1. The Royal Insurance Co., does business in the United Kingdom, India, and other parts of the world. It has done business in India for some 60 years. During that time it has carried on life assurance business. Previous to the year in question, the company had been assessed under Rule 35, Income-tax Rules, which pro- vides as follows:
The total income of the Indian branches of nonresident insurance companies (life, marine, fire, accident, burglary, fidelity guarantee, etc.) in the absence of more reliable data may be deemed to be the proportion of the total income, profits or gains, of the companies, corresponding to the proportion which their Indian premium income bears to their total premium income. For the purpose of this rule the total income, profits or gains of non-resident life assurance companies whose profits are periodically ascertained by actuarial valuation shall be computed in the same manner as is prescribed in Rule 25 for the computation of income, profits and gains of life assurance companies incorporated in British India.
2. Rule 25 is in these terms:
In the case of life assurance companies incorporated in British India whose profits are periodically ascertained by actuarial valuation, the income, profits and gains of the life assurance business shall be the average annual net profits disclosed by the last preceding valuation, provided that any deductions made from the gross income in arriving at the actuarial valuation which are not admissible for the purpose of income-tax assessment, and any Indian income-tax deducted from or paid on income derived from investments before such income is received, shall be added to the net profits disclosed by the valuation.
3. In the year 1925 the company began to have separate actuarial valuations of its Indian life assurance business made. Those valuations were made by an actuary in London. The beginning of the period was 1st January 1925, and the form of the valuation which is set out at p. 11 of the paper book follows the form that is prescribed by Section 8, Indian Life Assurance Companies Act of 1912, which provides:
Every life assurance company shall once in every five years, or at such shorter intervals as may be prescribed by the instrument constituting the company, or by its regulations or bye-laws, cause an investigation to be made into its financial condition, including a valuation of its liabilities, by an actuary, and shall cause an abstract of the report of such actuary to be made in the form set forth in Schedule 4.
4. The form in question in this case is that set out in Schedule 4 and is as follows:
Fund at 1st January 1925 ... ... 466,508 Claims paid and ... 184,571
Premiums ... ... ... 500,653 Surrenders ... ... ... 51,671
Interest, less ... ... 142,291 Bonuses paid in ... ... 4,229
Commission ... ... ... 49,214
Expenditure of ... ... 43,410
Taxes other than ... ... 191
Fund at 31st Deoember ... 776,166
___________ 1929 __________
5. Then follows the valuation balance sheet as at 3lst December 1929:
Fund ... ... ... ... ... ... ... ... 776,166
Net liability ... ... ... ... ... ... ... 672,147
SURPLUS ... ... ... ... ... ... ... ... 104,019
6. There was another valuation made for the Quinquennium 1930-34 which is as follows:
Fund at 1st January 1930... ... 770,588 Claims paid and ... 297,160
Less amount transferred to Surrenders ... ... ... 124,306
share-holders life profits ... 9,429 Bonuses paid in cash ... 10,564
_______ Commission ... ... ... 49,856
761,159 Expenses of management... 38,486
_______ Proportion of Home ... 12,336
Premiums ... ... ... 668,576 Taxes other than ... 692
Interest less income-tax... ... 187,320 Fund at 31st ...1,083,655
7. Underneath that is the actuarial balance sheet as at 3lst December 1934 which is as follows:
Fund ... ... ... ... ... ... ... ... 1,083,655
Net liability... ... ... ... ... ... ... ... 952,012
SURPLUS ... ... ... ... ... ... ... ... 161,643
8. Previous to the year of assessment now in question (1935-1936) the assessees had been assessed under the provisions of Rule 35 of the Income-tax Rules on a proportion of their total world life profits. But in the year in question they claimed to be assessed on an annual income which was one-fifth of the surplus of 161,643. The reasons given were that the proportionate assessment under Rule 35 was not accurate in that the premiums on the Indian business were higher relative to the sums assured and the profits, and that consequently a wrong proportion was arrived at as the assessable income. The assessees tendered the balance sheets I have quoted as evidence of what their real income was. The Income-tax Officer and the Assistant Commissioner and the Commissioner, however, proceeded to assess the company under the proportionate provisions of Rule 35 on the ground that the accounts provided by the company were not accurate, because (they said) the fund as at 1st January 1925, namely, 466,508, was not the correct figure, in that it was too low, and because of that the interest which was credited to the fund during the following quinquennium was too low ; consequently the fund as at 3lst December 1929 and the surplus on that date were both too low following upon that the fund as at the beginning of the quinquennium 1930-34, namely, 770,588 was too low with the result that the interest credited in that quinquennium was too low so that the fund as at 3lst December 1934 was too low and the surplus 161,643, one-fifth of which was to be taken as the income in the year of assessment, was also too low.
9. In short the income-tax authorities contended that the foundation of the accounts was wrong. They base their objection to that starting fund of 4,66,508 because they alleged that it did not really represent the life fund attributable to Indian business at that date, and they refer to the report on the Quinquennial Valuation of the Royal Assurance Co.'s (world) Life Department as at 31st December 1924 which was signed by the company's Actuary on 24th April 1925. Some figures from that report are extracted in the cage which has been stated, but not the whole of it, and I have referred to the report itself which, of course, was before the income-tax authorities at the time this case was stated and for sometime previous. At page 3 of the report it is stated:
The result of the valuation is to show a surplus of 1,611,626 as follows: Assurances Annuities Total Fund 15,187,179 1,420,341 16,607,520Liability 13,595,380 1,400,514 14,995,894Surplus 1,591,799 19,827 1,611,626
10. Then follow these words; they are the actuary's words :
I recommend that 311,617 should be carried forward as an unappropriated balance, and in order that the amount carried forward may bear the same proportion to liabilities in the Annuity section, as in the Assurance section, I recommend that 10,000 be transferred from the Assurance Fund to the Annuity Fund.
This leaves 1,300,009 available for distribution. I recommend that 130,000 should be allocated to the shareholders and that 1,170,009 should be applied for the benefit of participating policy holders.
11. With the rest of the report I have no concern. It is common ground that the starting fund as at 1st January 1925, viz. 466,508, is the amount at which the total liabilities of the company under its life policies in India were estimated actuarially at that date. The income-tax authorities however say that the sum of 311,617 has been carried forward from the total world fund, but that no part of it has been carried forward to the Indian fund which was just enough on 1st January 1925 to meet the company's liabilities in India, and that the correct amount of the fund should be a sum of 466,508 plus some part of that 311,617 which could be said to be attributable to the Indian life fund which had been in operation for a large number of years. Now, the Insurance company contended that 466,508 was the correct value to be placed on the fund as at 1st January 1925 whether that figure is correct or not I do not propose to say, but it seems to me that the income-tax authorities were wrong in assessing the company under the proportionate scheme of B. 35 on the ground that they had no more reliable data. In National Mutual Life Association of Australasia, Ltd. v. Commissioner of Income-tax Bombay Presidency and Aden Lord Thankerton at page 109 said:
The Income-tax Officer is only authorised to have recourse to the method of computation provided by Rule 35 in the absence of more reliable data.
12. Rule 35 applies to all kinds of insurance- Life, Fire, Marine, Accident, etc. One can well understand that the conditions as regards life assurance are different in India from what they are in other parts of the world, and it may be that the proportionate scheme under Rule 35 is not accurate but merely what was described in another case as a working rule. But were the data before the income-tax authorities more reliable As regards the method in which the profits and gains were computed in the two quinquenniums in question no objection can be taken. The computation was in accordance with the scheme laid down by statute for insurance companies resident in British India. The computation was made by an actuary and it is not alleged except in one particular that it is wrong. The only particular objected to is the amount of the fund as at 1st January 1925. If that fund is wrongly stated there is ample material before the income-tax authorities from which they can correct it. It seems to me that if that sum 466,508 is wrong it is possible from the data that I have read out previously from the actuary's Report on the Quinquennial Valuation of the company's Life Department at 3lst December 1924 for persons who are competent in such matters (and the income-tax authorities have such competent persons) to ascertain what was the correct and proper Indian life fund as at 1st January 1925, reckoned in the way that the income-tax authorities say that it should be done.
13. That being so, it appears to me that instead of the income-tax authorities rejecting the valuation and revenue accounts of the company in toto and proceeding to assess the company under the proportionate scheme of Rule 35, they ought first to have applied themselves to the data before them and ascertained first the correct amount of the fund and then the profits as at 1st January 1924 in accordance with the methods customarily employed and, in fact, designated by statute to be employed in estimating the profits of life assurance companies. For that reason I am of the opinion that the income-tax authorities having before them reliable data on which they could have computed in the usual way the income, profits and gains of this company for the year in question, were wrong in proceeding to assess this company under the proportionate scheme of Rule 85. The result is that the question asked in the reference is answered in the negative. No order is made as to costs. The insurance company is. entitled to a re-turn of the deposit before the Commissioner.
14. I concur in the order which has been made, but I wish to add a few observations to what has fallen from my Lord. If an assessee, on proper materials, can show to the income-tax authorities that his profits and gains are in fact less than they would be if ascertained by some conventional or arbitrary method, such as that prescribed by the concluding paragraph of Rule 35, it would be a serious injustice to relegate him to that method. Accordingly the rule provides that the method is only to be applied in, what is described as, 'the absence of more reliable data.' Beaumont C.J. in Manufacturers' Life Insurance Co. of Canada v. Commissioner of Income-tax Bombay ('38) 6 I.T.R. 321 observed at p. 329:
The introduction of the comparative is not, I think, very happy, because almost any data might be more reliable than the data on which the rule of the thumb is based. But I take it the rule means that it is to be applied unless the assessee provides some reliable data on which the Income-tax Officer can make a satisfactory assessment.
15. I respectfully agree with the learned Chief Justice's criticism of the language of the rule and I may add that in my opinion it is regrettable that in framing the rule, the Central Board of Revenue did not use ordinary English language, instead of importing what is really a Latin word, namely, 'data.' That word I consider to be misleading because it is used both for actual materials from which inferences are to be drawn, and also for assumptions from which inferences are to be drawn. But I think it is clear from the language of Rule 35, taken in connexion with the subject-matter, that the rule uses 'data' in the former sense of facts and materials. Now, if the contention of the assessees is right, as to which I offer no opinion, that the taxable income, as they have shown it based on the calculations whose starting point is 1st January 1925, is conclusive, then there can be no question but that those data are more reliable than a result arrived at by following the method laid down in Rule 35. Assuming that they are wrong, I agree that there were ample materials upon which the Income-tax Officer could fairly assess what the profits and gains of the Indian branch of the assessee company were. In addition to what is called the actuarial valuation, the Income-tax Officer had before him a statement which is Annexure 'A' to the letter of reference and sets out the total life fund of the company and the total of the actuarially ascertained liabilities.
16. From the actuarial valuation itself and from the explanation furnished by the Actuary he knew that the figure shown as the fund at 1st January 1925, was the actuarially ascertained liabilities in India. He also knew the subsequent history of the company because he had the actuarial reports and the quinquennial valuations. If he required any further information, as for example, the proportion of Indian premium income into world premium income he had it at his disposal. It seems to me that if what the Income-tax Officer had before him cannot be described as more reliable data within the meaning of the rule no life insurance company can ever be in a position to furnish more reliable data and, in consequence, the Indian branch of every non-resident life insurance company will have to be assessed under what I may call the proportionate method prescribed by, the rule. That is clearly not the intention; of the rule, because life insurance companies are expressly mentioned as a class of insurance companies whose total income can only be fixed by the proportionate method in the absence of more reliable data. In these circumstances what the Income-tax Officer should have done was to ask for further information if he needed it, and if he thought that the starting figure was hot justified, he should have worked on the basis of an increased starting figure for the fund and he could have decided fairly accurately, on the materials before him, what that figure should have been.