Can the telecom equipment industry afford accelerating technical advance?

Telecommunications Policy

Volumn 21, Issue 8, 1997, Pages 697-707

  Hicks, Donald A ^a   O'Brien, Daniel M ^b  

^a UNIVERSITY OF TEXAS AT DALLAS   (United States)

^b Political Economy Doctoral Program   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

COSTS; ELECTRONIC EQUIPMENT MANUFACTURE; LARGE SCALE SYSTEMS; PUBLIC POLICY; TELECOMMUNICATION SYSTEMS;

TELECOMMUNICATION EQUIPMENT INDUSTRY;

TELECOMMUNICATION SERVICES;

EID: 0031250456     PISSN: 03085961     EISSN: None     Source Type: Journal    
DOI: 10.1016/S0308-5961(97)00040-2     Document Type: Article

References (22)

1. Blackburn, J D (ed.), Time-Based Competition: The Next Battleground in American Manufacturing. Business One Irwin, Homewood, IL, 1991.
2. Stalk, G and Hout, T M, Competing Against Time. The Free Press, New York, 1990.
3. Schmenner, R W, The merit of making things fast. Sloan Management Review, 1988, 30 (1), 11-17.
4. Stonich, P J, Time: the next strategic frontier. Planning Review, 1990, 18, 5-7 and 46-48.
5. Bockerstette, J A and Shell, R L, Time-based Manufacturing. (Norcross, GA: Institute of Industrial Engineers, Industrial Engineering and Management Press). McGraw-Hill, New York, 1993.
6. Merrills, R., How Northern Telecom competes on time. Harvard Business Review, 1989, 67 (4), 108-114.
7. Stonich, op cit Ref 4.
8. Pisano, G P, Russo, M V and Teece, D J, Joint ventures and collaborative arrangements in the telecommunications equipment industry, In International Collaborative Ventures in US Manufacturing, Chapter 2, ed. D C Mowery. Ballinger, Cambridge, MA, 1989, p 38; note that the convergence of microelectronics and computer technology has both shortened switching product life cycles and made their development more costly, with development costs approaching $1 billion for equipment that will be obsolete within 5 years.
9. On this point see also Chapuis, R J and Joel, A E, Electronics, Computers and Telephone Switching, North-Holland Studies in Telecommunications, Vol. 13. North-Holland-Amsterdam, 1990.
10. Pisano et al, Ibid, pp 23-70.
11. Chapuis, R J and Joel, A E, op cit Ref 8.
12. Because much telecom equipment is manufactured by large multi-focus companies, a list of producers specializing in such equipment is not likely to be the same as a list of leading producers of such equipment, the bulk of whose efforts is focused elsewhere. Of the 21 market leaders in telecom equipment manufacturing, six specialize in such products and are therefore found on both lists.
13. Across the models for the 23 producers, the shift toward relatively greater R&D was statistically significant at the 0.05 level for 16 producers.
14. While in several cases it is clear that the fit of a linear (OLS) model specification to a specific firm's data is relatively poor, the goal of this portion of the analysis is not to converge on 'best-fit' models for individual firms. Rather it is to determine to what extent a common two-variable model for detecting long-term change yields a common pattern - slope sign and statistical significance - across all the firms in our database. In this way we can detect the ubiquity of the shifts under study.
15. Hutcheson, J D and Hutcheson, G D, Is Semiconductor Manufacturing Equipment Still Affordable?, VLSI Research Inc., Paper presented at the IEEE 1993 International Symposium on Semiconductor Manufacturing, Austin, TX, September 19, 1993; see also Ref 15.
16. Hutcheson, G D and Hutcheson, J D, Technology and economics in the semiconductor industry, Scientific American, January 1996, 274 (1), 54-62.
17. In addition to the net income figure, R&D expenditures, typically expensed and reported in a firm's P&L statement rather than in its cash flow statement, are included here in a firm's total income. R&D comprises a portion of the 'new' cash generated by previous capital investments. On this point see Ref 14, p 13.
18. Only three firms - DSC (0.18), Acme-Cleveland (0.07), and Cognitronics (0.00) - countered the dominant trend of declining new-cash ROI. None of these coefficients achieved statistical significance. Given the expected instability of coefficients generated from regressions involving fewer than two dozen data points, we place primary significance on the pattern of negative coefficients across the common model tested for each firm. Statistical significance at the 0.05 level was achieved for seven (35.0%) of the firms.
19. While one-year lag structures were employed in the phase charts below, the general patterns were found to be remarkably robust using lag structures of up to three years as well as specifications in which the averages of more than one year's R&D expenditures were included in the numerator.
20. Hicks D A. Evolving complexity and cost dynamics in the semiconductor industry. IEEE Transactions in Semiconductor Manufacturing, 1996, 9 (3), 1-9.
21. Hicks, D A and O'Brien, D M, Order amid complexity: patterned investment returns to rapid technical advance. Paper presented at the 24th Annual Conference of the European Association for Research in Industrial Economics (EARIE) in Leuven, Belgium, 31 August-3 September, 1997.
22. Scientific-Atlantic experienced a steady 23-fold (nominal) increase in revenues across the 1975-1994 period, with its R&D share of annual revenues steadily increasing from 3.4% (1975) to 7.4% (1994), even as its R&D share of investment increased from 34.7% to 63.4% over the same period. Meanwhile, the much smaller Andrew, whose revenues increased only eight-fold over the 1979-1994 period and whose R&D share of revenues actually declined from 6.2% to 4.6%, saw its R&D share of investment increase from 35.0% to 48.7% over the same period.