Transcript Short Background on the Semiconductor Industry

Short Background on the
Semiconductor Industry
• The leading industry in the IT revolution.
• Has significant influence on global economic cycles
• Relevant to almost all sectors of IT.
• Annual revenues of more then $150 billion, with average
year to year growth of 17%.
The Fabless Model
• only chip design.
• No foundry.
• Usually more focus then Integrated Device manufactures
(IDM).
• need to sale less to become profit and as a result are more
flexible and can address niche markets as well as main
segments.
The Fabless Model
Fablees History
• began becoming relevant in the 70’s.
• rapid growth in 90’s
• Today, the fabless companies total annual revenues
are more then $23 million which represent 15% of
semiconductor annual revenues.
• for the last decade fabless growth rates were higher
then the semiconductor growth rates and it is expected
that in 2010 fabless annual revenues will represent
about 50% of the semiconductors revenues.
Fabless Statistics
Fabless Companies Primary Business by Market Sector
Market Sector/Application
Percentage
Wired Communication
31%
Wireless Communication
23%
PC Peripherals
17%
Consumer
14%
PC
10%
Industrial Medical
2%
Defense Industries
2%
Automotives
1%
Fabless Statistics
Fabless Companies Primary Business by technology
Technology Area
Percentage
Analog/Digital
Logic
Microprocessors
DSP
Memory
37%
21%
18%
17%
7%
Fabless Statistics
Fabless Companies Geographic distribution
Geographic area of Fabless
companies
Percentage of total revenues
/Number of companies
North America
Europe
Asia
Israel
84% / 455
7% / 55
5% / 75
4% / 30
Fabless Statistics
Geographic Distribution of Foundries serving Fabless Companies
Company & location
Percentage
TSMC - Taiwan
UMC - Taiwan
Chartered semi - US
American Microsystems - US
Other - worldwide
41%
24%
10%
7%
18%
Background to the Israeli
Semiconductor Industry
• Very strong microelectronic academic departments from early 60’s.
• many Israeli engineers emigrated, in the 60’s, 70’s and 80’s to the silicon
Valley to work in semiconductor companies.
• First multinational semiconductor companies open Israeli R&D centers in
the late 60’s and early 70’s (Motorola 64, IBM 72 and Intel 74).
• Israeli engineers from silicon valley return to Israel in the 80’s and 90’s.
• In 1984, first multinational semiconductor company open a foundry in
Israel – Intel’s fab in Jerusalem.
• structure changes in the Israeli economy – restructure of the military
industries, the stability program, liberalization and globalization.
Background to the Israeli Fabless
Industry
• Small country for a semiconductor industry – fabless companies are
smaller companies and have more floatability in changing technology sector.
• Distance from the market – it is easier to export design then products.
• Weak marketing abilities and weak in selling to the final costumer – selling
to the semiconductor companies or selling through OEMs.
• Strong R&D abilities – the most important factor is R&D.
• Shortage of capital and weak capital markets – less capital needed to
fablees companies and less risk.
• Unstable area – fabless companies have more floatability in changing
location.
Israeli Statistics
• chip design and semiconductor is the most important sector in the Israeli
Hi-tech.
• In 2000, more then quarter of the M&A deals in the Israeli Hi-tech were
related to semiconductors and chip design.
• In 2000, Galileo was acquired by Marwell for $2.7B and Visiontech was
acquired by Broadcom for $700M.
•In 1999, DSPC was acquired by Intel for $1.6B and Libit was acquired by
TI for $465M.
• Today, there are at least 3 semiconductor startups among the top 10
private startup in Israel (Saifun, Mobilian, Ezchip).
Case Studies – the Sample
• 15 startup companies (11 chip design companies and 4 related industries).
• Fairly representative with small biases to more successful startups.
• Technologies applying to 7 different categories.
• The companies were established in 4 phases of the Israeli VC industry.
• large difference in success (failures to tremendous successes).
• Few geographic areas.
Methodology
• Background.
•Initial Mapping of the Israeli chip design companies.
• Deep research on 15 chip design startup companies.
• Interviews with most of these startup companies.
• Analyzing the sample results.
• Building an Initial Growth Model.
• Initial policy implications.
• Understanding the fundamentals of Italy chip design sector and adjusting
the model to Italy.
• Increasing the sample - checking and expanding the growth model.
• Policy Implications.
Entrepreneur Background
• Very strong educational background (10 PhD, 3 M.Sc EE and 2 B.Sc EE).
• Strong Work experience background (7 MNE, 5 SU, 3 IDF, 4 Academia).
• Strong managerial experience (7 manager positions, 2 entrepreneurs).
• Strong core additional team (5 very strong, 5 strong).
• Strong R&D position experience (15 R&D positions).
Trigger and Foundation
• Most startups were lead by the scientists and the technology developer (12
startups).
• We believe that half of the startup would have established even if the
market condition were less attractive (driven by very strong technology).
• We found few kind of triggers (development in academia, development
during previous works and problem solving).
• At least 7 startups were triggered by the Nasdaq good conditions.
• Two startups were in technology incubators.
• Two startup were spin-offs from established companies.
• most startup had long maturation process from initial idea to
establishment.
Business Model
• The technology innovation: 5 specific products, 6 platform technology
and 4 radical technology.
• Becoming more focus in areas of development: 6 more focus, 6 no
change and 3 broadened.
• Marketing model: 5 direct sales, 6 OEMs and 8 strategic and marketing
partnerships.
• Growth Profiles: 2 IPO and M&A, 2 IPO, 2 fast M&A, 1 M&A as last
rescue and 8 not clear yet.
Optimal Growth Profiles
Strong entrepreneur background
Additional strong initial team
Strong technology in a significant segment
Path 4
OEM or/and Strategic agreements
Fast M&A
Path 3
IPO
IPO
Path 2
M&A
Path 1
Explanation of the Model
Path 1: companies with all characteristics of optimal profile of growth, which
underwent successful IPO and were later acquired by a large multinational company
(2 companies in our sample).
Path 2: companies with all characteristics of optimal profile of growth, which had
gone trough a successful IPO (2 companies in our sample).
Path 3: companies with all characteristics of optimal profile of growth, which were
acquired by a large multinational company in the beginning of their marketing and
strategy alliances efforts (1 company in our sample).
Path 4: companies that seems to have all characteristics of optimal profile of growth
but are only in an initial phase of marketing and strategic agreements (2-5
companies in our sample).
Path1 (Very Successful Companies)
companies with founders with very strong background and experience and
very strong technology. These factors often leverage the company into
strong OEM and strategic agreements with leading multinational
companies. The companies’ initial success enables them to go through a
successful IPO, which helps them keep growing. Later on, when the
technology gap closed and the market consolidation in their segment began,
they were attractive enough to be acquired be one of the competitors for a
significant amount. In Israel, we identified 2 companies, which have gone
through path1.
Path2 (Tremendous Success or
Moderate Success)
path2 is divided to 2 types of companies. Type1 (tremendous success) companies with the same characteristics of companies in path1 (or
maybe even stronger background and technology), which went through
successful IPO and succeeded in leveraging their initial advantages
better than companies from path1, and as a result succeeded in staying
independent companies despite the consolidation on the market. These
companies are market leaders in their segment and become large
multinational companies. Type2 (moderate success) - companies with
weaker characteristics of companies in path1 (weaker background and
technology or technology which applies to less significant market
segments), which went through successful IPO but were not attractive
enough when the market consolidation began and as a result were not
acquired and remained a medium company. In Israel, we identified 6-8
companies, which have gone through path2 (most of them from type2
but same may become type1).
Path3 (Successful Companies or
Successful Technology Exits)
Paths3 is divided into 2 types of companies. Type1 – companies with very strong
technology in a significant segment that due to weak managerial capabilities or due
to entrepreneur preferences decided to make an early sale exit. These companies due
to their very strong technology, were able, in spite of their early stage, to be
acquired for a significant amount. Type2 – companies with strong technology but
with weak managerial abilities and/or work experience that their only chance to
have any profit from the startup was to sell their technology/company in an early
stage for a moderate amount.
In Israel, we identified 3-4 companies, which have gone through path3 (only 1 was
from type1).
Path4 (Young Startups)
companies that suit the model but are still too young to determine which path they
will pursue.
We believe that most companies that don’t suit the growth profile have small
chances of becoming successful companies in the chip design sector.
Most of the startups that suit the model fit into it sooner or later.
Research Summary
We believe that most companies that don’t suit the growth profile have
small chances of becoming successful companies in the chip design sector.
Most of the startups that suit the model fit into it sooner or later. We
identified between seven to ten companies in our sample that fit this model,
and five of these were already successful (between moderate success and
tremendous success).
We believe that in most cases our model can be helpful to policy makers in
choosing an appropriate policy/program to use in order to promote chip
design companies and the successful growth of the chip design sector.
We also believe that this model can be helpful to venture capitalists as a
checklist prior to any investment in this sector. Moreover, this model can
help the entrepreneur to decide whether they are ready to establish their
own startup.