information that the appellant would have given to the author we think they do carry some, albeit limited, weight. First, they would have recorded recollections closer in time to the relevant events. Second, because they are unlikely to have been self-serving in the sense that they were produced well in advance of the appellant’s work history being relevant for his appeal on national insurance contributions payments.

40. With money he saved from this employment the appellant was able to buy a camera and began working as a free-lance photographer taking photographs of TV and music stars of the day on location at music events and TV studios. He wrote a monthly column for a Swedish pop magazine. He got paid for each photo published. The money was insufficient to live on so he had to supplement this income with other income from other work. He worked as a kitchen hand for the Stockpot, a restaurant in London.

41. We accept the appellant’s evidence that at some time prior to 31 December 1965 he undertook a part-time journalism course. This enabled him to get a press card, a copy of which appears in the extract from the “My 60s book”. The expiry date set out there is 31 December 1965.

42. Some time between 1965 and 1967 the appellant also took on work as a general assistant to a flower shop run by a George Gillette in the East End of London. The appellant could not remember his pay but he did get use of the delivery vehicle and petrol which helped him with his photography business and free accommodation.

43. From 1967 to sometime in 1969 the appellant worked full-time for 2 and half years as an employee of Olga Records (UK) Limited. A friend of the appellant in Sweden had a record label and wanted to branch into the UK. The appellant was a director and was head of promotion and international matters. The company employed around 5 people.

44. From 1969 to 1971 the appellant started his own record label, Green Light Records. We noted this chronology was consistent with excerpts in the “My 60s” book which referred to the appellant running his own record label in 1969. The business involved acquiring rights and arranging releases in the Benelux countries. The appellant recollected that the income for this venture came from the Netherlands.

45. It was not until 1969 that the appellant had an accountant to assist him with his tax and contribution affairs.

46. In April 1971 the appellant joined Young Blood Record Co. Ltd as their international manager and worked there until around 1975.

47. In the period 1970 to 1974 the appellant was also involved with a night club in London’s West End, Club Flicka. He was remunerated according to a percentage of the takings.

48. From 1975 onward the appellant concentrated on his music publishing business “Olofsong” which owned rights to various songs.

49. From 1980 onwards the appellant helped his wife with a catering business she had started following the appellant’s idea to market Scandinavian style open sandwiches and with the help of contacts he had in the TV business. His wife funded the business with redundancy money she had obtained. He helped out with the business but he was not on its payroll.

50. From 1985 to 1987 the appellant was unemployed.

51. From November 2001 to 2004 the appellant was responsible for the care of his youngest daughter. He took up employment a wine adviser some time in 2004 and in the summer of 2006 he relocated permanently to Sweden.

52. One of the contentions made on behalf of the appellant was in relation to possibility of errors arising in national insurance records because of language barriers. While as a general proposition there is no reason to think that might not increase the risk of misunderstandings leading to errors, we noted that the appellant’s answers and understanding of questions put to him displayed a strong command of English. The appellant told us he thought his standard of English was “pretty good” back in 1971.

The National Insurance record and the procedures for updating it

53. Mr Greenshields gave evidence on the general administrative procedures and practices which existed in bringing contributions to account and the maintenance of the records of the insured population, including the security checks in place and the accuracy of the records. We found Mr Greenshields to be an informed and credible witness and made the following findings on the basis of his evidence. We were conscious that his evidence was as to procedures and practices that ought to have been followed at the relevant times and was not direct evidence of the actual creation and maintenance of records relating to the appellant in particular.

54. Between 1948 and 1975 National Insurance contributions were paid by buying a stamp of the relevant value from the Post Office and affixing it to a stamp card each week.

55. To obtain the stamp card a person had to register themselves into the scheme. There was a network of National Insurance offices which existed in most towns throughout Great Britain. A form CF8 had to be completed and presented to the office with proof of identity for example a birth certificate or passport.

56. The Department relied on the person providing accurate information as this was the only source from which such information was obtained.

57. Once the office was satisfied the local office prepared a form RF1 and RF2 and a contribution card.

58. We heard about the arrangements for creating numbers. The local office would be provided with a list they could use. The documents were sent in sealed pouches with transit sheets which were checked upon receipt. The procedures for delivery were designed with security in mind.

59. The RF1 was a permanent record of insurance. It had the person’s identity details along with other relevant information transcribed from the CF8.

60. There are around 50 million Forms RF1 held in total.

61. After completion forms RF1, RF2 and CF8 were sent to Records Branch New Registrations Section at Newcastle upon Tyne where the forms were checked for accuracy. Where there was an indication on the CF8 of previous insurance investigations were made to establish if a National Insurance Record already existed for that person, if it was the registration process stopped and the person was informed of their original National Insurance number.

62. If there were no discrepancies the RF1 was then sent to Graduated Record section who put the National Insurance number and the identity details into the computer and created a corresponding computer record used for the recording of Graduated Contributions. This computer record started from 1961.

63. The RF1 was then sent to a Records Branch Ledger Section. They were responsible for maintenance of the record, recording contributions and other notations to the person’s insurance liability and keeping a record of benefit claims.

64. The RF1 was filed in a numeric sequence based on the National Insurance number.

65. Form RF2 was an index slip which showed only the person’s identity details, date of birth and address which corresponded to form RF1.

66. Once New Registrations had checked the RF2 it was sent to Records Branch General Index. It was filed in an alphabetical sequence based on the person’s surname, first names and date of birth. The CF8 was retained for a certain length of time but then destroyed in line with the Department’s retention of documents policy.

67. It was necessary to know a person’s National Insurance number to get access to their record. If a person’s number was not known or it was incorrect it was possible to trace the correct number from the RF2’s in the General Index.

68. General Index identified cases where a duplicate RF2 was received for the same person but where this showed a different National Insurance number to the original. They then arranged for the records to be amalgamated and one of the records to be cancelled. The person was then informed which National Insurance number they were required to use.

69. Computer based records were subsequently created from each RF1. It was possible to search this from a computer based alphabetical index. The service began in November 1981.

70. The same importance was given to identifying duplicate NI numbers and the computerised system was programmed accordingly.

71. If a contributions card could not be matched to a National Insurance record it was referred to General Index Special section who tried to trace the correct number, investigating with the local office where the card had been surrendered if necessary. If the tracing was unsuccessful a record with a National Insurance number with the prefix “CR” for “constructed record” was created and the person who had provided the card was informed of the number.

72. Where an employer submits a return bearing no National Insurance number or the identity details are so vague that the record cannot be found the contributions are transferred to a suspense file until such time as the correct record is found.

Attempts to track down the missing contributions- searches performed of different name permutations

73. We heard from Mr Greenshields about the searches he had made, without success to find any unallocated contributions or records relating to the appellant.

74. These searches were performed by Mr Greenshields personally on 20 September 2010. Mr Greenshields performed a large number of different searches with various different spellings of the appellant’s name and combinations of his surname and first names including and excluding his date of birth. These included “relaxed search” with first three letters of surname and without date of birth which would have come up with different permutations of the appellant’s surname. Under cross-examination Mr Greenshields stated that his searches had not been double checked by anybody else.

75. The Tribunal asked HMRC whether it had been able to make any searches against the employer records for employers which the appellant had named and was told this was not possible as they were in accordance with records retention policy only kept for 6 years.

Appellant’s arguments

76. The appellant entered into National insurance before 21 July 1969 (6 April 1962).

77. He paid National Insurance between 1962 and 1969 but these contributions have not been taken into account.

78. Insufficient contributions / credits have been taken into account in calculating the appellant’s entitlement for periods after July 1969.

79. Between his entry into National insurance in 1962 and his departure from the UK in July 2006 the appellant should be regarded as having a full National insurance contribution record.

80. As a result of error the appellant has a National Insurance record for periods prior to 1969 that cannot now be traced. It is likely the appellant had a different National Insurance number which cannot now be located. There is a reasonable possibility of there being a second RF1 under a different National Insurance number on which there are further contributions and given that the appellant should be credited with those contributions accordingly.

(1) The appellant’s name is prone to being misspelt – see for example errors in the Respondents’ Statement of Case – incorrectly spelt on no fewer than five occasions. Similarly there may be errors in how national insurance numbers have been written down (a form sent from the Inland Revenue to the appellant provides an example of this). Anglicising of a foreign name must provide further scope for errors in transcription.

(2) It is not uncommon especially amongst visitors from abroad who may have made different visits, and because of language barriers, that there is a failure to understand the system and that there might be errors in records or multiple NI numbers.

(3) There are 100s of local offices and a number of staff in each qualified to deal with new persons registering and therefore obvious scope for inconsistency in practice.

(4) Forms RF1 are filed in NI number order – no means of finding them without NI number. The RF2 cross references filed in name order were destroyed in 2004.

(5) It is an impossible task without the NI number to find a specific form.

(6) Without checking each RF1 and this is not possible to do we can never be sure there is not a second record.

81. In relation to the onus of proof, while the onus rests on the appellant to show the decision should be varied, the appellant needs to prove he did not enter National Insurance on 21 July 1969 which put him in the position of trying to prove a negative. On this point the onus of proof rests with HMRC and it is for them to demonstrate the robustness of the records and systems on which their decision depends.

82. The Respondents have not put any direct evidence forward:

(1) There is no CF8 which forms the basis of the RF1.

(2) There is no evidence from anyone who has worked on the appellant’s record.

(3) The evidence is of procedures which should have happened, that is not evidence of what happened.

(4) There are no copies of correspondence; statements etc. that HMRC say were sent to the appellant.

(5) There are no copies of minutes of meetings that appellant may have attended.

83. While the destruction of records may be in accordance with a policy the appellant should not be penalised for a decision which is adverse to him on record destruction.

84. The appellant has his own evidence and secondary evidence as to his employments pre 1969 in the UK – this is stronger than HMRC’s reliance on form RF1 which is not supported by any other corroboration.

85. The other cases before the Tribunal to which we were referred were decided on their facts. The appellants were in such cases unrepresented and it is possible there is a question over the level of openness in what was placed before the Tribunal.

Respondents’ arguments

86. There is no evidence to suggest the RF1 is wrong and nothing to indicate any errors of law. There is no evidence he paid National Insurance contributions before the 1971-72 contribution year.

87. The appellant’s National insurance records both pre and post 6 April 1975 are correct. The records show he first registered for National insurance on 11 August 1971 and was treated as having entered into insurance on 21 July 1969.

88. Despite extensive searches having been carried out it has not been possible to trace an earlier National Insurance contribution record which relates to the appellant.

89. The appellant’s date of entry into the National Insurance scheme of 21 July 1969 was based on information he provided on form CF8. If he had completed the form to say he had been in the UK previously and had previous insurance a trace for his previous record would have been made and if not found his record would be been noted to the effect “previous insurance declared but not found”.

Discussion

90. The issue for the Tribunal’s determination is whether contributions have been paid in the period 1962 to 2006 additional to those which are set out on the appellant’s permanent record as put forward by the Respondents in their decision under s8 SSC(ToF)A 1999 (“the HMRC decision”), and against which the appellant has appealed.

91. Under the relevant legislation, if it appears to the Tribunal that the decision should be varied, the Tribunal may vary it but otherwise the decision shall stand good.

Onus of proof

92. In relation to onus of proof it was put to us by Mr Mitchell that while in general terms it was accepted that the onus lay on the appellant to show why the decision should be varied, here, because the appellant was required to prove a negative, in the sense that he had to show he did not enter the national insurance system in 1969, the onus should pass to HMRC to demonstrate the robustness of the records and systems on which their decision depends.

93. We do not accept that argument. The HMRC decision set out the contributions it maintains the appellant has paid and it is for the appellant to demonstrate why HMRC’s decision should be varied. The appellant argues that he has in addition to extra contributions post 1969 paid contributions in an earlier period. He is not required to prove that he was not registered for the first time from 1969; rather he must show that he paid contributions in an earlier period. As part of that it is open to the appellant, as he has done so in this appeal, to lead evidence to suggest the appellant was in the national insurance scheme earlier than 1969 and making payments of contributions before then. That does not require proof of a negative.

Other Tribunal’s decisions and approach to evidence

94. We were referred to a number of decisions where other tribunals had had to consider evidence in relation to the written record in relation to national insurance kept by the Department on the one hand and evidence from the appellant on the other. Gutteridge concerned the issue of whether the appellant had made an election available to married women not to pay contributions. Beamish concerned the issue of whether contributions had been paid and so deals with similar issues to this appeal. Similarly Rose and McGough amongst other matters deal with the accuracy of the appellant’s contribution record. Apart from McGough which was before a Social Security Commissioner all three cases were before the Special Commissioners. While none of the decisions are binding on this tribunal it is possible for them to be of persuasive value.

95. In each of those cases, detailed evidence was led, as in this case, on the national insurance record and the procedures for maintaining it. In this appeal Mr Mitchell drew our attention in his submission to the lack of primary evidence. We do not for instance have evidence of various communications and documents passing between the appellant and HMRC and the predecessor Departments. We do not have the appellant’s CF9, the stamp cards relating to contributions paid, shortfall letters / statements sent to the appellant. We noted that there was a similar lack of primary evidence in the other cases but that the Special Commissioners and Social Securtiy Commissioner nevertheless found themselves able to make findings of fact on the basis of secondary evidence and that they did this having duly considered the relevance and weight of other evidence brought on behalf of the appellant.

96. It seems to us that the persuasive value of these cases is limited in that those cases turned on the particular weight the Special Commissioners and Social Secruity Commissioner were able to attribute to the evidence before them. While those decisions are helpful in providing examples of the fact that it is open in principle for tribunals, to make findings of fact from secondary evidence, and also of the balancing that must be undertaken in assessing evidence they do not provide us with any short-cut around the exercise of assessing and weighing the particular evidence put before us. Given this we do not need to deal with Mr Mitchell’s submission that limited weight should be put on these other cases because the appellants there were unrepresented. We would nevertheless remark that there is nothing in those other decisions which struck us as suggesting the Commissioners’ evaluation of HMRC’s evidence, was materially less robust or their consideration of relevant arguments for the appellant was materially less comprehensive due to the lack of the appellants’ representation in those cases.

The appellant’s evidence

97. As mentioned above we found the appellant to be a credible witness and we had no doubts that his account of his recollections were genuine and honest. His oral evidence together with certain documentary evidence in the form of biographical material, which we accept, given the circumstances of its genesis does not give rise to concerns that the evidence is self-serving, has enabled us to make certain findings of fact in relation to the appellant’s work in periods preceding 1971. The relevance of such findings is that they lay a preliminary foundation for a contention that contributions were paid in respect of such employments and self-employments. They do not of themselves provide direct evidence of payment of contributions however.

98. While we have been able to make some findings of fact the limitations of relying on the appellant’s recollections from matters from 40 to 50 years ago must be acknowledged. We think distinctions may be drawn between the types of information one might reasonably remember over that length of time with some types of information being more amenable to recollection and therefore given with greater lucidity than others. Places of work, who a person worked with and the sequence of a person’s different jobs /activities are we think more likely to stick in the mind, the legal status of whether a person was working as an employee, a partner, or self-employed maybe less so. We are less confident about relying on a person’s recollections of visits to contributions offices, indeed the appellant was frank about, quite understandably, not remembering the details of when these took place and what was said. The appellant’s evidence covers recollections of deductions being made from certain employments in respect of contributions, and of payments being made in relation to certain self-employment income. We were of the view these recollections could be given very limited weight. We are not suggesting the appellant was in any way dishonest about these recollections but we suggest these are the sorts of matters it would not be reasonable to expect a person to necessarily remember one way or the other with any degree of certainty over the length of time in issue here.

99. The limited weight attached to such recollections does not of itself mean the contributions were not paid. The likelihood of them being paid needs to be assessed in the light of all the evidence that has been put forward.

Completeness of HMRC’s record of contributions / procedures in place for recording contributions and retrieving records

Was there a duplicate record / unallocated contributions?

100. The appellant puts forward an explanation for why the national insurance records we were shown, both the permanent RF1 form and computer based records are not reflective of the contributions the appellant has paid. He suggests there is a reasonable possibility that there is a duplicate record or records which if found would show the missing contributions. The system for recording and tracing contributions and for identifying and amalgamating duplicate records is susceptible to error. The likelihood is increased where a contributor has a surname such as the appellant’s which is prone to being misspelt.

101. We heard from Mr Greenshields that a search would be made for a prior record if the contributor had stated on the form CF2 that they had previously been subject to insurance. But, if they did not state this, the implication we draw from this is that a search would not be made. From Mr Greenshield’s evidence a search for a duplicate would be picked up when the RF2 was sent in as the General Index section would pick up if someone with the same name had two different national insurance numbers. We did not hear evidence as to what extent this search would look only at identical names or whether it would be broader so as to capture misspelt names.

102. We were told that there is now a computer based system for tracing National Insurance records from an alphabetical index, that there is a computer record for each RF1 and that the same importance is given to identifying duplicate NI numbers. Mr Greenshields explained how extensive searches had been performed on the computer using various different possible spellings of Mr Olofsson’s name and a “relaxed search”.

103. Mr Greenshields accepted that these searches were not double checked. We recognise that tracking down unallocated contributions is a task that must be executed in a fully comprehensive and diligent manner given the effects the outcome of the searches may have on a contributor’s entitlements. We noted Mr Greenshield’s experience with contributions records, and the serious approach he took their maintenance and were satisfied that Mr Greenshields had carried out all the searches he could reasonably be expected to perform and that he had executed and recorded his searches with the necessary degree of care. We note that even if a duplicate record had not been identified by the General Index search for duplicates because the appellant’s name had been spelt different (see [101]), the duplicate record would we think have been identified by these computer searches.

104. Apart from his evidence in relation to the searches he has performed, Mr Greenshield’s evidence is of course only evidence of what ought to have happened in terms of procedures and checks. We do not have direct evidence of what actually happened in relation to how those procedures and checks were operated in fact and in particular in relation to the appellant. Nevertheless it is open to the Tribunal to draw inferences from evidence on what ought to have happened and find that on the balance of probabilities that is in fact what happened.

105. In relation to the destruction of original contributions records and forms (which it is not disputed has been in accordance with policy) Mr Mitchell argues the destruction of records is adverse to the appellant and the appellant should not be penalised for the policy decision to destroy original records.

106. In this regard we note the discussion on retention of records in the Rose case where the Special Commissioner considered the approach of Social Security Commissioner decision of Commissioner Mitchell in R(IS) 11/92. The point in the Social Security Commissioner case covered whether an appellant should be entitled to the benefit of an adverse assumption against the Secretary of State responsible for social security records because the Secretary of State had organised the destruction of the appellant’s records. The Special Commissioner in Rose refined the Social Security Commissioner’s approach and rejected the argument that HMRC should have an adverse inference drawn against them because forms issued to the appellant could not be produced, those forms having been destroyed in accordance with a documents retention policy.

107. While Mr Mitchell has framed his argument slightly differently, taking account of the above, we can see no basis for either ignoring secondary evidence which has quite reasonably been put forward given the absence of the primary evidence, or making any presumptions in favour of the appellant due to the destruction of primary records. The task remains that of considering all the available evidence and making findings of fact on the balance of probabilities. We would add that without the benefit of the primary records it cannot necessarily be assumed that their absence is in any case wholly adverse to the appellant.

Accuracy of record / newspaper cuttings

108. The Tribunal had before it newspaper articles put forward by the appellant which suggest that there are 9.3 million individual records that have not been matched up since 2004.

109. We noted the article contains a quote to the effect that where the contributions cannot be matched the contributions are retained until such time as they can be matched.

110. Mr Greenshields’ evidence covered how records were subject to audit checks. Nevertheless he accepted in cross-examination that errors could be made. He put the possibility of this as low “say 1%”. Mr Mitchell made the point that even if there was an error rate of 1% when dealing with 50 million records would give rise to a significant number of records with errors (500,000).

111. The appellant also highlighted that there had been errors on the RF1 certainly in relation to information given in relation to dates the appellant was said to have been in full-time education and ambiguity about the notations in relation to his bankruptcy.

112. Mr Mitchell contends that without checking each RF1 which is not possible to do we can never be sure there is not a second record. We do not disagree with that from the point of view that a search of all the RF1s in existence while possible in principle would be hugely disproportionate as a matter of practice, but we do not think this argument means we must find that there is a duplicate record. The standard of proof we must employ, and we do not understand this to be a matter of contention between the parties, is one of a balance of probabilities.

113. Putting to one side the relevance and weight of the appellant’s evidence which we consider further below, we would be satisfied, on the basis of HMRC’s evidence that on the balance of probabilities the procedures for identifying duplicate records did operate as described in the relevant period and it more likely than not that a duplicate record would have been identified by the general index and subsequently the computer checks. Even if the appellant’s name was misspelt, or it was possible that somehow there was a record of the appellant’s where there was a transposition error in the appellant’s national insurance number, we think it is more likely than not that this would have been picked up in the comprehensive computer based checks for different name permutations that Mr Greenshields performed.

114. While it cannot be ruled out that there were errors in sending RF1s from local offices to Longbenton, given what we have heard about the security procedures in place for sending and logging documents between the offices, and audit processes, we think it is more likely than not that any RF1 completed in respect of the appellant prior to 1969 would have been sent to Longbenton.

115. Come 1975 and the move to the new computer system, while it cannot be ruled out that there might be errors transposing the RF1s to the computer record it is more likely than not in our view that if there was a duplicate RF1 it would be included in the transposition to the computer and would then have been picked up by the computer search.

117. Neither the newspaper cuttings, (even discounting any issues as to the limited evidential weight to be attached to them), Mr Greenshield’s concession on accuracy, or the errors and ambiguities in relation to the notations in relation to the appellant’s periods in full-time educations and bankruptcy persuade us the position is otherwise.

118. In relation to the newspaper cuttings even if it were to be assumed that this level of unmatched contributions subsisted in the period 1961-2 onwards we take into account that in this matter we have heard evidence about the specific attempts that have been made to find contributions not on the appellant’s record. We are satisfied on the balance of probabilities that there are not unmatched contributions recorded elsewhere and if there were they would have been found by the searches undertaken.

119. In relation to error rates we approach Mr Greenshield’s percentage estimate of error with caution as no evidence was given supporting how the percentage was derived. The relevance of a general error rate however that is defined also has to be considered in relation to the issue here which is whether there was an error which led to the creation of at least more than one record and the attribution of contributions to that other record. On the evidence before us, although not impossible we were satisfied that the likelihood of an error leading to an additional record being made was extremely low.

120. While, having heard the appellant’s evidence we find it more likely than not that the notation on his record about his period of full-time education was incorrect because he was in education, but not full-time in an earlier period, we find, taking Mr Greenshield’s evidence into account, that the information would only have been as accurate as the information given by the appellant. The appellant understandably is not in a position to recollect in any detail his attendance at the interview of what was said. When asked why he had felt the need to register in 1971 if he had already been registered previously he thought this may have been because the finance/ accountancy person at Youngblood Records had suggested this. Without evidence to the contrary we find that on the balance of probabilities the notation on full time education reflected the interviewer’s understanding of information provided by the appellant. The fact it is incorrect does not in our view throw into doubt the accuracy of contribution record in so far as it records contributions paid given the system that was in place for recording contributions received. The inaccuracy does not in our view make it any more likely that there is a duplicate record or unallocated contributions. Similarly any issue over whether the period of the appellant’s bankruptcy was correctly recorded or not does not in our view make it more or less likely that there is a duplicate record or unallocated contributions.

121. Before moving on to balance the weight of the respective pieces of evidence we need to deal with certain issues raised by the Respondents which we have considered but which we think are of limited relevance in determining whether contributions of a particular class were paid.

Relevance of letters being sent highlighting shortfalls

122. Despite there not being direct evidence of the actual letters sent from HMRC and its predecessors to the appellant, given the systems in place for logging the letters on the RF1 and computer records we find it more likely than not that the letters that the RF1 stated were sent to the appellant were actually sent. These letters appear to request return of contributions cards. We heard how the computer based record showed that a statement informing the appellant of his benefits position for 1977/78 was issued and we were shown an example of the sort of letter that would have been issued. We noted this related to the specific contribution year and not the full contribution record. For the remaining years in dispute the computer record showed letters had been issued either requesting the return of the contribution card or that a statement was issued.

123. There was nothing to suggest any of these letters would have highlighted to the appellant that his contributions record only started in January 1971. We therefore think they are of limited relevance to the question of whether contributions were paid or not paid certainly in the periods prior to 1971. They may be of more relevance for contribution periods after 1971 in the sense that where a letter highlights deficiencies in contributions to a contributor for a year that has recently passed, it might be expected that a contributor who disputed that would query the matter then. Even then, on the facts of this case and given what we have heard about the appellant’s focus on his music business and photography career, his reliance on others to handle his financial affairs from 1969, and that the appellant might not have appreciated the significance of deficient records, it would not surprise us if the shortfall letters did not prompt a response from the appellant even if the record of contributions paid was incorrect. We therefore place limited weight on the shortfall letters for the issue of whether the record is correct for post 1971 periods.

Relevance of paying contributions late

124. We were referred to the judgment in Beamish that the appellant had been a slow payer, and asked to note the appellant was a slow payer. We do not see how that helps establish one way or the other whether the contributions pre 1971 were paid or not or whether additional contributions to those shown on the appellant’s record were paid post 1971. Certainly in relation to periods when the appellant was employed it is irrelevant as it would be for the employer to deduct amounts in respect of contributions and pay over the contributions.

Absence of notation on record “previous insurance declared but not found”

125. HMRC submitted that if the appellant had completed his form CF8 to show that that he had been in the UK previously and had had previous insurance a trace would have been carried out, and if unsuccessful a note of this would have been made. No such note was made. Mr Greenshield’s evidence did not deal with the particular notation but even if it had done and we had made a finding of fact that this was the notation that would have been made its primary assistance would be in showing that it was more likely than not that the appellant had not declared previous insurance in the UK.

126. In relation to whether the absence of such a notation makes it more likely than not that previous contributions were in fact paid, we think such absence would have been of only limited relevance. The relevance of such absence to whether previous contributions were paid is dependent on making an assumption that the appellant would be more likely to declare previous insurance if previous contributions had been paid than to have not declared previous insurance. While that assumption does not seem unreasonable as a general proposition particularly given the example form of the application for national insurance form CF8 alerts the reader to the importance of giving this information given the effect on benefit claims, we did not have sufficient evidence before us to establish that it necessarily followed that any lack of declaration by this particular appellant indicated it was more likely than not that contributions had not been paid previously.

Weighing up the evidence

127. We have on the one hand the appellant’s own oral evidence of his periods of employment and self-employment and recollections of deductions of contributions being made, and in the case of self-employment of contributions being paid. Some of the employments and self-employment are also mentioned in biographical material about the appellant. On the other hand we have HMRC’s record of the contributions it maintains the appellant has paid, evidence from an officer of HMRC as to the procedures surrounding the creation, maintenance and retrieval of national insurance contributions records, and the officer’s own evidence of the attempts he has made to locate any missing contributions.

128. We have outlined above the limitations of relying on the appellant’s recollections given the length of time that has passed, and the points of detail that it would be reasonable to expect him to remember with any degree of reliability. In particular we have not from the appellant’s evidence been able to make a finding of fact that deductions of contributions were made or that the appellant paid self-employed contributions.

129. While employment and self-employment are pre-requisite to payment of employed earner or self-employed earner contributions, any inference that as a result contributions were in fact paid has to be weighed against the documentary evidence held by HMRC, and the evidence around the record creation, maintenance and retrieval.

130. It is our view that the documentary evidence on the appellant’s national insurance record and the evidence on procedures surrounding it should be given significant weight. We are satisfied that while not failsafe, the procedures that would have been place for recording contributions payments were highly robust. The documentary evidence and surrounding evidence on national insurance procedures outweighs the inferences that might otherwise be drawn from the fact of the appellant’s employments and self-employments prior to 1971 (which facts are themselves based on recollections going back 40 to 50 years ago). We disagree with the appellant’s submission that there is a reasonable possibility that there is a duplicate RF1. While the possibility of a duplicate record cannot be ruled out, we find the likelihood of that to be the case to be extremely low and on the balance of probabilities we find that there is no duplicate record.

131. It is possible to be in employment but not have had deductions, or to have had deductions but not have had them paid over by the employer. While those situations, which would involve an abdication of responsibility, intentional or otherwise on the part of the employer and for that reason would not necessarily be that common place we think it is more unlikely, given the robustness of the processes surrounding the national insurance record that, contributions having been paid, there was an error in failing to create a contribution record. Similarly we think it is more unlikely than the scenario of a remiss employer that a duplicate national insurance record was created which was not then picked up in the extensive searches made and more unlikely that contributions were paid but then not recorded on a suspense file or recorded but then not traced. The absence of a record of payment of contributions when viewed against the backdrop of the procedures for creation, maintenance and retrieval of records suggests to us that on the balance of probabilities no payment was made.

132. In relation to periods of self-employment, where it would be up to the self-employed contributor to make the payment, it strikes us as being quite possible that the contributor might fail to make the payment. Again, given the robustness of record-keeping the absence of a record of payment suggests to us that on the balance of probabilities no payment of contributions was made.

Missing contributions post 1971

133. The appellant maintained that the record for the 2 years 1977 and 1978 was incorrect in that he had continued to work for Young Blood Record Co Ltd, and the record for 1980 through to 1986 were incorrect because he was involved in his wife’s catering business Kathie’s Kitchen Ltd. We noted however that it was not clear from the appellant’s oral evidence that his work with Young Blood Records Co Ltd did continue past 1976. In relation to the catering business our understanding of the appellant’s role was in generating the idea for the business, and that it was not one in which had an active ongoing financial or other role that would suggest it was likely that full contributions were paid during 1980 through to 1986.

134. For the contribution years 1990 to 1991 the appellant queries why these are incomplete on the record given his recollection that contributions were accounted for in relation to income from publishing rights of his songs held through Olofsong Music.

135. Weighing up the appellant’s recollections as to contributions being paid in the post 1971 period against the appellant’s national insurance record, and procedures governing their creation and maintenance of that we give greater weight to the record. The periods in issue are still a significant way in the past and we are not persuaded the appellant’s recollections of contributions being paid, although truthfully recollected, can be given as much weight as the record.

136. In relation to periods of employment and self-employment post 1971 it seems even more improbable to us (than in relation to the pre 1971 period) that the appellant having registered for national insurance, and a national insurance record having been created, and some contributions having been recorded on that record, that further contributions which had been paid would not have been put onto the record.

137. We find that on the balance of probabilities no additional contributions were paid by the appellant.

What if appellant suffered deductions as an employed earner but amounts were not paid over? Regulation 60 of the Social Security (Contributions) Regulations 2001

138. Given we have been able to make findings for the purposes of this appeal as to some of the employments the appellant held but have found that contributions were not paid we hope it will be useful to mention the following regulation, to which were alerted to in the Gutteridge case.

139. Regulation 60 of the Social Security (Contributions) Regulations 2001 covers the treatment of unpaid primary Class 1 contributions where there is no consent, connivance or negligence on the part of the primary contributor, for the purposes of entitlement to contributory benefits.

140. The regulation is only available where contributions have not been paid. That is the finding here. The appellant worked in a number of places. From the detail of his evidence we were satisfied that he was in full-time employment as a waiter at the Piccadilly hotel, and later with Olga Records (UK) Limited for the periods set out in our background findings.

141. While it seems more probable than not that the work at the Stockpot as a kitchen hand and at the flower shop as a general assistant were on an employed earner basis we were unable on the evidence before us to make any specific findings as to the period over which the work took place and the number of hours worked. Furthermore, the nature of the work, the casual hours and likely rates of pay suggests that primary contributions might not have been due in any event. That is particularly the case with the work for the flower shop which seems to some extent to have been on a “favour given for a favour received” basis.

142. On the evidence before us we think it is more likely than not that the appellant’s work as a photographer and in running the label Green Light records was not on an employed earner basis.

Conclusion

143. We are not satisfied that on the balance of probabilities that contributions pre 1971 were paid or that there were contributions paid post 1971 which are not reflected on the record. The contributions record is an accurate record of the contributions which have been paid by the appellant. Accordingly the appeal is dismissed.

144. In view of our findings and Regulation 60, the appellant may wish to ask HMRC to consider whether there is any scope to treat any unpaid contributions from the appellant’s periods as an employed earner as paid.

145. This document contains full findings of fact and reasons for the decision. Any party dissatisfied with this decision has a right to apply for permission to appeal against it pursuant to Rule 39 of the Tribunal Procedure (First-tier Tribunal) (Tax Chamber) Rules 2009. The application must be received by this Tribunal not later than 56 days after this decision is sent to that party. The parties are referred to “Guidance to accompany a Decision from the First-tier Tribunal (Tax Chamber)” which accompanies and forms part of this decision notice.

SWAMI RAGHAVAN

	Mr JAN EVERT INGEMAR OLOFSSON	Appellant

	- and -

	THE COMMISSIONERS FOR HER MAJESTY’S	Respondents
	REVENUE & CUSTOMS

First-tier Tribunal (Tax)

Introduction

Evidence

Law

Background facts

History of decision / appeal

Computer based record

The appellant’s work and education history in the UK

Appellant’s arguments

Respondents’ arguments

Discussion

Onus of proof

Other Tribunal’s decisions and approach to evidence

The appellant’s evidence

Completeness of HMRC’s record of contributions / procedures in place for recording contributions and retrieving records

Was there a duplicate record / unallocated contributions?

Relevance of letters being sent highlighting shortfalls

Relevance of paying contributions late

Weighing up the evidence

Missing contributions post 1971

Conclusion

TRIBUNAL JUDGE

RELEASE DATE: 3 August 2012

	[Home] [Databases] [World Law] [Multidatabase Search] [Help] [Feedback]
	First-tier Tribunal (Tax)
You are here: BAILII >> Databases >> First-tier Tribunal (Tax) >> Olofsson v Revenue & Customs [2012] UKFTT 490 (TC) (03 August 2012) URL: http://www.bailii.org/uk/cases/UKFTT/TC/2012/TC02167.html Cite as: [2012] UKFTT 490 (TC)

TRIBUNAL:	JUDGE SWAMI RAGHAVAN
	CHARLES BAKER FCA

First-tier Tribunal (Tax)

Introduction

Evidence

Law

Background facts

History of decision / appeal

Computer based record

The appellant’s work and education history in the UK

Appellant’s arguments

Respondents’ arguments

Discussion

Onus of proof

Other Tribunal’s decisions and approach to evidence

The appellant’s evidence

Completeness of HMRC’s record of contributions / procedures in place for recording contributions and retrieving records Was there a duplicate record / unallocated contributions?

Relevance of letters being sent highlighting shortfalls

Relevance of paying contributions late

Weighing up the evidence

Missing contributions post 1971

Conclusion

TRIBUNAL JUDGE

RELEASE DATE: 3 August 2012

Completeness of HMRC’s record of contributions / procedures in place for recording contributions and retrieving records

Was there a duplicate record / unallocated contributions?