MEMORANDUM FINDINGS OF FACT AND OPINION
DAWSON, Judge: This case was assigned to and heard by Special Trial Judge Darrell D. Hallett pursuant to section 7456(c), Internal Revenue Code of 1954, 1 and Rules 180 and 181 of the Tax Court Rules of Practice and Procedure.2 The Court agrees with and adopts his opinion which is set forth below.
OPINION OF THE SPECIAL TRIAL JUDGE
HALLETT, Special Trial Judge: Respondent determined deficiencies in petitioner's Federal income tax and additions to tax as follows:
| | Additions to Tax |
| | Section | Section | Section |
| Taxable Year | Deficiency | 6651(a)(1) | 6651(a)(2) | 6653(a) |
| 1975 | $ 2,696.69 | 0 | 0 | 0 |
| 1976 | 2,529.97 | $ 74.08 | $ 8.23 | $ 190.75 |
After concessions, the remaining issues for decision are (1) What was the amount of tokes received by petitioner during the years 1975 and 1976 in connection with petitioner's employment as a "21" dealer at the MGM Grand casino 2a in Las Vegas; (2) whether petitioner is liable for the section 6653(a) addition to tax for 1976; and (3) whether petitioner is entitled to a medical expense deduction with respect to the cost of installing a therapy spa in his home.
This case was tried with 12 other cases which were consolidated for purposes of trial only and which involve similar issues.
FINDINGS OF FACT
Some of the facts have been stipulated and are found accordingly.
Petitioner Sidney Keen was a resident of Las Vegas, Nevada, at the time the petition was filed. He worked as a "21" dealer at the Las Vegas MGM Grand casino during the calendar years 1975 and 1976. He worked approximately 2,178 hours during the year 1975, and 2,223 hours during 1976. These hours included regular, overtime, and sick pay hours. Petitioner normally worked an eight-hour shift.
The MGM Grand is located on the corner of the Las Vegas Strip. It is adjacent to Caesar's Palace, the Dunes, and the Flamingo Hilton Hotel. During the years in question, the MGM casino was almost as large as a football field. It had 930 slot machines, 10 crap tables, 6 roulette wheels, 2 baccarat tables, 3 big six wheels, and a 200-seat keno lounge with closed-circuit TV to other areas. The casino also had a 16-table poker parlor and 4 casino bars.
In addition, the casino had 60 "21" tables, each with a seven-player capacity. Approximately 200 dealers were employed during a 24-hour shift to service these tables (not all tables would be in operation at any particular time). Approximately 80 percent of the "21" tables in operation had a $ 2 minimum betting limit. Other tables had higher limits, such as $ 5 and $ 25.
Petitioner and the other "21" dealers spent their shifts, except for brief relief periods, playing "21" as the dealers with patrons of the casino. There was a constant turnover of patrons during each shift. The number of players at a table varied throughout the shifts, depending upon such things as the time of day or night, the season of the year, and the relationship to the beginning or ending of a show being held at the MGM Grand.
Some patrons, but by no means all or necessarily even a majority, left a tip, which is referred to in the industry as a "toke", for the dealer during or at the completion of the patron's play. Tokes were received either directly from the patron in the form of cash or chips, or they were received as a consequence of a winning bet placed for the dealer by the player. Tokes received as a consequence of a winning bet resulted from the player's stating when the bet was placed that the winnings, if any, were "for the dealer." If the bet was lost, then the dealer got nothing. If the bet was won, he usually, but not always, received the winnings.
The "21" dealers were required to place all of the tokes received by them in a box kept near the tables. These boxes containing the tokes were periodically picked up by a designated individual and taken to a central place in the casino. The total tokes collected during each 24-hour shift were counted by designated dealers serving as an informal group referred to as the "toke committee." The toke committee obtained a dealer sign-in sheet, which each dealer working during the shift was required to sign, and determined by reference to that sheet the number of dealers working during the shift. Committee members divided the total number of tokes equally among all of the dealers, except that each dealer working overtime received an additional one-eighth share for each hour of overtime worked, and members of the committee would often receive an extra $ 5 or $ 10 per day for their work on the committee. The result of the count and distribution normally was written on the back of the sign-in sheet and was provided to an employee working in the casino cage. This employee checked the count of the total tokes, resolved any discrepancies, and placed each dealer's determined share in an envelope. The dealers would then pick up the envelope at the main bank cashier's window and sign a payroll sheet signifying receipt of the tokes. Tokes were always whole dollar amounts, i.e., no small change was ever included in the envelopes.
The records reflecting the daily determination and distribution of tokes for the years 1975 and 1976 were requested by respondent's agents by summonses served upon responsible MGM officials. The records were not available at the time the summonses were served. The records were destroyed by an MGM employee as a matter of convenience shortly after the daily distribution was made.
On February 11, 1975, an opinion was issued by the United States District Court for the District of Nevada in Olk v. United States, 388 F. Supp. 1108 (D. Nev. 1975), which held that tokes received by casino dealers were nontaxable gifts for Federal income tax purposes. This opinion was well publicized and well known among the casino employees in the State of Nevada. The decision was appealed to the United States Court of Appeals for the Ninth Circuit. By opinion issued on June 1, 1976, Olk v. United States, 536 F.2d 876 (9th Cir.1976), the Court of Appeals reversed the District Court and held that tokes are taxable income to dealers. Plaintiff's motion for rehearing in the Olk case was denied by the Ninth Circuit on July 14, 1976, and certiorari was denied by the Supreme Court of the United States, 429 U.S. 920 (1976).
In 1977, respondent's agents in the State of Nevada undertook a project to determine the toke income of MGM "21" dealers for the years 1975 and 1976. In this respect, respondent's agent Joe Page, the project coordinator, arranged for approximately 300 returns covering the years 1975 and 1976 and involving individuals who were dealers at the MGM Grand during those years to be selected for examination. Of the 300 returns, approximately 30 were selected for detailed analysis. Mr. Page made some effort to obtain information from the taxpayers involved and from available public records concerning the income and expenditures of the individuals in this group. After some analysis of public records information and personally meeting with approximately 23 or 24 of the group, Mr. Page concluded that he could "complete" a detailed financial analysis of six out of the group. This detailed financial analysis consisted of what is commonly referred to as a "source and applications of funds analysis."
Source and applications of funds analyses considered complete by Mr. Page related to three of the petitioners in the cases consolidated with this case for trial, Mr. Keen, Mr. Chappell, and Mr. Kozloff, and to three other taxpayers whose cases were not involved in these consolidated proceedings. However, the results of the source and applications of funds analyses with respect to two of the non-petitioning taxpayers, as well as Mr. Page's work-sheets concerning these two taxpayers, were admitted into evidence with deletions to protect the confidentiality and identity of these taxpayers. These taxpayers are referred to in the record as taxpayer "F" and taxpayer "A". 3
In connection with the source and applications of funds analyses for the five dealers referred to above, the total hours worked by each dealer during the year 1975 were adjusted by a reduction of 5 percent to allow for shifts for which the dealer may have been paid a salary but received no tips because of sick leave or lack of business, and to allow for possible error in the records ofthe MGM Grand. For the year 1976, respondent made an additional 10 percent reduction in hours to account for the culinary strike that year and a change in the employer's method of accounting for holidays worked. The resulting figure, divided by eight, is referred to as "adjusted shifts."
Respondent determined the total sources of funds and the total applications of funds for each of the five taxpayers for the year 1975. The excess of total applications of funds over total sources of funds for each dealer was determined by respondent to represent the dealer's taxable tokes for the year 1975. That figure, divided by the number of the dealer's adjusted shifts for the year, produced respondent's determination as to the tokes received per adjusted shift. The following is a summary of respondent's determination in this regard:
| Unaccounted | Adjusted | Rate Per Adjusted |
| Taxpayer | Funds (Tokes) | Shifts | Shifts |
| Kozloff | $ 14,888.68 | 252.58 | $ 58.95 |
| Chappell | 10,150.52 | 248.42 | 40.86 |
| Keen | 10,518.83 | 265.29 | 39.65 |
| "F" | 6,601.47 | 261.01 | 25.29 |
| "A" | 9,376.58 | 254.60 | 36.83 |
Based upon the results of the source and applications of funds analyses and other information respondent determined that petitioner and all other MGM dealers whose returns were examined for the years 1975 and 1976 received tokes for these years totaling $ 44 per adjusted shift.
Petitioner Sidney Keen kept no records of his toke income during the years 1975 and 1976. Petitioner declared tokes of $ 497 to his employer, MGM Grand during 1975 and reported no additional amount of tokes for that year on his 1975 return. During 1976 petitioner reported tokes of $ 1,974 to his employer, and reported no additional amount of toke income on his 1976 return.
Petitioner's adjusted shifts for 1975 and 1976, determined by making the reductions to total shifts referred to above, are 265.29 and 241.61, respectively.
Respondent's source and applications of funds analysis for Mr. Keen included as an application of funds $ 3,389.68 for food. This figure was an estimate by respondent based upon the United States Bureau of Labor Statistics estimates as to the national average cost of food for a family of three individuals. (Mr. Keen is married and has one child). This figure must be adjusted, however, to reflect the fact that Mr. Keen, unlike the "average" breadwinner presumably included in the Bureau of Labor Statistics estimate, was provided two meals per shift by his employer at no cost. A reasonable adjustment for this item is $ 5 per shift or $ 1,395. Respondent also included as an application of funds in Mr. Keen's analysis an item for "entertainment" of $ 886.42. This figure too was based upon the Bureau of Labor Statistics estimate. However, respondent included in Mr. Keen's applications of funds checks totaling $ 633.45 which were paid by Mr. Keen to Western Airlines in connection with vacations taken during 1975. To this extent, respondent's determination involves a duplication, and the applications of funds determined by respondent should be reduced by $ 633.45. Likewise, respondent based Mr. Keen's automobile expenses on estimates, and arrived at a figure of $ 1,212.78. However, included in numerous checks used by respondent in determining the total applications of funds for Mr. Keen was $ 743.12 in checks representing automobile expenses. A further adjustment should be made to Mr. Keen's source and applications of funds by reducing applications of funds by $ 743.12 to account for this item.
Mr. Keen testified that he had $ 3,000 cash on hand at the beginning of 1975, and that this "cash hoard" was used to purchase a therapy spa, the cost of which was included in respondent's application of funds. The evidence does not establish, however, that Mr. Keen's "cash hoard" was any more or less at the end of 1975 than it was at the beginning of 1975.
Adjusting respondent's source and applications of funds analysis for Mr. Keen for food, entertainment, and automobile expenses, as referred to above, results in an adjusted shift toke rate of $ 31.07. 4
Respondent's source and applications of funds for Mr. Chappell must also be adjusted based upon the evidence adduced at trial. In this regard, the evidence establishes that Mr. Chappell received a veteran's disability pension of $ 80 per month, totaling $ 960 for the year 1975. This amount was not included by respondent as a source of funds of Mr. Chappell. Mr. Chappell testified at trial that he believed he and his wife received cash gifts from his mother during 1975 totaling $ 3,000. However, a supplemental stipulation and Mr. Chappell's mother's Federal gift tax returns disclose that there were substantial cash gifts to Mr. Chappell and his wife during 1976, but no gifts during 1975. Adjustment of Mr. Chappell's source and applications of funds analysis for the disability pension results in an indicated toke rate per adjusted shift for Mr. Chappell of $ 36.99.
Respondent's source and applications of funds analysis for Mr. Kozloff should be adjusted with respect to total sources to reflect an addition of at least $ 3,400 for cash on hand as of January 1, 1975. This sum represents a January 3, 1975, currency deposit to a bank account of Mr. Kozloff as well as two other extraordinary currency deposits made by Mr. Kozloff in April 1975. Mr. Kozloff's source and applications of funds analysis, as determined by the respondent, should be further adjusted by at least $ 1,329 because the Bureau of Labor Statistics estimate for food expenditures used by respondent does not reflect that Mr. Kozloff, like Mr. Keen, was provided two free meals per shift by his employer. These two adjustments to respondent's source and applications of funds analysis result in an indicated toke rate per shift of $ 40.22 for Mr. Kozloff.
The source and applications of funds analysis for taxpayer "F" includes as an application of funds for food and entertainment only actual check or documented cash expenditures by the taxpayer during the year. No amounts have been estimated for these amounts or other expenditures.
The source and applications of funds analysis for taxpayer "A" includes an estimate for food based upon the Bureau of Labor Statistics figures. Therefore, the same adjustment as was made in the cases of Mr. Keen and Mr. Kozloff to reflect the fact that taxpyer "A" did not expend money for two of his three meals during working days is necessary. This adjustment amounts to a reduction in applications of funds in the amount of $ 1,272.50, which results in a revised indicated rate of tokes per adjusted shift of $ 31.78 for taxpayer "A".
Mr. Frank Martins was a "21" dealer at the MGM Grand during the years 1975 and 1976. For the year 1976, he reported tokes on his 1976 return amounting to $ 40 per day. Mr. Martins kept no records as to his actual tips for the year 1976, and the amount he reported on his 1976 return was based upon an estimate made at the time the return was prepared. Mr. Martins' accountant advised him in connection with the preparation of the 1976 return to report tokes of $ 40 per day.
During 1975, petitioner paid $ 3,400 for the installation of a "health spa" at his home. The health spa was installed as a result of advice from a local doctor that it would be beneficial in relieving pain of severe multiple joint arthritis suffered by petitioner's wife. Installation of the health spa required removal of landscaping which had been acquired at a total cost of approximately $ 1,900.
ULTIMATE FINDING OF FACT
Petitioner's toke income for the years 1975 and 1976 was $ 32 per adjusted shift.
OPINION
The issue for decision here is purely a factual one, namely, what was the amount of tokes received by petitioner during 1975 and 1976. In view of the Ninth Circuit's decision in Olk v. United States, 536 F.2d 876 (9th Cir. 1976), there is now no question that tokes are taxable income. 5
Several material facts are undisputed. Petitioner actually received tokes throughout the tax years 1975 and 1976. He kept no contemporaneous records reflecting the amount of these tokes, nor were records maintained or preserved by petitioner's employer, MGM Grand, showing the actual amount of tokes distributed to petitioner and the other "21" dealers.
In these circumstances, there is no question that the respondent can resort to indirect methods of proof to reconstruct the amount of petitioner's unreported income. Mendelson v. Commissioner, 305 F.2d 519 (7th Cir. 1962), affg. a Memorandum Opinion of this Court, cert. denied 371 U.S. 877 (1962); see section 446(b). Further, the source and applications of funds method of income reconstruction has long been accepted by this Court. Vassallo v. Commissioner, 23 T.C. 656 (1955). The method is based on the assumption that the amount by which the taxpayer's application of funds during a period exceeds his known sources of funds for the same period is taxable income. Of course, just as in any case involving an indirect method of proof, it is open to petitioner to point out areas or specific instances in which the method used by the respondent failed to reflect true income. See Meneguzzo v. Commissioner, 43 T.C. 824 (1965). In particular, the respondent's source and applications of funds analysis should be adjustd where evidence is submitted showing that it does not reflect as sources of income nontaxable items, such as funds accumulated at the beginning of the year, and used during the year, and/or amounts included as applications of funds which do not truly reflect expenditures made by a taxpayer during the year.
Petitioner contends, however, that the evidence concerning the source and applications of funds analyses of other taxpayers cannot be used as a basis to reconstruct petitioner's toke income. We disagree. There is clearly an element of commonality among petitioner and the other MGM "21" dealers who worked on a full-time basis during the years in question which makes evidence as to toke income received by one dealer relevant in determining the income received by others. This is so because the evidence is clear that tokes were, throughout both tax years in question, distributed evenly among all the dealers working during a 24-hour shift. Accordingly, there is not the likelihood that each dealer's daily tokes would vary substantially, depending upon such things as the location of a dealer's table assignment, the 8-hour period during which the dealer worked within a 24-hour shift, and the whim of the individual player the dealer happened to draw. In addition, this petitioner and others involved in the cases consolidated with petitioner's case for trial did not contend or testify that the total tokes received for the full year 1975 and 1976 varied significantly among individual dealers, because of different days on and off duty. Indeed, the evidence is virtually undisputed that whatever the toke income of petitioner should be, the amount should be substantially the same for the other dealers.
This being the case, the evidence as to toke income received by other dealers, consisting of the source of applications of funds analyses of those individuals, is relevant and was properly utilized by the respondent in determining petitioner's toke income.
In theory, a source and applications of funds analysis should show precisely the amount of a taxpayer's unreported income. In fact, since it is almost always necessarily based upon estimates and assumptions, it is at best a reasonable indication of the amount of unreported income. In these cases the source and applications of funds analyses do no more than indicate a range within which the average amount of toke income should lie. In this respect, respondent's calculations for the five individuals, unadjusted for items in controversy, show the range from a low of $ 25.29 to a high of $ 58.95. This is so even though the evidence is virtually uncontradicted that these individuals should have approximately the same toke income for the year 1975.
Because of these factors, we believe that undue emphasis should not be placed upon the results of any one particular source and applications of funds analysis. We have adjusted each individual analysis for items we believe clearly warrant adjustment. After making these adjustments, the analyses indicate the following:
| Taxpayer | Rate |
| Kozloff | $ 40.22 |
| Chappell | $ 36.99 |
| Keen | $ 31.07 |
| "F" | $ 25.29 |
| "A" | $ 31.78 |
In arriving at our conclusion as to the appropriate rate for petitioner, we have considered both the range of rates indicated by the five analyses, and the relative weight which should be afforded each individual analysis. For example, we believe the analysis of Mr. Kozloff should be afforded relatively less weight than that of the other individuals, because of the considerable uncertainty that exists in determining the amount of increase or decrease in cash accumulations by Mr. Kozloff during the year, as well as in determining his cash expenditures.
Further, we reject respondent's argument on brief that the analyses regarding taxpayers "F" and "A" should be afforded "little weight," apparently on the grounds these two individuals were not called to testify at trial. We find this argument somewhat surprising in view of the fact that the results of respondent's source and applications of funds analyses regarding these two taxpayers, as well as Mr. Page's detailed workpapers supporting the results, were introduced into evidence by respondent over petitioner's objection. We overruled petitioner's objection to the admission of these analyses based upon Rule 803(8c), Federal Rules of Evidence, which permits admission of a government investigative report, even though it constitutes hearsay, unless the Court finds the report lacks "trustworthiness". Respondent not only vouched for the trustworthiness of the reports regarding taxpayers "F" and "A", but introduced no evidence regarding adjustments, if any, that should be made to Mr. Page's computations for these taxpayers.
We have also considered the testimony of Mr. Martins, who was a "21" dealer throughout the year and who voluntarily reported toke income based upon $ 40 per shift. However, we do not believe that his testimony alone warrants a finding that petitioner's toke income for the year 1976 was at least $ 40 per shift. We must consider that Mr. Martins, like petitioner, kept no contemporaneous records of his income during the year 1976, and he based the amount reported solely upon an estimate arrived at between him and his accountant. Simply because the estimate favors the respondent's determination in this case more than it does petitioner's contention does not mean it should be accepted as any better evidence that it is, namely, an estimate.
Finally, we have considered respondent's determination and our opinion upholding that determination with respect to the toke income of dealers at the Las Vegas Hilton for the years 1972 and 1973. See Williams v. Commissioner, T.C. Memo. 1980-494. Because respondent determined the same rate per shift ($ 44) for the dealers involved in this case and for the dealers in the Williams case, and based the determination in this case at least in part upon the determination with respect to the Hilton dealers, particular emphasis at trial was placed by the parties upon a comparison of the Hilton and the MGM Grand. The evidence in this case establishes that the average amount of tokes received by the dealers at the Hilton was greater than that received by the dealers at the MGM Grand. This is because the MGM Grand has a substantially higher percentage of "low stake" tables than does the Hilton, and has significantly more tables and more dealers who share in the toke pool than the Hilton does. Likewise, respondent relied upon the determination of dealer's toke income in the Golden Nugget for the years 1972 and 1973, which was upheld by our opinion in Hannifin v. Commissioner, T.C. Memo. 1980-482. It should be noted that respondent's determination and our opinion in Hannifin were based upon actual contemporaneous records of toke income for the years involved (covering approximately six months of each year). The determined toke income involving the dealers in the Golden Nugget ranged from $ 17 to $ 21. However, the evidence in this case establishes that the toke income of the Golden Nugget dealers was less than that of MGM Grand dealers, largely because of the differences in clientele and betting practices.
We have considered the petitioner's toke income for each year 1975 and 1976. We conclude that the evidence in this case establishes that the amount of the toke income did not significantly vary between these two years.
Considering the evidence as a whole, we conclude and have found as an ultimate fact that petitioner's toke income was $ 32 per adjusted shift.
Respondent asserted the addition to tax provided by section 6653(a) for the year 1976, but not for 1975. Respondent's failure to assert the addition to tax for 1975 recognizes the uncertainty in the law that existed during that year as a result of the decision in Olk v. United States, 388 F.Supp. 1108 (D. Nev. 1975). We believe substantially the same uncertainty existed during 1976 and this uncertainty warrants the conclusion that petitioner was not negligent nor did he intentionally disregard the rules and regulations. We recognize that generally the substantial under-reporting of tip income, where the taxpayer has kept no records or failed to produce accurate records, has been held to justify the imposition of the addition to tax. See, e.g., Zibilich v. Commissioner, T.C. Memo. 1972-92. However, the unique circumstances of this case show that during much of the tax year in question petitioner may have in good faith believed that the law did not require him to report tokes for tax purposes. Therefore, although the uncertainty was resolved by the time the 1976 return was due to be filed, there is justifiable reason for failing to have records upon which to base the amount reported on that return.
Finally, we must consider petitioner's claim for a medical expense deduction of $ 3,440 with respect to the installation of a health spa at his home during 1975. A capital expenditure for an improvement to a taxpayer's residence may qualify as a medical expense deduction if the expenditure was for the primary purpose of and directly related to the medical care of the taxpayer (or his spouse) and dependents. However, the expenditure may only qualify for deduction to the extent it exceeds the increase in value of the taxpayer's residence attributable to the capital improvement. Section 1.213-1(e)(1)(iii), Income Tax Regs.; Haines v. Commissioner, 71 T.C. 644 (1979); Ferris v. Commissioner, 582 F.2d 1112 (7th Cir. 1978), reversing and remanding a Memorandum Opinion of this Court.
We are satisfied here that petitioner had the health spa installed primarily on the doctor's advice and to alleviate his wife's medical condition. We also conclude that the expenditure was directly related to his wife's medical condition, and was not lavish nor of a luxury nature. Cf. Ferris v. Commissioner, supra.However, in order to qualify the full expenditure for deduction, petitioner must also establish that the improvement did not increase the fair market value of the residence. Petitioner argues that it did not, particularly in view of the cost of landscaping which had been acquired within a relatively short period of time prior to installation of the health spa and which had to be removed to make room for the spa. However, we conclude that the health spa added value to petitioner's residence beyond the value of landscaping removed, and that petitioner's deduction should therefore be limited to $ 1,500.
Decision will be entered under Rule 155.