By 1988, the personal computer industry was rapidly shaking off its experimental image and entering mainstream business and household life. For economists or policymakers hoping to quantify the surge in PC manufacturing, ISIC 2910—Manufacture of machinery and equipment not elsewhere classified—serves as a necessary, if blunt, starting point. The code covers an expansive terrain, including everything from agricultural machinery to vending machines, so isolating PC makers requires an extra layer of investigation.
The first step is to extract a registry of firms coded under ISIC 2910 for the relevant geography—say, the United States, Japan, or Western Europe. This registry, while comprehensive, sweeps up companies that have little or nothing to do with computers. Filtering for true PC manufacturers means turning to trade publications, hardware catalogues, and market research reports from the period. Companies that consistently appear in product reviews, advertise in technology magazines, or are profiled in business press as computer makers—IBM, Compaq, Toshiba, Apple, and others—should form the initial core list.
A further clue comes from regulatory filings and import/export data, where available. Many countries tracked “computing machinery” as a separate subcategory in customs documentation, providing another way to triangulate which ISIC 2910-registered firms were actually assembling PCs. Hardware expos, conference proceedings, and patent records from the late 1980s often mention new entrants or innovative upstarts in the field, rounding out the roster.
Next comes measuring production volume. Industry analyst firms like IDC, Dataquest, and local equivalents regularly published reports detailing units shipped, market shares, and—sometimes—breakdowns by product line or manufacturer. Annual reports from the larger firms, particularly if they were publicly traded, often disclosed unit production figures or at least year-over-year growth. Not all firms reported consistently, and some figures were bundled with other “information equipment” categories, so cross-referencing is vital.
To correlate production volumes with chipset supply chain data, analysts should consult trade journals, semiconductor industry reports, and, where possible, statements from chipset suppliers themselves. The supply of microprocessors—most famously from Intel, Motorola, or AMD—was a key bottleneck and a marker of industry momentum. Data on chipset shipments (number, destination, type) is sometimes available in the annual reports of the major suppliers or in the pages of semiconductor market studies. Overlaying PC production surges with spikes in chipset shipments can reveal both coordination and constraints within the industry. During periods of rapid innovation or shortage, news stories about delays or allocations help explain why some manufacturers expanded faster than others.
Another layer is the role of OEM relationships and contract manufacturers, which were already becoming significant by the late 1980s. Some ISIC 2910-registered firms built their own systems end-to-end, while others assembled PCs based on third-party chipsets and motherboards, or even acted as “white box” producers for local brands. Documentation of these relationships—whether in press releases, analyst notes, or supplier lists—adds nuance to the simple production figures.
The data trail is, as ever, imperfect. Product lines shifted quickly, firms entered and exited the market, and reporting standards were anything but uniform. Every step—how companies were classified, how figures were interpolated, how supply chain links were drawn—should be logged with care.
Still, by layering ISIC 2910 firm registries, production volume data, and chipset supply chain information, analysts can construct a reasonable map of the personal computer industry’s expansion in 1988. The landscape is uneven, reflecting local peculiarities and global trends alike, but in the aggregation of these sources, the momentum of the PC revolution becomes unmistakably clear.