delivered the opinion of the Court.
This appeal marks our second review of the 1978 tax assessment on certain Fairfax County property owned by David E. Nassif, doing business as Suffolk Properties. In Fairfax County v. Nassif 223 Va. 400, 290 S.E.2d 822 (1982) (Nassif I), this Court held, among other things, that the trial court erred in ruling as a matter of law that actual contract income must be used by the County in appraising property by the capitalization of income method. Consequently, we ruled that the trial court erroneously reassessed the property in an amount less than the County’s assessment. In Nassif I, we said further that “as a general rule, economic rent is the measure to be used in capitalizing income for fair-market-value determination; however, contract rent is relevant evidence of economic rent.” 223 Va. at 405, 290 S.E.2d at 825. In Nassif I, we also stated that “the assessor erred in failing to consider contract rent in making the assessment.” Id., 290 S.E.2d at 825. We remanded the case for further proceedings consistent with the views expressed in Nassif I, directing that on remand the County’s assessment be accorded the presumption of validity.
On remand, the taxpayer’s expert testified that he made new calculations, using economic rent instead of contract rent but that *474in accordance with the mandate of Nassif I he considered the contract rent as relevant evidence of economic rent. On remand, Nassifs expert valued the property at approximately $5,500,000.
By contrast, the County, in effect, resubmitted its original 1978 appraisal of $7,113,000 and said it was “going with” that appraisal. With regard to the impact of contract rent on the determination of economic rent, the County’s appraiser said the contract rent was too low to be considered.
Further, on remand the trial court refused to permit Nassif to challenge the County’s appraisal on any ground other than that of the impact of contract rent upon the original economic rent valuation. The trial court held that the opinion in Nassif I forbade consideration of any other issues.
Ultimately, the trial court concluded that because of the presumption of validity that attached to the original assessment, that assessment had to be upheld. The trial court said, in effect, that Nassif failed to meet the burden imposed upon him to overcome the presumption. Though the trial court questioned how the County’s assessment could remain the same if contract rent was considered as directed in Nassif I, the court nevertheless accepted the County’s valuation.
On appeal, Nassif makes three assignments of error. First, he contends the trial court improperly limited the scope of the proceedings on remand. Second, he contends the trial court erred in holding that he failed to overcome the presumption of correctness of the County’s assessment. Third, he contends the trial court erred by not properly applying the requirement of Nassif I that contract rent was to be considered as relevant evidence of economic rent in the determination of fair market value by the capitalization of income method. We find merit in all of Nassif's assignments of error; therefore, we will reverse the judgment of the trial court.
Facts
Nassifs property consists of three adjacent parcels of land improved by a multi-story office building that contains 229,549 square feet of leasable space. In 1964, 94% of the space was leased to the federal government for an original term of ten years with two renewal options of five years each. The total lease term was twenty years. However, the lease contained no rent-escalation clause or expense-pass-through provision.
*475In 1974, at the time of the first renewal option, the lease was renegotiated with the lessee taking over responsibility for utilities and char services. The rental rate was adjusted to reflect those and other changes.
Seventy-five parking spaces were included in the lease to the federal government. However, 558 additional spaces were rented on a monthly basis to individuals. In 1976, the rental from the individual spaces yielded $30,000.
The property was assessed in 1977 using the capitalization of economic income method. That year, the assessed value of the property was $5,336,245.
In 1978, the property was reassessed using the same appraisal method. The assessor, Frederick C. Wyse, admitted that, in making the 1978 appraisal, he ignored the actual contract rent. The 1978 value was determined to be $7,113,842.
On July 13, 1978, Nassif filed an application under Code §§ 58-1145 and 58-1148 alleging that the 1978 assessment was erroneous. The dispute was heard by the trial court which concluded that the County erred in failing to use actual income in valuing the property. The County appealed. We reversed and remanded.
In Nassif I, we said the following about the further disposition of the case:
Accordingly, it is necessary to remand the case so that further proceedings may be conducted in light of the views expressed in this opinion. In the further proceedings, the trial court shall accord the County’s 1978 assessment the presumption of correctness to which it is entitled and then determine from the evidence whether Nassif has borne his burden of overcoming the presumption.
223 Va. at 405, 290 S.E.2d at 825 (citations omitted). On remand, the parties differed as to the issues that were to be considered.
After a hearing and argument by counsel regarding the scope of the new trial, the trial court ruled that the issues on remand were limited by what we said in Nassif I. In a letter opinion, the trial court wrote as follows on the subject: “The issue shall be the economic rent and the consideration given to the contract when economic rent is determined.”
*476The case came on for trial on remand on November 22, 1982. The parties agreed that the record of the first trial should be admitted as evidence in the trial on remand.
Nassifs sole witness on remand was McKenzie Downs. In the first trial, Downs had valued the property based on actual rent. As noted earlier, we reversed the judgment based on actual rent because the trial court had required, as a matter of law, that actual rent be used in appraising the property based on the capitalization of income approach. In the first trial, Downs valued the property at $5,250,000.
On remand, Downs testified that he recalculated his appraisal according to the dictates of Nassif I. He said he went into the market and obtained economic rent data. He then made adjustments to the economic rent based on the level of services provided and such things as the age, condition, location, and parking situation at the comparable buildings. He took into consideration the actual rent at the Nassif building itself and concluded that though the actual rent was $3.58 per square foot the economic rent was $4.30 per square foot. Using that figure and attributing value to the 558 individually leased parking spaces, he arrived at an estimated total gross income of $1,098,333. He next applied a vacancy rate and rent loss figure and arrived at an adjusted gross income of $1,085,810. Next, he subtracted estimated expenses and arrived at a net income prior to taxation of $603,757. This figure was then capitalized at the rate used by the County, 12.57%, giving a value for the property of $4,803,158. Downs testified further that had he been allowed to use his own capitalization rate, he would have valued the property at $5,189,411. Downs, in a report that was made an exhibit, estimated the value of the property at $5,500,000.
The County’s case on remand was based on Wyse’s testimony from the first trial, along with the live testimony of Harry Middleton, who had replaced Wyse. Wyse was the County official who actually made the 1978 appraisal. Wyse testified that as of January 1, 1977, the County was assessing at 100% of value. Wyse defined fair market value as the highest price in terms of money the property will bring in a competitive and open market under all conditions requisite to a fair sale, the buyer and seller each acting prudently and knowledgeably and assuming price was not affected by undue stimuli. Wyse said this was the definition he used in appraising property. However, he went on to say that fair market *477value for tax purposes may be different from fair market value for sale purposes where a leasehold interest is involved.
Wyse also defined economic rent. He said it is competitive rent which the property would bring in a competitive market. He admitted, however, that because of the long-term lease with the federal government, the Nassif property was not in a competitive market.
Wyse said he valued the land based on comparable sales of raw land. He determined from this market data that the land was worth $5.00 per square foot.
Wyse’s economic rent figures were based on a survey of properties in the area. According to Wyse, his survey showed that the appropriate competitive rent was not less than $7.50 per square foot. Wyse admitted that he did not adjust his comparables to make them conform to the level of services provided by Nassif to his lessees. He admitted further that his comparables included services not provided by Nassif. Wyse also was aware that the actual rent at the Nassif building was $3.58 per square foot.
With regard to the role played by the long-term lease in his calculation of economic rent, Wyse said he was not concerned with the fact that the lease was entered into in 1963 nor that it had considerable years to run. He admitted, however, that a buyer in the market would be concerned with the lease.
Wyse assigned a net value of $6.00 per square foot to the Nassif building. Wyse testified that if the property were leased in 1978, it would generate gross revenues of $1,339,986. From this figure, Wyse subtracted expenses and arrived at a figure of $801,982 net economic income before taxes. He then capitalized this figure and, using a capitalization rate of 12.57%, arrived at a value of $7,113,000.
Wyse admitted that he had not made a fee appraisal of the subject property. He admitted too that a prudent purchaser would not look at average values but at the precise property itself. He also said that he does not use actual rent unless it is economic. According to Wyse, contract rent has little or no role to play in determining economic rent.
Middleton, in essence, adopted Wyse’s $7,113,000 valuation of the subject property. According to Middleton, he reviewed Wyse’s assessment. He said, “I looked at this rent and I confirmed these figures.” Middleton testified further that he reviewed the actual contracts on the Nassif building. However, when asked what ef*478feet the contract rent had on the determination of economic rent he said, “I feel that the contract rent in this case here is low and does not reflect the current 1978 market.” Middleton was asked again about the impact of the contract rent on the County’s valuation of the property, he merely said that after studying the market he “concurred that the economic rent that was used on the assessment or was used with the appraisal, which set the assessment for 1978, was correct.”
On cross-examination, Middleton defined economic rent as “that rent which a market is currently paying for space.” He said he was aware of the contract rent, but since it was not comparable to the 1978 market, it did not change his estimate of value.
Middleton admitted further that he made no adjustments to his comparables to take into account differences in the various properties. He admitted too that where a comparable provides for parking in the lease there should be an adjustment when making a comparison to the Nassif building which provided only limited parking.
Middleton made the further admission that he made no determination of the economic rent for the Nassif building itself. He said, “[n]o, not that particular building for that because we are looking at a uniform and equitable assessment of all buildings and I’m looking at economic rent in the area in the general proximity, general office space, of what it would rent for.” He was then asked whether his interpretation of “uniform” meant the determination of an average rental. He replied, “[w]ell, it isn’t an average rental; but, a rental that suits the — that the market presents, the rental market presents.”
At the conclusion of the evidence, the trial court made clear that it had great trouble with the County’s case, yet, upheld the assessment on the basis of the presumption of correctness. The trial court said it was deciding the case on the basis of Nassif I “and the narrow issue that is before the Court concerning the economic and contract rent.” The trial court commented on the well-documented nature of Nassif's case and the logical nature of the adjustments made to comparables. The trial court then noted that it was strange that, in light of Nassif I, the County could advance the identical assessment. The trial court made the following remarks on that point:
*479It is somewhat strange when the Supreme Court rules that the record shows conclusively that the assessor did not give consideration to the contract when he arrived at the amount of assessment in this case and then within the last year, a few days, the County appraiser and the expert on behalf of the County looks at the same information and looks at the contract and says that does not affect the assessment in his opinion and the use of comparables without any adjustment.
The trial court was of the view that it was “under strong admonishment” from this Court to afford the 1978 assessment the presumption of correctness. It is apparent that only because of this, the trial court upheld the assessment.
General Principles Concerning Review of Assessments
The principles that guide disposition of cases of this kind are well settled and have been often recited. Recently, we restated them in Bd. of Sup. v. Donatelli & Klein, 228 Va. 620, 627, 325 S.E.2d 342, 345 (1985). There, we noted that the burden is on the taxpayer to show that the property is assessed at more than fair market value or that the assessment is not uniform in its application; that a clear presumption favors the validity of the assessment; and that this presumption can be rebutted only upon a showing of manifest error or a total disregard of controlling evidence. Further, in Donatelli, we reiterated the traditional definition of fair market value, which is, “the price property will bring when offered for sale by a seller who desires but is not obliged to sell and bought by a buyer under no necessity of purchasing.” Id. at 628, 325 S.E.2d at 345 (citations omitted).
In Nassif I, we defined economic rent. There, relying on Shaia v. City of Richmond, 207 Va. 885, 895, 153 S.E.2d 257, 265 (1967), we said that economic rent is the amount a typical lessee should be willing to pay for the right to use and occupy the premises for a stated period. Nassif I, 223 Va. at 402 n.1, 290 S.E.2d at 823 n.1.
*480Application of these and other principles to the present facts makes clear that the trial court erred in upholding the County’s 1978 assessment of the Nassif property.
Scope of Proceedings on Remand
The County contends, for two reasons, that on remand the case was limited to a consideration of the narrow issue of the effect of contract rent on the determination of economic rent. First, the County contends that in Nassif I, we so limited the case. Second, the County contends that Nassif's failure to assign cross-error to the trial court’s failure to rule in his favor on grounds, in addition to the one which was held dispositive, forever forecloses consideration of the full case. We will consider the second issue first.
At the beginning of this odyssey, Nassif challenged the 1978 assessment on multiple grounds. He contended, for example, that a 33!/3% increase in valuation from 1977 to 1978 where there were no changes either in the property or in market conditions was unwarranted; that there was a lack of uniformity of assessments because different assessors were using different approaches to valuation in the same year; that in making the 1978 assessment the assessor improperly doubled the value attributable to the land as compared to the 1977 assessment; that in calculating economic rent for the Nassif building, the assessor improperly included the actual income from the 558 parking spaces even though the economic rent was based on figures from buildings where parking was included in the rental and thus contained a built-in factor representing parking; and, that the assessor ignored the actual rent generated by the property. The trial court disposed of the case on the last issue alone. Nothing was said about the other issues. The County now contends that when it appealed, Nassif should have cross-appealed and assigned cross-error to the trial court’s failure to rule favorably on the other asserted bases for challenging the assessment.
The County’s argument is not persuasive. When this Court rules that the judgment of a trial court is erroneous it does not matter whether that judgment is erroneous for one reason or ten; it is no longer viable. Unless we say otherwise, the slate is wiped clean, with the result that on remand the parties begin anew. It would serve no useful purpose, we think, to require a prevailing party to assign error to his failure to win on all points in order to *481protect his right to a full and complete trial should his apparent victory be reversed and the case remanded.
Next, Nassif I does not contain a limitation or restriction of the kind suggested by the County. When we limit issues on remand we do so with words of limitation or restriction. For example, we sometimes are heard to say “the case is remanded for a new trial limited to the issue of damages.” No such language can be found in Nassif I. Indeed, the language of Nassif I is quite to the contrary. We remanded for further proceedings in light of the views expressed in that opinion. Then we wrote that “the trial court shall accord the County’s 1978 assessment the presumption of correctness to which it is entitled and then determine from the evidence whether Nassif has borne his burden of overcoming the presumption.” Nassif I, 223 Va. at 405, 290 S.E.2d at 825 (emphasis added). The emphasized phrase shows that there was no limitation on the evidence. The word “evidence” is not given a restrictive modifier.
All the evidence should have been considered on remand. We hold that the trial court erred in restricting Nassifs efforts to challenge the validity of the 1978 assessment.
Sufficiency of The Evidence To Overcome The Presumption In Favor of the County’s Assessment
Nassif contends the presumption of correctness of the 1978 assessment was overcome for at least three reasons: First, because the County simply confirmed an assessment that this Court had already declared infirm; second, because Middleton, in confirming the original assessment admitted that he used a definition of economic rent at odds with the one stated by this Court in Shaia and reiterated in Nassif I; and third, because the assessor, in fact, did not consider contract rent as relevant evidence of economic rent. Nassif's position is well taken.
In Nassif I, though we reversed the trial court’s judgment assessing the property at a value lower than the County’s $7,113,840 assessment, we refused to reinstate the County’s assessment. We explained our treatment of the matter in the following terms:
The County concedes that contract rent should be considered in this type of case. Yet, the record shows conclusively that *482the assessor ignored contract rent in appraising the property for the 1978 assessment. Thus, while the trial court erred in refusing to consider economic rent in testing the correctness of the County’s assessment, the assessor erred in failing to consider contract rent in making the assessment.
Nassif I, 223 Va. at 404-405, 290 S.E.2d at 824-25 (emphasis added).
On remand, two things occurred. The $7,113,840 assessment was, at the direction of Nassif I, accorded the presumption of correctness and the record from the first trial was incorporated by reference.
Nassif submits that although, at the beginning of the new trial, the 1978 assessment was presumed correct, the moment the record from the first trial was incorporated into the evidence of the second trial the 1978 assessment lost its presumption of correctness because this Court, in Nassif I, had already ruled that Wyse had erroneously ignored contract rent in calculating economic rent. Nassif explains that if the record in Nassif I had not contained sufficient evidence to overcome the presumption of correctness of the 1978 assessment, this Court would have simply reinstated the County’s assessment and entered final judgment for the County. Nassif goes on to argue that by remanding the case for a new trial this Court effectively ruled that neither Nassifs appraisal nor the County’s appraisal was correct. Consequently, Nassif argues, when the evidence from the initial trial was incorporated into the retrial, that evidence necessarily overcame the presumption in favor of the 1978 assessment.
The County does not respond directly to Nassif s explanation of how the presumption of correctness of the 1978 assessment was overcome on remand. The County contends, instead, that the presumption of correctness was obviously intended by this Court to attach to the 1978 assessment “as reconsidered.” The County’s argument suggests that the presumption of correctness was not present at the beginning of the new trial, but somehow sprang into existence once the County’s witness testified that he had considered contract rent as relevant evidence in determining economic rent. According to the County, “[o]nce the assessor considered contract rent as relevant evidence in the remand, the presumption thereupon attached to that assessment. . . .” (Emphasis in the original.)
*483The County misunderstands Nassif I. The effect of that opinion was to restore the parties to their original positions at the time the challenge to the assessment was first filed and to give guidance to the trial court on remand. We clearly intended that the presumption apply at the beginning of the trial, as in the usual case, and that the taxpayer have the burden to overcome that presumption. Nassif I did not say that the existence of the presumption was tied to testimony from the County that it had considered contract rent in calculating economic rent.
Nassif s argument on overcoming the presumption is the correct approach. In any proceeding where a taxpayer challenges an assessment, the case begins with the assessment entitled to the presumption of validity, and the taxpayer bears the burden of overcoming the presumption. When we wrote Nassif I, we did not know how proof would proceed in the new trial. There was no way for this Court to know that the record of the first proceeding would be incorporated into the evidence of the new trial. However, as soon as that was agreed to by the parties and permitted by the trial court, the presumption of validity was necessarily overcome because, at that point, it was clear that the assessment was based on a methodology which we had already declared improper.
Nassif submits that even if we were to adopt the County’s position that the presumption arose “once the assessor considered contract rent as relevant,” the County would still not prevail because, as a matter of fact, the County never properly considered contract rent in calculating economic rent. We agree with Nassif. In our opinion, when the case was tried on remand, the County disregarded what we said in Nassif I. Our review of the evidence makes plain that instead of considering contract rent as relevant evidence of economic rent, the County simply did what it had done in the first case and resubmitted an assessment which had given no effect to contract rent. Nassif I would be an exercise in futility if all it did was to require an assessor to “think” about contract rent, then reject it as meaningless in arriving at economic rent. Nassif I means more than that.
Since we handed down Nassif I, we have had occasion to discuss it in two assessment cases. In Donatelli, we commented that in Nassif I, “we held that contract rent is relevant evidence of economic rent and cannot be disregarded by the appraisers.” 228 Va. at 630, 325 S.E.2d at 347 (emphasis added). In Arlington County Board v. Ginsberg, 228 Va. 633, 640, 325 S.E.2d 348, 352 *484(1985), we made reference to Nassif I and said: “As a general rule, determination of fair market value by capitalization of economic rents is the preferred method, but consideration of contract rents is required in ascertaining economic rents." (Emphasis added). Thus, in determining economic rent, contract rent must be factored into the formula; it cannot be disregarded.
Here, the first assessor, Wyse, completely ignored contract rent. We held that to be error. Then, the second assessor testified that he did not use contract rent because it was too low and that he confirmed the first assessment. It follows, in light of Nassif I, that what happened on remand must also be error.
Nassif also submits that the 1978 assessment is in error because it was based on an improper conception of economic rent. We defined economic rent in Nassif I as the amount that a typical lessee should be willing to pay for the right to use and occupy the premises for a stated period. Nassif I, 223 Va. at 402 n.1, 290 S.E.2d at 823 n.1. That definition focuses upon the property that is being valued and plainly indicates that the determination of economic rent must be specific to the property under review as opposed to some abstract or theoretical property.
At trial, Middleton gave the following definition of economic rent, “Economic rent is that rent which a market is currently paying for space.” He explained further that by “market” he meant “the general market of office space throughout the County.” He also admitted that he did not determine the economic rent for the Nassif property itself. During trial, he was asked whether the approach he used resulted in an average rental for the County. He responded as follows: “Well, it isn’t an average rental; but, a rental that suits the — that the market presents.”
Whether the County’s approach is a true averaging or whether it results in theoretical, market-wide values it is at odds with the definition of economic rent that we have laid down because it leads to values unrelated to the specific property being appraised. In Donatelli, another assessment case from Fairfax County, we criticized the County’s economic rent model in the following terms:
The economic model used by the County is based on income and expenses of commercial properties throughout the entire county area. The model does not allow for adjustment in values where the subject property is situated in an area in *485which it cannot generate the level of income produced by identical property in a more commercially suitable location. Thus, in the name of uniformity, the County’s model can produce assessment values that are greater than actual fair market values because the properties’ geographical locations and other individual characteristics are not considered.
228 Va. at 630, 325 S.E.2d at 347 (emphasis added). In Donatelli, we said further that the taxpayer’s appraiser, McKenzie Downs — the same appraiser who testified here on behalf of Nassif — had complied with the Nassif I mandate but that the County’s economic rent approach violated Nassif I. More specifically we wrote, “the economic rent approach employed by the County, disregarding actual leases, was based on a theoretical return obtainable under optimum conditions which may never exist.” Donatelli, 228 Va. at 632, 325 S.E.2d at 348 (emphasis added).
Here, it is apparent to the Court that the evidence before the trial court on remand was sufficient to overcome the presumption of validity of the County’s 1978 assessment. That assessment was erroneous. Indeed, it is clear that the only evidence offered on remand that met the requirements of Nassif I and our other opinions was Nassif's evidence. Consequently, we will reverse the judgment of the trial court which upheld the 1978 assessment of Nassifs property at $7,113,840; assess the subject property at $5,500,000 in accord with the evidence; enter final judgment on that assessment, and remand the case for further proceedings consistent with this opinion. See Smith v. City of Covington, 205 Va. 104, 135 S.E.2d 220 (1964); Skyline Swannanoa v. Nelson County, 186 Va. 878, 44 S.E.2d 437 (1947).
Reversed and remanded.