|
Citation: 2004TCC448
|
|
Date: 20040618
|
|
Docket: 2000-3460(GST)G
2000-3463(GST)G
|
|
BETWEEN:
|
|
27 CARDIGAN INC. and 33 CARDIGAN INC.,
|
|
Appellants,
|
|
and
|
|
|
|
HER MAJESTY THE QUEEN,
|
|
Respondent.
|
REASONS FOR JUDGMENT
Bowie J.
[1] These two appeals concern the
amount of goods and services tax (GST) that is exigible under the
provisions of the Excise Tax Act, Part IX, (the
Act) on the deemed self-supply of condominium units in an
apartment building to the two Appellant corporations. The parties
have entered into an agreement as to many of the facts of the
case, and they agree that the issue to be decided is the fair
market values of the units at the dates of the self-supply. There
is also a subsidiary issue as to penalties. I shall reproduce
their agreement in its entirety.
The Reassessments
1. The
Appellants appeal from Notices of Reassessment each dated
February 23, 2000 for the period October 1, 1994 to September 30,
1997 (the "Reassessments"), which were issued by the
Minister of National Revenue (the "Minister").
Background
2. Each of the
Appellants is a corporation which was incorporated under the laws
of the Province of Ontario, the principal place of business of
which is c/o Cornerstone Properties Inc., P.O. Box 3117, Terminal
"A", London, Ontario, N6A 4J4 and the fiscal period and
the taxation year of which at all material times ended on
September 30.
3. By
agreement of purchase and sale dated September 1, 1994 (the
"Agreement") the Appellants purchased from Richmill
Development Corporation ("Richmill"), a corporation
related to the Appellants, designated individually-titled units
(the "Units") and their appurtenant interests in a
187-unit condominium project that Richmill was constructing at
675 Richmond Street, London, Ontario, (the "Richmond
Property").
4. As
evidenced by Schedule 1.1 to the Agreement, entitled
"Details of Purchase Price and Allocation", 33 Cardigan
Inc. ("33 Cardigan") purchased 93 of the Units and 27
Cardigan Inc. ("27 Cardigan") purchased the remaining
94 Units in the Richmond Property for consideration totalling the
amount of $10,544,000, of which $5,262,481 (49.9% of the purchase
price) was paid by 33 Cardigan and $5,281,519 (50.1% of the
purchase price) was paid by 27 Cardigan.
5. The
Agreement provided that the transfer of possession of the
Richmond Property to the Appellants would be deemed to take
effect as at October 1, 1994 (the "Effective
Date") but that title to the Units would not be transferred
until a date after the registration of the Richmond Property as a
condominium project (the "Transfer Date")
6. Under the
Agreement Richmill was obliged, inter alia, to pay realty
taxes and maintain insurance coverage with respect to the
Richmond Property, to act as agent for the Appellants while
carrying out the completion of the proposed condominium project
and to account to the Appellants during the period between the
Effective Date and the Transfer Date.
7. As of the
Effective Date, the construction of the condominium project at
the Richmond Property was approximately 60% completed. The cost
to complete the condominium project was $7,253,000. The total
cost of the condominium project on the Richmond Property was
$17,797,000, of which 96.2%, namely, $17,120,714, relates to the
residential units.
8. On June 30,
1995 a preliminary draft appraisal report authored by
Tony Best of Lansink Best & McIver Limited was obtained
and used to estimate the value of the Units that were leased to
tenants after October 1, 1994.
9. Between
October 1, 1994 and September 30, 1997, the Units were leased to
individuals who were the first to occupy the Units.
10. Richmill estimated and
remitted GST on the Units in its reporting periods ending July
31, 1995 to October 31, 1996, inclusive, and in its reporting
period ending January 31, 1997, pursuant to subsection 191(1) of
the Excise Tax Act, R.S.C. 1985, c.E-15, as amended (the
"Act").
11. On June 23, 1998 the
Richmond Property was registered as condominium project M.C.C.
No. 388.
12. As the Appellants were
the beneficial owners of the Units at the time that they were
supplied, Richmill was not required to report GST pursuant to
subsection 191(1) of the Act nor was it entitled to claim
ITCs in respect of the Units. Therefore, by Notice of Assessment
No. 00000000395 dated November 24, 1998 issued to Richmill for
the period between October 1, 1994 and September 30, 1997 (the
"Richmill Assessment"), the Minster disallowed the
$596,167.30 of input tax credits claimed and refunded the
$416,167.30 of GST reported by Richmill in respect of the
Richmond Project. The Minister reassessed the Appellants to allow
them $596,167.30 of ITCs which was set off against the amount of
GST owing in respect of the Richmond Property.
13. The Appellant obtained
an appraisal report dated September 14, 1998 prepared by Lansink
Best & McIver Limited (the "Lansink Appraisal
Report") estimating the fair market value of the Richmond
Property as of August 1, 1995 to have been $9,850,000 (the
"Lansink Amount").
14. On the basis of an
appraisal report dated August 27, 1999 prepared by the London Tax
Services Office of the Department of National Revenue and
containing an estimate of the fair market value of each of the
Rental Units as of their initial dates of occupancy, the Minister
assumed in making the Reassessment that the fair market value of
all of the Units was the aggregate amount of $16,693,000 (the
"London T.S.O. Amount").
The Reassessments
15. By the Reassessments
the Minister assessed GST against the Appellants and penalty and
interest thereon.
16. The Appellants
objected to the Reassessments by Notices of Objection dated May
17, 2000.
17. The Minister confirmed
the Reassessment by Notice of Decision dated July 27, 2000.
Penalties
18. The Respondent
concedes that no penalties ought to have been assessed against
the Appellants for the period in which Richmill reported GST in
respect of the Richmond Project except to the extent of the
amount, if any, of GST underreported.
Issues in the Appeals
19. The remaining issues
to be decided in these appeals are:
(a) what the fair
market values of the Units were at the time of the "deemed
self-supply"; and
(b) in the event
that the London T.S.O. Amount is established to reflect properly
the fair market value of the Units at the relevant time, which
the Appellants do not admit but deny, whether the Appellants are
liable for the section 280 penalties on the amount of
underreported and unremitted GST.
20. The Appellants submit
that the fair market values of the Units at the time of the
"deemed self-supply" of the Units are as reflected in
the report of Ben Lansink dated March 17, 2003.
21. The Respondent submits
that the fair market values of the Units at the time of the
"deemed self-supply" of the Units are as reflected in
the report of Warren Shannon dated March 5, 2003.
[2] The London T.S.O. amount is said
in paragraph 14 to be $16,693,000. That is pleaded in paragraph
12 of each Notice of Appeal, and admitted at paragraph 2 of each
Reply. However, the actual amount is $16,963,000: See Exhibit
A-2, Tabs 8, 13 and 14.
[3] It is not in dispute that upon the
first occupation of each of the 187 condominium units by a
tenant there was a deemed self-supply of that unit from the
Appellant that owned it to itself, giving rise to the imposition
of GST at the rate of 7% based on the fair market value of the
unit. The tax is deemed to have been collected by the Appellant,
as a supplier, from itself as recipient of the supply.[1] As the various units
were first occupied at various dates between October 1, 1994 and
September 30, 1997, the determination of the aggregate liability
for tax imposed on each of the Appellants becomes more complex
than might at first blush appear. It is necessary to establish
the fair market value of each individual unit at the date of its
first occupation by a tenant. To this end, each party led the
evidence of a professional real estate appraiser. For the reasons
that follow, the evidence given by Mr. Ben Lansink for the
Appellants roved to be much more useful than that of Mr. Warren
Shannon for the Respondent.
[4] The building at 675 Richmond
Street is a 17-storey concrete and glass apartment building,
containing the 187 apartments that are the subject of these
appeals. There are 1, 2 and 3-bedroom apartments in 12 different
configurations, ranging in size from 918 to 1,764 square feet. In
addition, the building contains commercial, retail and office
space on the main floor, parking below grade, and it shares a
swimming pool with the adjacent building at 695 Richmond Street.
None of these amenities are included in the GST assessment under
appeal. They are relevant only to the extent that by their
existence they may contribute some additional value to the
apartments. The building was, of course, new at the relevant
time, and so in excellent condition. The construction was of high
quality. The building covers slightly less than 75% of the 0.413
hectare parcel of land on which it sits.
[5] Mr. Ben Lansink gave evidence for
the Appellants. He is a well-qualified appraiser, although his
qualifications and experience are mainly theoretical. He does not
appear to have recent experience in the marketplace, either as a
principal or as an agent for buyers or sellers. He has been
engaged in private practice as a fee appraiser since 1974, and
has been accredited by the Appraisal Institute of Canada since
1982. In that time, he has acted for a wide variety of both
government and private sector clients, including numerous
financial institutions, first nations, national and international
law firms and accounting firms. He has given evidence on a number
of occasions before the Supreme Court of Ontario and the Ontario
Municipal Board.
[6] Mr. Lansink acknowledged that this
case presents a novel problem for the appraiser, involving as it
does 187 separate appraisals of 187 separate condominium units,
each of which must be appraised as of the date that it was first
occupied. Those dates are scattered over the period between
November 1994 and September 1997. Mr. Lansink applied the direct
comparison approach in estimating the value of these 187 units.
To do so, he looked at six sales of units in a condominium
apartment building at 155 Kent Street that took place between
October 1994 and March 1995, six sales in a building at 600
Talbot Street between April 1994 and April 1995, and five sales
in 695 Richmond Street between October 1994 and February 1995. He
considered the sales at 695 Richmond to be the most
comparable, because that building is adjacent to 675 Richmond,
and in fact shares some amenities with it. 600 Talbot is in a
better location, and so he made some downward adjustment to those
sales. He was of the opinion, based on the average prices of
single dwellings in London and St. Thomas, that there had been no
appreciable change in values during the period from 1994 to 1997,
and he therefore made no adjustments for time. From this data he
estimated the value of apartments in the building at 675 Richmond
Street to have been in the range of $70.00 to $80.00 per square
foot throughout the period from November 1994 to September 1997,
from which he derived an estimated value of $75.00 per square
foot.
[7] In a table found on page 21 of his
report, Mr. Lansink then calculated the value for an apartment in
each of the 12 different configurations, simply by multiplying
the area of each, in square feet, by $75.00. That table goes on
to establish the aggregate value of the 187 units to be
$17,339,250, which Mr. Lansink then divides by 187,
estimating the average value of an apartment in the building to
be $92,723, which he rounds up to $95,000. The purpose of this
exercise seems to be to enable him to use that average apartment
value for purposes of the next part of his evidence.
[8] Mr. Lansink then went on to
hypothesize that the market for condominium apartments in central
London could not absorb 187 apartments at a rate greater than 30
per year. He therefore concluded that, even with what he called
aggressive marketing, they would not all be sold in less than six
years. He then proceeded to compute a discount from the average
value that he had previously established to be $95,000, designed
to take this into account. He also estimated what he described as
marketing expenses. These include such items as real estate
agent's commissions, conveyancing costs, management fees and
interest associated with the unsold units for an average of three
years, together with a further 10% of the aggregate of these
amounts for "contingencies". He then computed the present value
of the total proceeds of sale of the 187 apartment units over a
six-year period, net of the marketing expenses, to be $9,300,000
(rounded), or $40.23 per square foot. He proceeded from there to
his final 187 estimates of value by the comparative approach
by multiplying the area, in square feet, of each of the 187
apartments by $40.23. The results are set out in a table on pages
24 to 29 of his report.
[9] Mr. Lansink also looked at value
using the income approach, not, he said, in order to estimate the
fair market value of the apartments, but to satisfy himself as to
the highest and best use. At page 3 of his report he says this
about highest and best use:
As Though Vacant: Construction and use of a building
housing residential dwelling units together with complimentary
retail, commercial and office uses and other site improvements as
legally permitted by land use controls.
As Improved: Continuation of the existing use, as a
building that houses 187 residential rental dwelling units,
basement parking and main floor retail, commercial and office
space. The value conclusion, assuming one owner of each of the
Units, is estimated at $10,557,938 if each unit is rented. The
value is $9,300,000, assuming there are 187 individual owners of
the Units. Therefore, the highest and best use of each Unit is
rental to a tenant for residential purposes.
I do not understand why he concerned himself with the highest
and best use of the land as if it were vacant, or that of the
entire building. He had earlier made it clear that he appreciated
that the subject of the appraisal was not the land if vacant, or
the building, but the individual residential units in the
building. His opinion of value based upon the income approach, he
said, was $44.00 to $46.00 per square foot. His opinion as to
value based upon direct comparison to nearby sales, before
discounting for the supposition that there would be 187 units
placed on the market at one time, was $70.00 to $80.00 per square
foot. I do not understand how his opinion based on the income
approach could lead him to believe that renting one of these
units to tenants was a higher and better use than selling it to a
person who wished to reside in it. The only explanation is that
instead of looking at the units individually, as he had
acknowledged should be done, he estimated the value of the
residential part of the building as a whole, and then ascribed
that to the various apartments on a per square foot basis. His
evidence did not make it clear that he had not included some
expenses in his analysis that should properly have been excluded
as applicable to the non-residential portions of the
building. Nor does he take into account that the owner of a
single condominium unit in the building would be required to pay
substantial condominium fees, including reserves of the kind that
he includes in his computation of the expenses. I have little
doubt that these factors have the effect of considerably
inflating the estimate of expenses properly chargeable against
the income from an individual unit.
[10] I also do not accept that for the
purpose of applying the income approach to one condominium
apartment, it is appropriate to derive a capitalization rate from
the sales of entire apartment buildings that changed hands at
prices between $1.75 million and $6.15 million. Such an
investment cannot properly be compared to the purchase of one
apartment for a price somewhere between $66,000 and $195,000.[2] As Mr. Lansink
said more than once in his evidence that he did not rely on the
result of his income approach to value except for the purpose of
establishing the highest and best use, I do not intend to
consider it further. I have no doubt that the highest and best
use of each of these apartment units is to be used as a dwelling
by the owner, or to be rented by the owner to a residential
tenant.
[11] I also have no doubt that the most
appropriate method of valuation of the units is by direct
comparison to sales of similar units in the same neighbourhood
during the period between November 1994 and September 1997, at
which time values of such units were quite stable. Indeed, Mr.
Lansink made that very clear at the conclusion of his evidence in
chief.
Q. Would you just
explain why you favoured the direct comparison approach to reach
your final conclusion over the income approach? I know the
differences aren't great but I would just like to know the
reason why you favoured one over the other.
A. Real estate
appraisers like to joke with the statement that three approaches
to value are market, market and market. And I think it's
true, the market value is really the measuring of the individual
actions of people in the marketplace. and the income approach in
this particular example is really not applicable because it's
- you would not use the income approach on one unit. Nobody would
buy a single unit for the income it produces. You would only buy
the building where the economies of scale would come into play
such as property management and so on. But to buy one unit for
the income it produces, you would be making very little return on
your money.
Q. Did you make the
income approach as kind of a test check?
A. It was a test
check and it was also for me to be able to conclude my highest
and best use.
My highest and best use in my opinion is not selling the 187
units. The value in this building is greater if it's sold to
one individual who then collects the rent and manages the
building. That is the highest and best use and that's the
highest value. That's really the purpose of the income
approach, was to illustrate that.
Q. But on the other
hand it's not the value of each individual unit.
A. No
My job, according to your instructions and the way I've
written my report, is 187 dwellings. It's individual. I'm
doing 187 appraisals and I'm showing 187 individual values.
And if I'm not, if it's just one deed, then I would
re-look at that value. I would say the value is closer to my
income approach as opposed to the direct conversion approach. I
would rethink that if you were to instruct me differently.
I accept that the value of a condominium apartment in the 675
Richmond Street building, during the period from November 1994 to
September 1997, if only one were on the market, would have been
about $75.00 per square foot, as Mr. Lansink's evidence
suggests.
[12] I turn now to the discount from that
figure that Mr. Lansink applied. For the reasons that follow, I
do not accept his opinion that a discount of about 45% should be
applied to that estimate of value because there are 187 units in
the building.
[13] First, there is the question whether
any discount should be applied at all. Mr. Lansink's theory is
that for the purpose of each of his 187 appraisals, he must
assume not only a hypothetical sale by a willing vendor to a
willing purchaser of the particular unit being appraised, but
also that this hypothetical sale takes place not in the London
market as it existed on the date at which each unit is being
appraised, but as it would have existed if all 186 other units
had been placed on the market that day. The Excise Tax Act
imposes a tax on hypothetical transactions that the statute deems
to have taken place, although they did not in fact take place at
all. Each deemed supply and receipt of a unit is a separate one,
and the determination of the fair market value of the unit on
which it operates is unaffected by any other hypothetical
transactions that may be deemed to take place at or about the
same time. That is because those deemed sales and receipts of
supplies never take place, and indeed the units are not offered
for sale, and so they are not capable of affecting the
market that exists in reality. It is in the context of that real
market that value must be determined.
[14] I emphasize that the present case bears
no similarity to cases such as Henderson v. M.N.R.,[3] where the
Minister's assessor discounted a block of shares from $10.78
to $8.00 to account for the effect on the market that would
result from offering a very large holding for sale at one time.
In that case, the statutory mandate was to ascertain the fair
market value of one large block of shares on one specific date.
Nor does it resemble Wosk's Ltd. v. British Columbia
(Assessor of Area No. 16 - Chilliwack).[4] That case arose out of the
valuation for municipal taxation purposes of 296 serviced lots in
the Village of Harrison Hot Springs, B.C., a village of only 600
persons living in 233 homes. There a discount from value based on
the absence of potential immediate buyers for all 296 lots was
approved by the Court of Appeal. Both of these cases, however,
deal with the actual market conditions existing on the date that
all the shares, and all the lots, were to be valued. In
Henderson, the appraisal was of one block of shares in its
entirety. In Wosk's case, the Assessment Appeal Board
effectively discounted the value of the lots from the estimated
value of a single lot time 296 by finding that the highest and
best use was to be purchased in bulk for future resale. Although
the lots were not being actively marketed, they were all
available for sale. In the case before me none of the condominium
units were for sale at all. My task is to determine as of each of
the many valuation dates between November 1994 and September 1997
what was the highest price that a willing vendor of the property
could expect to receive from a willing purchaser, both well
informed and acting at arm's length, on the basis of the
market existing at that time, not on the basis of a hypothetical
market that did not, and never would, exist. The deeming
provision in subsection 191(2) of the Act operates
independently in respect of each unit that is deemed to be
supplied and deemed to be acquired by the builder. It does not
deem every other unit in the building to be for sale on the
market at the same time.
[15] Even if I were to accept Mr. Lansink's
theory that each appraisal must proceed on the assumption that
all 187 units would be exposed for sale throughout the period
from 1994 to 1997, I would not accept his methodology as being a
reasonable one. There are a number of aspects to it that simply
cannot be justified. First, I do not accept his opinion that only
30 units could be sold each year. When asked about the study of
the market on which he based this opinion, he said that it was
based on resales in the three buildings that he considered to be
comparable, and on which he had formed his initial estimate of
value. His evidence was that in 1995 there were a total of 54
units sold in those three buildings.[5] The charts at pages 19 and 20 of his
report show the sales in those buildings on which he based his
estimate of value. There are six sales at 155 Kent from October
1994 to March 1995, six sales in 600 Talbot between May 1994 and
February 1995, and five sales between October 1994 and February
1995 in 695 Richmond. These are resales, and so did not result
from aggressive marketing. In the Kent Street and Richmond Street
buildings only one of those units was listed for significantly
more than one month, and even it sold in less than two months.
The average exposure time for the Talbot Street building was
about 3.3 months, presumably because the prices in that building
were considerably higher. This evidence, limited in scope though
it is, suggests to me that the limit on the number of sales,
particularly in the two most comparable buildings, is not caused
by a lack of buyers but by a lack of units offered for sale. Mr.
Lansink was asked in his direct examination how he arrived at his
conclusion that six years would be required to sell 187 units.
His answer was "Experience, judgment". His reliance on the resale
evidence in cross-examination appears to have been an
afterthought. Although he was cross-examined at some length
on this point, he did not refer to any actual experience in
connection with the marketing of condominiums. I find this part
of his evidence to be at best speculative, and not at all
objective. To adopt it would offend the principle stated by
Mahoney J. as he then was, in National Systems of Baking of
Alberta Ltd. v. The Queen[6] and specifically approved by the Federal
Court of Appeal.[7]
The Court is not justified in jumping with an expert to a
conclusion that is sustained only by the evidence of his
expertise; it simply must have evidence as to facts so that it
can both understand and evaluate the process leading to the
conclusion and the validity of the conclusion itself.
[16] Mr. Lansink testified that the
discounting for volume should begin by removing more than one
third of the value of the units to account for what he called
"marketing expenses". I have described these above. On page 35 of
his report he quotes the definition of market value as it appears
in the Canadian Uniform Standards of Professional Appraisal
Practice published by the Appraisal Institute of Canada for the
guidance of its members.
Market Value is defined as "the estimated amount for which a
property should exchange on the date of valuation between a
willing buyer and a willing seller in an arms-length transaction
after proper marketing wherein the parties had each acted
knowledgeably, prudently and without compulsion."
This is not significantly different from the definition
generally accepted by Canadian courts. The amount for which a
property should exchange means the price at which it likely would
be bought and sold; it does not mean the net proceeds to the
seller after paying the costs of selling. Mr. Lansink sought to
justify his departure from the classic definition by likening his
approach to the development approach often applied to the
valuation of raw land. The very significant difference is that a
purchaser of a parcel of raw land will consider the costs
associated with developing it in formulating the price that he is
willing to pay. That makes the computation of both the costs of
subdividing and the probable proceeds of sales of lots a relevant
consideration when estimating the price at which the transaction
is likely to take place. In considering the price for which a
condominium residence will likely sell, no such considerations
are relevant. To put it another way, Mr. Lansink's task was to
estimate the price at which each unit was likely to sell on its
particular date of first occupancy. Instead, he set out to
estimate the present value of the likely net proceeds of sale of
all the units at unspecified dates in the future. I appreciate
that he did this in order to solve what he perceived to be a
difficult appraisal problem, but that gives him no mandate to
depart from the long established meaning of the statutory words
"... the fair market value of the unit at [the date of first
occupation]".
[17] A further inconsistency in the
discounting by Mr. Lansink is his failure to give credit for the
rent that would have been paid by the tenants of the unsold units
during his hypothetical six-year marketing period. He has been
careful to include all the costs that he could think of,
including the removal of tenants in order to be able to give
vacant possession when the time came. He gave no good reason why
the rents received during the period should not have been taken
into account as well. Nor did he include any increment in the
prices that would be achieved towards the end of his six-year
period. There was evidence that prices did not rise between 1994
and 1997, but the evidence suggests that they would have risen
somewhat after 1997.
[18] It is noteworthy, too, that Mr.
Lansink, although he discounted value by 45% on the basis of his
notion that he had to take into account a hypothetical glut on
the market, did not produce in his evidence any analysis of the
dates of first occupation of the units. Had he done so he would
have seen that July, August, September and October 1995 were the
only months in which more than 10 of the units became occupied
for the first time. If a glutting of the market were a relevant
concern for him to take into account, which I do not accept, then
the hypothetical glut was not nearly of the magnitude that he
apparently thought. Rather than the 187 units being
hypothetically exposed for sale in one day, as he seemed to
suppose, they would have come onto the market quite gradually
between November 1994 and June 1995, and again from November 1995
to September 1996. While I do not think it strictly relevant, I
have appended to these reasons as Appendix "A" a chart
showing the dates of first occupation of the 187 units
month-by-month. When seen in its true light the hypothetical
problem, if there were one at all, is much less than Mr Lansink
would have us believe.
[19] Indeed, it is only during the second
half of 1995 that more than 30 units per year were occupied for
the first time. Even on his evidence, the market could absorb all
the units that hypothetically were sold during all of 1994, the
first six months of 1995, all of 1996 and all of 1997. Mr.
Lansink failed to consider the possibility that a modest discount
from the $75.00 per square foot value otherwise ascertained would
entice sufficient buyers to the market. There is no reason to
believe that residential real estate is not a highly elastic
commodity.
[20] Mr. Warren Shannon was called by the
Respondent to give evidence as to his opinion of the value of the
187 condominium units that make up the residential portion of 675
Richmond Street. Mr. Shannon has been a member of the Appraisal
Institute of Canada for some 30 years. He was certified as a
rural appraiser initially, in 1988 he was accredited by the
Canadian Institute with the designation AACI, and he is presently
so certified until October 2007. He worked as an appraiser for
Central Mortgage and Housing Corporation from 1972 to 1977. Since
then he has worked for Revenue Canada (now CCRA), first in
Hamilton and St. Catharine's, then as a team leader in
southwestern Ontario, and most recently as Regional Manager in
that area. He has not, so far as I know, had any experience
buying and selling in the real estate market, either on his own
account or as an agent for others.
[21] Mr. Shannon first formed an opinion as
to value in relation to each of these apartment units in August
1999 when he was asked by the Chief of Appeals in the London Tax
Services office of Revenue Canada to appraise each residential
condominium unit in the building at 675 Richmond Street. At that
time, he prepared a written appraisal report that is in evidence
at Tab 8 of Exhibit A-1. In explaining his methodology, he said
at page 20:
An income approach to value was not undertaken as it is not
typically reflective of the motives of or value to owners wishing
to copy the units as their principal residence.
A cost approach was not undertaken as it is not within the
scope of this appraisal to value the complete building, only the
individual units at their date of occupancy.
The direct comparison approach is considered applicable as
there are numerous sales at the effective date of appraisal to
compare to the subject and estimate a market value.
In that regard various sales were reviewed and analysed in
comparison to the subject.
His reference to "the subject" was clearly meant to
refer to all 187 subjects that he was appraising. He then went on
to examine 43 sales of condominium apartments in the adjacent
building that he considered to be comparable. Of these, 11 took
place in 1995, 14 in 1996 and 18 in 1997. After making some
adjustments for the relative ages of the buildings, and for time,
he established values for various categories of apartments in 675
Richmond in each of the years 1995, 1996 and 1997 as follows:
|
|
1995
|
1996
|
1997
|
|
1 bedroom
|
$75,000
|
$78,000
|
$82,000
|
|
2 bedroom
|
94,000
|
98,000
|
102,000
|
|
3 bedroom (1,624 sq. ft.)
|
114,000
|
118,000
|
125,000
|
|
3 bedroom (1,764 sq. ft.)
|
123,000
|
129,000
|
135,000
|
He applied these value estimates to the various apartments,
and the aggregate of these opinions of value became what the
parties referred to in the pleadings and in the agreed facts as
the London T.S.O. Amount.[8]
[22] Exhibit R-1 at the trial is Mr.
Shannon's written appraisal report dated March 5, 2003 filed
to comply with Rule 145. In it he estimated value by the cost
approach, the income approach and the direct comparison approach.
Quite clearly, he understood when he embarked on this that his
task was to form 187 opinions of value; one for each apartment.
He had done that in 1999. Now, however, he chose to estimate the
value of the residential portion of the entire building by each
of these three methods, and then apportion his final estimate of
value for the building among the various apartments in proportion
to the relative percentages of the total rental revenue for the
building that each unit produced. In my view, this approach is
fundamentally flawed for a number of reasons. First, there is no
reason to believe that the value of the residential portion of
the building, undivided would be equal to the sum of the values
of the individual condominium units. No such assumption can be
made, because the highest and best use of the building undivided
is not necessarily the same as the highest and best use of each
single condominium unit. In 1999, he had described the highest
and best use as being "as an apartment condominium". In
2003, he described it as "Existing use 187 residential
units". Neither of these is a useful observation. However,
the fact that in 2003 he chose to value the entire residential
portion of the building by comparison to sales of apartment
buildings suggests to me that he was not appraising the
apartments in the same use then as he had previously. Second, the
dates at which values for the units must be established are
spread over the period beginning in December 1994 and ending in
August 1997. Mr. Shannon effectively ignored both these
problems. With respect to the dates, when cross-examined, he took
the position that values did not change significantly during this
period. This is completely at odds with his August 1999 opinion
where he stated that "Based on all the data however it is
considered that values increase 5% from 1995 to 1996 and 10% from
1995 to 1997".[9] In Exhibit R-1 he simply said:
As a result of this analysis, it is the appraiser's
opinion that the Fair Market Value of the subject property, as at
the effective dates of the Appraisal was:
$17,356,000
This Value relates to the residential portion of the property
only.
He acknowledged earlier in his report that the effective dates
referred to here are between August 1994 and November 1997, but
he simply proceeded as though there were no need to consider
trends in the market at all.
[23] Another serious flaw in Mr.
Shannon's evidence is that in the final analysis he relies
almost totally on the historical cost for his opinion as to
value. Although his evidence contains sections devoted to
estimating value by the income approach and by the comparative
approach, he explained that the latter two were relied on simply
as a check upon the value estimated by the cost approach. His
cost approach estimate is based upon two numbers; one is what he
called the building replacement cost new, and the other is the
estimate of the value of the site as if it were vacant. As the
building was new at the effective date of the appraisal, he did
not consider there to be any physical, functional or economic
depreciation. He derived the cost to build the residential part
of the building at the effective dates from the total historical
cost of the project, from which he subtracted the cost of the
land ($630,000) and the 3.8% of construction costs that he was
told was attributable to the non-residential part of the
building, producing a replacement cost of $16,514,654. This
number, of course, has nothing to do with any estimate of the
cost that would be incurred to reproduce the building at the date
on which it was completed, or any other date. It is essentially
the historical cost of a disastrous construction project, to
which he added his estimate of the current value of the land if
vacant. Mr. Trent Krauel testified at some length about the
construction project. He said that it was a financial disaster
for various reasons, with substantial cost overruns, delays, and
financing problems that drove the cost far beyond budget. His
evidence satisfied me beyond any doubt that this building was
completed only at a cost substantially beyond what it would have
been worth if offered on the market at its completion.
[24] Mr. Shannon said at page 13 of his
report, in relation to this replacement cost:
... These costs have been reviewed and compared to other
known costs and manuals. The costs are considered reasonable in
the subject market area at the effective date of appraisal.
When asked about this comparative examination on
cross-examination, Mr. Shannon was not able to give any
information at all about it. I have serious doubt that he
undertook any real review or comparison at all. His evidence on
this point cannot be reconciled with that of Mr. Krauel. I accept
Mr. Krauel's evidence that the project was, from the
builder's point of view, a financial disaster.
[25] All of this is simply to say that Mr.
Shannon did not apply a high level of expertise to his estimate
of value by the cost approach. But there is a more fundamental
problem with his opinion, and that is that it is contrary to
accepted appraisal practice to estimate value of a single
residential dwelling by the cost approach if there is a market
for the property being appraised, and comparable sales of similar
property in the vicinity can be found. Mr. Shannon accepted this
principle in giving his evidence, and he also accepted that in
the present case, there was a market for the apartments at the
relevant time. He went on to make an estimate of value by the
direct comparison method, as well as by the income approach. He
said, however, that these latter two were only intended to be a
check on the accuracy of his estimate by the cost approach. His
estimate by the direct comparison method however, was not based
on sales of comparable condominium units, but on sales of
buildings that he considered to be comparable to 675
Richmond.
[26] Mr. Shannon's estimates of value by
the income approach and by the direct comparison approach cannot
be considered reliable. The income approach requires the
appraiser to determine the net income producing potential of an
investment property and apply to that an appropriate
capitalization rate, thereby determining, at least in theory, how
much a prudent investor would be willing to pay for the property.
One could certainly quibble with the methodology by which Mr.
Shannon arrived at the net income that he used. He apparently did
not inquire into what inducements, if any, had been given to the
tenants on entering into their leases. Such inducements are
common in the industry, and certainly will have some effect on
the real income realized. Mr. Shannon simply worked from the
landlord's list of rents to be paid each month. His estimate
of the operating costs to be deducted is largely based on the
owner's cost experience for such things as utilities,
repairs, maintenance and insurance, and he allowed 4% of
effective gross income for management expense. He based this on
his discussions with property management firms in the area. The
greater difficulty, however, lies in his determination of the
capitalization rate to be applied to the net income. He based
this on two sales of apartment blocks, one consisting of three
condominium buildings in London held by one owner who leased the
apartments, and the other a complex in Kitchener, Ontario,
consisting of two 18-storey buildings. At the bottom of
page 15 of Exhibit R-1 he said this:
The capitalization rate was chosen from market sales. Those
sales indicate:
|
ADDRESS
|
DATE
|
PRICE
|
NET INCOME
|
CAP RATE
|
|
1669/70/71 Jalna
|
05/98
|
$15,250,000
|
$1,180,350
|
7.74
|
|
305-315 Margaret
Kitchener
|
03/97
|
$33,000,000
|
$2,435,400
|
8.17
|
From this data, he concluded that a capitalization rate of 8%
was appropriate. The capitalization rate produced by the
Kitchener property is in fact 7.38%, however, rather than 8.17%
as calculated by Mr. Shannon. This error has the effect of
decreasing the estimate of value by about $1 million, assuming
that but for the error he would have chosen a rate of 7.5% rather
than 8%. Such an error does not inspire confidence in Mr.
Shannon's opinion. Nor does the fact that he was willing to
base the capitalization rate on complexes significantly older
than the subject building, one of which is in a city
significantly closer to the Metropolitan Toronto Area without any
consideration of the effect of these factors.
[27] Mr. Shannon's estimate of value
based on the direct comparison method is equally flawed. He based
it on two sales of apartment buildings that he considered to be
appropriate comparables. One of these was the sale of
675 Richmond Street itself to Daniel Drimmer in July, 2002
at a price of $17,976,123 for the residential portion alone,
based on $96,129 per residential unit. The other was the sale of
a 142-unit building at 55 William Street in Waterloo, Ontario in
December, 2000 for $20,300,000. Both these sales took place
significantly later than last of the relevant dates for the
present case. It is not necessarily fatal that the sales are
after the relevant dates, but in order to be probative they must
be shown to be free from extraneous factors, and it must be shown
that values have not changed materially during the period between
the valuation date and the dates of the later sales: see
Roberts and Bagwell v. The Queen.[10] Not only did the
Respondent fail to establish that the market was stable from 1994
(or even 1996) until the times of these later sales, but Mr.
Shannon accepted that it was a period of rising prices, both in
his earlier report, and on cross-examination. I can find no
assistance in the evidence of Mr. Shannon.
[28] My conclusion as to value, then, is
this. Each of the 187 units had a fair market value on the date
of its first occupation of $75.00 per square foot of floor space.
For the reasons that I have given at paragraphs 12 to 14 above, I
make no adjustment to this amount for the hypothetical glut on
the market that caused Mr. Lansink to discount this amount
by about 45%. If I were persuaded that this is a proper factor to
take into account, then the allowance that I would make would be
to discount the values of those 96 apartments first occupied in
the months of July, August, September and October, 1995 by 10%. I
consider this to be sufficient to ensure that all the apartments
would be absorbed by the market as they became available. The
effect of such a discount would be to decrease the aggregate
value of the units by $895,860, divided more or less evenly
between the two Appellants.[11]
[29] While the evidence is not entirely
clear, it appears from the agreed facts and from admissions read
in from the examination for discovery of an officer of the Crown
that Richmill reported and paid GST on the basis of the draft
appraisal report that it had obtained from Mr. Best in June
1995.[12] That
report concluded at pages 51-52 that the aggregate value of
the 187 suites making up the residential portion of 675 Richmond
Street, was $12,000,000, as at August 1, 1995. Two things should
be noted. First, the later appraisal furnished by Mr. Lansink in
September 1998 was not completed until some time after the last
reporting period that relates to these apartments. Second, Mr.
Best's appraisal of the apartments was $2,150,000 higher than
Mr. Lansink's 1998 appraisal. The Respondent has conceded
that no penalties should have been assessed against the
Appellants for the period prior to January 31, 1997,[13] except to the extent
that the GST was underreported by Richmill. The remaining issue
with respect to the penalty under section 280 of the Act,
then, is whether the requirement of due diligence has been met,
even though Richmill (until January 31, 1997), and the Appellants
(from February to September 30, 1997) reported and paid GST on
the basis of estimates of value that were less than the
Minister's estimate, and less than the values that I have
arrived at.
[30] Did Richmill (in the period prior to
January 31, 1996) and the two Appellants (in the period after
January 31, 1996) exercise due diligence in reporting and
remitting GST in respect of the deemed self-supply of these
187 apartment units? In my view they did. They were
confronted with a difficult valuation issue and, quite properly,
they turned to an accredited member of the Appraisal Institute of
Canada for an opinion as to the values of the various units.
Having obtained that opinion, they based their reporting and
remittances of tax on it. Counsel for the Respondent suggested in
argument that they deliberately minimized the values of the units
and thereby under-reported and under-remitted tax in order to
alleviate the corporate group's cash flow problems. This
theory was not put to Mr. Krauel on cross-examination, nor is
there anything in the evidence that gives any support to it. I
have no reason to believe that Mr. Best gave, or was asked to
give, anything other than his honest opinion as to value.
Richmill and the Appellants did all that is reasonable to require
of them to establish the fair market values.
[31] As the aggregate fair market value of
the units owned by each of the Appellants, as determined at
paragraph 27 above, is greater than the aggregates determined by
the Minister for the purpose of the reassessments, the appeal
must fail insofar as the tax exigible is concerned. However, the
appeals will be allowed and the reassessments referred back to
the Minister for reconsideration and reassessment on the basis
that the Appellants are not liable to a penalty under section 280
of the Act. The Respondent has been successful on the
issue which occupied almost all the time at trial, and is
therefore entitled to one set of costs.
Signed at Ottawa, Canada, this 18th day of June, 2004.
Bowie J.
Email this Content