Burroughs Corp. v. State Board of Equalization

153 Cal. App. 3d 1152, 200 Cal. Rptr. 816, 1984 Cal. App. LEXIS 1854

• Court: California Court of Appeal
• Decided: March 30, 1984
• Docket: Civ. 22579
• Status: Published

Opinion

Opinion

SPARKS, J.

In this case we review a challenged application of the “Sales and Use Tax Law” (Rev. & Tax. Code, § 6001 et seq.) to the computer industry. Plaintiff Burroughs Corporation appeals from a judgment of the Superior Court of Sacramento County which rejected its claim for a refund of nearly $800,000 for use taxes paid in the period between July 1, 1972, and December 31, 1975. This dispute arises from Burroughs’ use of computer components, manufactured by itself, for purposes of testing other computer components which it also manufactured. Burroughs contends that since the components used for testing (referred to as captive components or captives) were ultimately refurbished and sold or leased at full price, it should not be liable for the use tax levied on its use of the components for testing. The State Board of Equalization (Board) contends that the first use of captives to test and evaluate other components coming off the assembly line constitutes a taxable intervening use of the captives. The Board alternatively contends that the captives are taxable under the primary purpose test. We hold that the primary purpose test, defined by case law and promulgated by regulation, applies here. Judged by that test, Burroughs was liable for use taxes levied on the materials used to manufacture the captive components. Accordingly, we shall affirm the judgment.

Facts

This dispute arose from Burroughs’ manufacture of computer systems during the period in question. Burroughs operated four manufacturing plants in California. These plants are referred to as the Santa Barbara plant, the Westlake plant, the Pasadena plant, and the Mission Viejo plant. Each of the plants was responsible for a different aspect of Burroughs’ manufacturing activity. The components from the different plants would be brought together to form a complete computer system.

A computer system consists of a family of components which, when brought together into a single entity, will perform the data processing functions the system was intended to accomplish. Within the different components of a system there are numerous options which may be selected by a customer. When a customer selects a computer system, the various manufacturing plants in turn receive an order for the component with the options necessary to complete the system. As each component is completed it must be tested. Due to the nature of the industry, it is not sufficient that a component be tested by itself. A computer component which itself operates properly may not be compatible with other components of the system it is intended to be operated with. In other words, it may not be able to communicate or “interface” with the other components. One of the main reasons for this is the constant change in the design of the computer components. The designs utilized are in a constant state of flux. As new designs are incorporated into various components, a problem may develop in the way other components interface with them. For this reason each component must be subjected to “interactive” testing, that is, as each component is completed it must be connected to other components of a complete system to ensure that it will be compatible with the other components with which it must operate.

Burroughs utilized what it refers to as “captive components” or simply “captives” for interactive testing. There were several reasons for using captives for testing: First, captives served the special function, not served by other testing equipment, of testing an entire system for compatibility. Second, because Burroughs engineering group was located in its manufacturing plants, and field engineers who installed the systems were not necessarily qualified to isolate and resolve the engineering or design problems, testing was best done at the plant. Third, there was the risk that in attempting to resolve a problem a field engineer might create unique and incompatible machines in customers’ offices. Fourth, the cost of shipping the various components to the same place for interactive testing was prohibitive. For these reasons captives were employed and components of the system tested at the plant before they came together for the first time in the field for installation.

Captive and customer orders were handled identically within Burroughs. The manufacturing branch would get an order from the marketing depart ment requesting a specific captive configuration. Some captives were manufactured in the same plant requesting the captive. Others were manufactured at one of Burroughs’ three other plants. A captive would be a complete unit in that it could have been shipped to a regular customer except that it was generally ordered without power cables because Burroughs had permanent cables in their plants. The captives would have “skins” (covers) on them but these were usually removed once they were installed for testing in order to facilitate testing and adjustment.

Once the captive itself was tested for compatibility with other Burroughs’ components, it became a known entity which could be used to test unknown entities, the newly manufactured components. Captives were used over and over again to test newly manufactured equipment. Burroughs’ policy was that captives should be rotated once every 12 months. In practice, the majority of captives rotated between one and two years, but some were held for considerably longer periods of time, in one instance for eight years. The 12-month captivity period was believed to be the optimum balance between costs of shipping and installing captives, on the one hand, and the risk of obsolescence, on the other. During the period of captivity engineering changes were made in the captives so that they would be up to date.

As a percentage of total production, the number of captives in use at the four manufacturing plants for the years 1972 through 1975 varied from year to year and plant to plant: Pasadena-14.1 to 7.3 percent; Mission Viejo26.4 to 20.4 percent; Santa Barbara-15.5 to 8.5 percent; and Westlake-6.6 to 4.3 percent.

When a captive was rotated it was sent back to its plant of manufacture and a new captive was installed. The old captive would be refurbished, outstanding engineering changes would be incorporated into it, and ultimately it would be sold or leased at full value. The costs of refurbishment and modification averaged 30 percent of the standard cost of the unit at all but the Westlake plant, where it averaged 32 percent.

For accounting purposes Burroughs kept a special account for captives. Although an investment credit was claimed on the captive account, the captives were not depreciated. Under Burroughs’ accounting policy, a captive not rotated within 12 months was transferred to another account and capitalized and depreciated. At trial it appeared that 85.3 percent of the units used as captives were rotated, refurbished, and sold or leased. 1

When Burroughs purchased the materials used to construct the components it gave the vendors resale certificates and no sales tax was paid at that time. 2 When a component was designated as a captive, Burroughs, under protest, paid a self-assessed use tax based upon the value of the materials used to construct the captive. Burroughs then brought a timely action for a refund of the amounts paid as use taxes for the captives which were ultimately rotated and returned to marketing for sale or lease. (See Rev. & Tax. Code, § 6933.) In the refund action, Burroughs’ principal contention was that because it had to pay a sales tax when the captives were finally sold or leased to consumers, it should not also have been required to pay a use tax for the material incorporated into the testing captives. Burroughs argued that captives were manufactured primarily for sale or lease, and not for testing.

The trial court applied the primary purpose test, most recently reaffirmed in Kaiser Steel Corp. v. State Board of Equalization (1979) 24 Cal.3d 188, 198 [154 Cal.Rptr. 919, 593 P.2d 864], and concluded that there were two substantial purposes for purchasing the materials which ended up in captives, “neither of which categorically and objectively can be designated as ‘primary.’” The court pointed out that the burden of proving one primary purpose was on Burroughs, and that it had failed to carry that burden. 3 Accordingly, the court denied the claim for a refund of use taxes. This appeal followed.

Discussion

Revenue and Taxation Code section 6051 imposes a sales tax on all retail sales. 4 The tax, computed as a percentage of gross receipts, is imposed upon all retailers for the privilege of selling tangible personal property at retail. (Lockheed Aircraft Corp. v. State Bd. of Equalization (1978) 81 Cal.App.3d 257, 265 [146 Cal.Rptr. 283].) Section 6007 defines a “retail sale” or “sale at retail” as “a sale for any purpose other than resale in the regular course of business in the form of tangible personal property.”

Section 6201 alternatively imposes a use tax upon storage, use or consumption in this state of tangible personal property purchased from any retailer for the purpose of storage, use or consumption in this state. 5 The use tax, like the sales tax, is not imposed upon tangible personal property held for resale in the regular course of business. (§§ 6008, 6009.)

The sales tax and the use tax are designed to be complementary. 6 Consequently, transactions subject to sales tax are exempt from the use tax. (§ 6401.) “As property covered by the sales tax is exempt under the use tax, all tangible personalty sold or utilized in California is taxed once for the support of state government. ” (Southern Pac. Co. v. Gallagher (1938) 306 U.S. 167, 171 [83 L.Ed. 586, 589-590, 59 S.Ct. 389].) The purpose of the use tax is to place local retailers upon an equal footing with their out of state competitors, since these competitors are not liable for the California sales tax. (Chicago Bridge & Iron Co. v. Johnson (1941) 19 Cal.2d 162, 165 [119 P.2d 945].) It has been said that all property not actually covered by the sales tax is subject to the use tax. (Bank of America v. State Bd. of Equal. (1962) 209 Cal.App.2d 780, 792 [26 Cal.Rptr. 348].)

The Revenue and Taxation Code contains provisions to insure that the sales tax and the use tax will in fact complement each other. Thus, section 6094, subdivision (a), provides: “If a purchaser who gives a resale certificate makes any use of the property other than retention, demonstration, or display while holding it for sale in the regular course of business, the use shall be taxable to the purchaser ... as of the time the property is first used by him . . . .” Likewise, section 6244, subdivision (a), provides: “If a purchaser who gives a resale certificate or purchases property for the purpose of reselling it makes storage or use of the property other than retention, demonstration, or display while holding it for sale in the regular course of business, the storage or use is taxable as of the time the property is first so stored or used.” 7

As we have noted, a retail sale does not include a sale for the purposes of resale in the normal course of business. Burroughs claims it manufactures the captives for resale and hence is not liable for any use tax. Burroughs consequently frames the issue this way: “Is a temporary use by a computer systems manufacturer of sample components for interactive testing, in which the component is both testing and being tested, subject to use tax when those components are subsequently incorporated into complete computer systems and sold or leased as new equipment, thereby subjecting them to a second tax at their highest value?” The Board reframed the question in this manner: “The issue, stated in terms of the statute, is whether the use of captives by Burroughs is a ‘use . . . other than retention, demonstration or display while holding it for sale in the regular course of business,’ within the meaning of Revenue and Taxation Code sections 6094, subdivision (a), and 6244, subdivision (a), subjecting Burroughs to the use tax imposed by section 6201 of said code.”

In our view, neither party correctly states the question. As we shall demonstrate, the dispositive question is whether the primary purpose in using the captives was to test other components, a taxable use, or to prepare the captives for resale, an exempt use.

When personal property is both used and resold by a manufacturer a conflict frequently arises over its taxability. The “primary purpose” test has been developed by the courts to resolve this conflict in close cases. 8 In determining whether a transaction is taxable or exempt, “the California courts have consistently looked to the primary intent of the purchaser or the primary purpose of the purchase. [Citations.]” (Kaiser Steel Corp. v. State Board of Equalization, supra, 24 Cal. 3d at pp. 192-193.) A review of those cases will be instructive.

In Kirk v. Johnson (1940) 37 Cal.App.2d 224 [99 P.2d 279], the taxpayers were cattle dealers in the business of selling dairy cows. The original purchase of their cows by dairymen for milk production was taxed in full. The taxpayers claimed that the sales of the cows should have been exempt because the cows were eventually sold by the dairymen for beef. They contended they were making sales for resale because: “When dairy cows become unprofitable because the cost of their feeding and care exceeds the proceeds derived from the sale of their milk, the cows are then usually fattened by feeding concentrates containing a high degree of protein, and then sold to slaughter houses to be butchered and resold as meat and meat products.” The Court of Appeal rejected the contention, holding that the resale of the cows was only secondary and incidental to the chief purpose, which was to derive a profit from the sale of milk. (37 Cal.App.2d at p. 227.)

In People v. Puritan Ice Co. (1944) 24 Cal.2d 645 [151 P.2d 1], ice was sold to vegetable packers and shippers for use in preserving the perishable products during transit. The Supreme Court held that the ice sales constituted taxable retail sales despite the fact that the purchasers separately charged their customers for the ice. “The essence of the matter is that the purchasers of the ice are acquiring it for purposs [sfc] other than resale. They are not engaged in the ice selling business. They are selling vegetables and the use of the ice or purported sale thereof to the purchasers of the vegetables is merely an incident of that activity. It is common knowledge that the dominant purpose for the use of ice in shipping perishable produce is to preserve the produce by means of refrigeration . ...” (24 Cal. 2d at pp. 651-652.)

In Safeway Stores v. State Bd. of Equal. (1957) 148 Cal.App.2d 299 [306 P.2d 597], the plaintiff purchased cardboard containers which were used for package and delivery of canned goods to retail merchants. When the canned goods were unpacked the boxes were used by the retailers for packaging goods purchased by customers. The plaintiff contended the boxes were purchased for resale, but the Court of Appeal disagreed: “It then made, however, a substantial use of the cartons other than for retention, demonstration, or display while holding them for sale in the regular course of business. This use consisted of using the empty cartons as receptacles for packing its manufactured products and transporting them to its retail outlets.” (148 Cal.App.2d at p. 303.) Likewise, in H. J. Heinz Co. v. State Board of Equalization (1962) 209 Cal.App.2d 1, at page 6 [25 Cal.Rptr. 685], the plaintiff was held liable for tax for the use of boxes and cans which it used to store and deliver tomato paste. In Luer Pack. Co. v. State Bd. of Equalization (1950) 101 Cal.App.2d 99, at page 104 [224 P.2d 744], a use tax was payable on cellulose sausage casings used to manufacture wieners, but removed prior to sale of the wieners. In Am. Distilling Co. v. St. Bd. of Equalization (1942) 55 Cal.App.2d 799 [131 P.2d 609], the plaintiff was held liable for tax upon chemicals used in the fermentation and distillation of commercial alcohol which did not become a part of the alcohol.

There are also cases in which the primary purpose test has resulted in a finding that no tax was due. In Hawley v. Johnson (1943) 58 Cal.App.2d 232 [136 P.2d 638], the plaintiff was a car dealer who used certain cars as demonstrators in his business. The Court of Appeal agreed that the use of the cars as demonstrators was not within the contemplation of the use tax: “So long as the depreciation in sales value of the property held for sale is a reasonable incident of the effort to market the goods we are satisfied that the goods so used are not subject to the use tax ... .” (58 Cal.App.2d at p. 239.) More recently, in McConville v. State Bd. of Equalization (1978) 85 Cal.App.3d 156 [149 Cal.Rptr. 194], this court held that breeding mare horses to get them in foal while being held for sale was not a taxable use. In that case the evidence established that the value of a breeding mare that has not foaled or is not in foal is substantially reduced, making it difficult to sell. Breeding the mares was, therefore, a reasonable incident of the effort to sell the mares. (85 Cal.App.3d at p. 161.) We nevertheless upheld the imposition of the use tax because the plaintiff had treated the mares as capital assets and had taken depreciation for purposes of federal and state income taxation. (Id., at p. 162.) We held that the taxpayer’s treatment of the mares for income tax purposes manifested an intent not to resell them.

In this case, the trial court relied upon the primary purpose test as set forth in Kaiser Steel Corp. v. State Board of Equalization, supra, 24 Cal. 3d 188. Kaiser Steel Corporation (Kaiser) was engaged in the manufacture and production for sale of steel, pig iron, and other allied products. It purchased certain materials to remove impurities from the molten metal. The removal of impurities was accomplished by combining these materials with molten metal to form slag. Fifty-two percent to 97 percent of the materials became components of the slag. The rest of the materials were either incorporated into the steel or dissipated in the manufacturing process. Kaiser sold the slag to a company that removed it from the premises for a nominal amount. The main benefit to Kaiser of this sales arrangement was in getting rid of the slag, (Id., at pp. 190-191.)

The case dealt only with the taxation of that portion of the materials which ended up as slag. (The materials which were incorporated into the final steel products were never subjected to a use tax.) (Id., at pp. 191, 197, fn. 5.) Kaiser sought a refund of use taxes paid on the materials incorporated into slag claiming that it had purchased those materials for the purpose of reselling them in the form of slag. The Supreme Court rejected that claim and denied the requested refund. (24 Cal.3d at p. 198.)

Acknowledging that none of the earlier primary purpose cases dealt with the purchase of materials for manufacturing, the Kaiser court nevertheless concluded that the primary purpose test applied to manufacturing: “Thus, if property is purchased as an aid in the manufacturing process, it is taxable despite the fact that some portion remains in the finished product or that an incidental waste or by-product results. Conversely, if the property is purchased for incorporation as a component of the finished product, it is not taxable despite the fact that some portion may be lost or otherwise dissipated in the manufacturing process.” (Id., at p. 193.)

The court ultimately decided that the Board’s conclusion, that Kaiser’s purchase of materials which eventually ended up as slag was primarily for aid in the manufacture of the steel, was eminently reasonable. (Id., at p. 195.) The court reasoned: “Given the economics of steel versus slag production, the resale of the slag is as collateral to the production of steel as the resale of cows in Kirk was collateral to dairy farming. It is as incidental to Kaiser’s business as the resale of ice in Puritan and the sale of carbon dioxide and fusel oil in American Distilling. The primary-purpose test of Puritan and Kirk is applicable to the manufacturing industries.” (Ibid.)

In the process of reaching its decision, the court addressed Kaiser’s argument that because its use was simultaneous (one use which both aided in manufacturing and also produced slag) rather than successive, the primary-purpose test did not apply. The court rejected the argument noting that the Board made no distinction in its rulings based upon simultaneous as opposed to successive use: “Where there are simultaneous uses but one or primary purpose for the purchase, the entire unit of material is taxed or not taxed, depending on that purpose . ...” (24 Cal.3d at p. 195.)

Finally, the court justified the extension of the primary-purpose test to the manufacturing process: “The primary-purpose test, which we affirm in this decision, finds support in tax policy. The test assures that the entire amount of materials will be taxed at its highest value. Permitting an exemption when materials are incorporated into the finished product is rational, since the materials enhance the value of the product which will eventually be sold and taxed at its highest value. Kaiser’s position would permit use of materials costing millions of dollars in aid of manufacturing a profitable article, while avoiding taxation in the reselling of a relatively worthless waste byproduct. In other words, instead of materials being taxed at their highest value, they will be taxed at a tiny fraction of that value. Such a result does not comport with the legislative scheme as indicated by the court decisions and administrative interpretations.” (24 Cal.3d at p. 198, fn. omitted.)

Burroughs contends that the primary-purpose test is simply an application of the one-tax principle and serves only to prevent tax avoidance. Since Burroughs was not involved in a tax avoidance scheme, it contends that the primary purpose test does not apply. That test, it argues, should therefore not be applied here to support a claimed duplicatory tax. Before reaching that contention, we first address the subsidiary issue, posed by the Board and not discussed in Kaiser, why the express language of section 6094, recited above, does not control this case. That section, as we have noted, imposes a use tax on a purchaser who has given a resale certificate to the seller if the purchaser “makes any use of the property other than retention, demonstration, or display while holding it for sale in the regular course of business, ...” (See also Rev. & Tax. Code, § 6244.) Kaiser did not begin and end its analysis by simply asking and answering the question prompted by section 6094, namely, did the manufacturer do something other than retain, demonstrate, or display the property? Why not?

Burroughs suggests that the answer lies in the “sale exception” of section 6009. We agree. Section 6009, it will be recalled, excludes from the definition of “use” any employment of the property which entails “the sale of that property in the regular course of business.” Without this exclusion, “use” would otherwise include virtually every transitory utilization of property, including the transporting, assembling, and limited individual testing of property. These activities constitute a traditional part of the manufacturing process. If these were “uses” within the meaning of the section 6094, then they would also be taxable because they are not among the exclusions (retentions, demonstrations and displays) listed in that section.

Clearly some uses not excluded by a literal reading of section 6094 are not taxable. Regulation 1525 is clear proof that section 6094 has not been read literally by the Board charged with the duty of administering and enforcing the tax. (See Rev. & Tax. Code, §§ 20, 7051.) Regulation 1525 expressly excludes from tax the use of property for the purpose of incorporating it into a manufactured article. The key is the purpose of the use, not its timing. 9 As the Kaiser court noted, “In 1945, . . . regulation 1525 was substantially modified to clarify and restate in its present form the rule that the purpose of purchase determined the taxability of raw material purchases by manufacturers. The Board has utilized the ‘primary purpose’ test in its tax rulings interpreting regulation 1525.” (24 Cal.3d at p. 194; italics added; fn. omitted.)

The “sale exception” of section 6009 must therefore be read to mean that the use of property for the purpose of fabricating it into a saleable product is not taxable. That exception, however, takes us full circle back to the starting point of the Kaiser court: we must now determine if the primary-purpose test applies here. If so, we must then determine whether Burroughs sustained its burden of proof by establishing that the primary purpose of using the captives was to prepare them for resale.

We conclude that the primary-purpose test properly applies here. The case before us involved the same ultimate question regarding taxation that was involved in Kaiser: when is the use tax properly imposed where property is purchased and used for more than one purpose. Burroughs’ conclusion that the primary-purpose test only applies to tax avoidance schemes constitutes too narrow a reading of Kaiser. As we have recounted, the Supreme Court there explicitly held that the “primary-purpose test . . . is applicable to manufacturing industries.” (24 Cal.3d at p. 195.) The court further recognized that “[i]t is settled law that the eventual resale of personal property by a person who purchases such property for use will not prevent the original sale of such property from being a retail sale subject to tax. (§§ 6009, 6094, subd. (a).)” (Id., at p. 197.) Thus the purpose of the purchase, and not whether the property will eventually be subject to another tax, is the controlling factor. As the Court of Appeal observed in the analogous context of a sales tax in Standard Oil Co. v. State Bd. of Equal. (1965) 232 Cal.App.2d 91 [42 Cal.Rptr. 543], “ ‘As is apparent from this statutory definition [of retail sale], the purpose of the buyer in making the purchase is the determinative factor. If the principal purpose of the buyer is to use the property, rather than resell it, the sale is a retail sale [for purposes of the sales tax]. This is true even though the buyer might intend to resell the same property when he no longer has any use for it. ’ ” (Id., at p. 95, quoting from Sho Sato, The Sales Tax and Capital Transactions, 45 Cal.L.Rev. 450, at p. 453, italics added by the court.)

Having concluded that the primary-purpose test applies here, we are satisfied that the trial court correctly determined that Burroughs did not carry its burden of proving by a preponderance of the evidence that its primary purpose in purchasing the materials used to manufacture captives was to sell or lease them.

As we have seen, the trial court, in its statement of decision, concluded: “The difficulty, however, is that there are two equally substantial purposes, neither of which categorically and objectively can be designated as ‘primary.’” Burroughs argues that the use of the words “categorically” and “objectively” suggests that the trial court applied a burden of proof approximating “beyond a reasonable doubt.” Burroughs contends that application of the proper “preponderance” standard would have yielded a different result, namely, a finding that the primary purpose for captives was for incorporation into the finished product.

There is no merit to this contention. The trial court could not have been more explicit on the matter of standard of proof. In its conclusions of law, it stated: “In view of the foregoing conclusions, the court believes it appropriate to apply section 500 of the Evidence Code requiring a party to carry the burden of proof ‘. . . as to each fact the existence or nonexistence of which is essential to the claim for relief . . . that he is asserting.’ This is the plaintiff’s burden in this case and it has not been fulfilled. . . . [f] . . . For the reasons there stated, there are two equally substantial purposes for the manufacture of captives, neither of which can be designated as the primary purpose. Plaintiff has, accordingly, failed to carry its burden of proof of establishing nontaxability by a preponderance of the evidence.”

Moving to Burroughs’ principal contention, that captives were manufactured primarily for resale, we first outline the standard of our review. It is well established that the question of whether a sale or use falls within the resale exemption is one of fact for the trial court, and the determination of that factual issue by the trial court upon conflicting evidence is conclusive upon an appellate court. (People v. Puritan Ice Co., supra, 24 Cal.2d at p. 651.)

The trial court made these factual findings, some of which we have already summarized: “The captives consisted of central processing units, peripheral controls, and peripherals manufactured by Burroughs which were ordered by one of Burroughs’ manufacturing plants, installed in test stations, and used for varying periods to perform testing and product evaluation functions. The ‘captives’ were used in Burroughs’ California plants (1) in product test centers where they interactively tested every unit coming off the production line, (2) in engineering laboratories where the engineers tested new designs and troubleshot problems encountered on the production line and in product test centers, (3) in software labs where master control programs (software) were developed, and (4) in product evaluation laboratories where Burroughs’ products were tested in operation under stresses which are likely to be encountered by Burroughs’ customers in various situations, i.e., various temperatures, voltages and power surges, and radio wave interference. This testing was ‘interactive,’ the purpose being to determine whether the various configurations of components and master control programs would communicate properly with one another. Both the captives and noncaptives were being tested. The results of a given test would reveal a fault in either the captive, the noncaptive, or both. The interactive testing had to be and was continuous for two reasons. One reason was that the engineering changes were constantly being made in the hardware components of a system and its master control program throughout the life of the system. A second reason for this continuous interactive testing is that there were millions of potential combinations of configurations of components within a given system. ... [f] 4. Several important facts emerge from the extensive testimony of plaintiff’s witnesses: [t] a. Each captive was ordered in a particular configuration by the manufacturing plant that needed it. [f] b. Captives were not complete units; they were ordered without power cords. Captives generally were not the same as customer installed units. Cosmetically, panels had not been included so as to facilitate fast connect/disconnect operations. Also, reliability changes had not normally been incorporated in the units. [1] c. Burroughs’ policy was to rotate captives within one year of being placed in captive status. The rotation policy was not, however, rigidly enforced. Captives generally were used for periods of zero to two years before rotation occurred but have been held as long as eight years, [t] d. Captives were always refurbished before they were sold. The captives had to be brought up to shipping level, requiring both logic changes and reliability changes. Most of the logic changes could be made at the captive plant, but the reliability changes had to be made in the modification center at the plant of manufacture. Captives were, in effect, overhauled or remanufactured to upgrade them to the equivalent level of new equipment before they were sold by Burroughs’ Marketing Division to an outside customer, [f] e. While in captive status, captives generally were a known quantity which had passed the tests, whereas a noncaptive was an unknown quantity which must be tested. Burroughs had a reasonable degree of confidence that a captive at a test station was functioning properly, but there was no guarantee that a defect in a captive could not be discovered at any time, [t] f. There is a minimum standard number of tests that the equipment goes through. Captives have the ability to test a number of options, in one instance 140. [t] g. Captives were used in the Engineering Department to test engineering design changes. [f] 5. The reasons for the policy of one-year rotation periods are twofold. First, frequent engineering changes force the cost of rework to increase at an increasing rate for longer periods of time. Thus, the cost of rework would eventually exceed the standard cost of building a new unit. Second, the cost of installing and setting up the ‘captive’ unit is too costly to do more often than once a year. . . . [t] 9. The cost of refurbishing prior to the sale or lease of a captive is considerable. At Pasadena, Mission Viejo, and Santa Barbara the average cost of refurbishing central processors and related items was 30 percent of a standard manufacturing cost.”

Based upon those findings, the trial court reached these conclusions: . 2. The use of the captives for testing of other work in the process of manufacture, for product evaluation, and for development of master control programs was not an incident of the effort to market the computer systems in question. The uses of captives demonstrated by the evidence was in the manufacturing process, not the marketing process ... [f] 4. Burroughs’ rotation policy and practice did not demonstrate that the use of captives was an incident to the effort to sell computers. The rotation policy and practice was evidence of a manufacturing objective, that is, it was established as a balance between the cost of setting up, tearing down, refurbishing, and recrating the captives for shipment on the one hand, and the need to have current models as captives on the other. In implementing its rotation policy, Burroughs did make efforts to assure that the captives would be marketable when rotated.”

We conclude that the evidence supports the trial court’s factual findings and that these findings further support its conclusion that no one primary purpose was established. We attach particular importance to the following facts. First, the captives were specially ordered to perform particular testing functions; they were not ordered to fill customer orders. Second, the captives were used primarily to test noncaptive components. Finally, rapid technological change required that new captive components be installed for testing. Presumably, captives would have been manufactured and used for testing even if refurbishment for sale was not economical.

In short, the trial court was not convinced that captives were manufactured primarily for resale. The court therefore concluded that Burroughs had failed to carry its burden of proof of establishing the nontaxability of the captives. Since we cannot say that this conclusion is unsupported by the record, we are bound by it.

The judgment is affirmed.

Regan, Acting P. J., and Evans, J., concurred.

1 The complaint sought the refund of amounts paid as use taxes for all of the captives utilized by Burroughs during the period in question. At trial Burroughs moved to amend the complaint to seek a refund only for those captives which were eventually refurbished and sent to marketing for sale or lease.

2 A resale certificate relieves the seller from liability for sales tax if it is taken in good faith from one who is engaged in the business of selling tangible personal property and who holds a seller’s permit. (Rev. & Tax. Code, § 6092.)

3 In an action for refund of sales or use taxes, the taxpayer has the burden of proof to show that he is entitled to his claim. (Honeywell, Inc. v. State Bd. of Equalization (1982) 128 Cal.App.3d 739, 744 [180 Cal.Rptr. 479].)

4 Revenue and Taxation Code section 6051 provides in relevant part: “For the privilege of selling tangible personal property at retail a tax is hereby imposed upon all retailers at [stated rates] of the gross receipts of any retailer from the sale of all tangible personal property sold at retail in this state . . . .”

All further statutory references are to the Revenue and Taxation Code.

5 A use tax is “[a]n excise tax . . . imposed on the storage, use, or other consumption in this state of tangible personal property purchased from any retailer ... for storage, use, or other consumption in this state ....”(§ 6201.)

6 “It is the intent of the use tax merely to supplement the sales tax by imposing upon those subject to it a tax burden equivalent to that of the sales tax with the same specific exemptions in each case. ” (Traynor, The California Use Tax (1935) 24 Cal.L.Rev. 175, 176.)

7 “Use” is defined by section 6009 as including “the exercise of any right or power over tangible personal property incident to the ownership of that property, . . . except that it does not include the sale of that property in the regular course of business.”

“Storage” is defined by section 6008 as including “any keeping or retention in this State for any purpose except sale in the regular course of business or subsequent use solely outside this State of tangible personal property purchased from a retailer.”

8 A Board regulation, dealing with the use of property in manufacturing and generally applicable to manufacturers, producers and processors, provides: “(a) Tax applies to the sale of tangible personal property to persons who purchase it for the purpose of use in manufacturing, producing or processing tangible personal property and not for the purpose of physically incorporating it into the manufactured article to be sold. Examples of such property are machinery, tools, furniture, office equipment, and chemicals used as catalysts or otherwise to produce a chemical or physical reaction such as the production of heat or the removal of impurities. [H] (b) Tax does not apply to sales of tangible personal property to persons who purchase it for the purpose of incorporating it into the manufactured article to be sold, as, for example, any raw material becoming an ingredient or component part of the manufactured article.” (Cal. Admin. Code, tit. 18, § 1525, hereafter Reg. 1525.)

In Kaiser, the Supreme Court concluded that this regulation constituted a statement of the primary purpose test. (24 Cal.3d at p. 193.)

9 Citing the so-called “first use” rule, the Board also argues that when the captives were first used for interactive testing with other components a taxable use occurred. As Burroughs correctly points out, the distinction between the “first use” and the “primary purpose” tests is not merely semantic. First use implies that the use that is first in time is taxable regardless whether that use is the primary use for which the property was purchased. Under such a first use test, any less than insignificant use would probably suffice. The primary purpose test, in contrast, presupposes two or more uses with taxation ultimately dependent upon which use is paramount.

In our view, it is not the timing of the use which is decisive; it is rather whether a substantial intervening use of property has occurred at any time which is primarily undertaken for a purpose other than its subsequent resale. Thus Burroughs could not avoid a use tax by first placing the captives on display for a week, and then using them exclusively for testing for a year or more thereafter and ultimately refurbishing and selling them. Such a substantial intervening use tends to negate any claim that the property was purchased primarily for resale rather than for testing.