Innovation and the Economic Frontier

8/14/2019 Innovation and the Economic Frontier

1/12

TTeecchhnnoollooggyy iinnnnoovvaattiioonn aanndd tthhee

lliimmiittss ttoo eeccoonnoommiicc ggrroowwtthh

WWhhyy eeccoonnoommiieess ddoonnttggrrooww ffaasstteerr??

bbyy JJaavviieerr CCaabbrreerriizzoo--EEMMBBAA ccllaassss 22000088

AApprriill,, 22000088


2/12

Javier Cabrerizo- Insead EMBA 2008-Implementation Essay- Macroeconomics

- 2 -

Table of Contents

INTRODUCTION .......................................................................................................................................3

TECHNOLOGY IS FROM MARS, ECONOMY IS FROM VENUS...............................................................4

IT AND PRODUCTIVITY: ROBUST AND MIXED RELATION.......................................................................7Why IT productivity improvements do not propagate equally across countries? .........................8Why IT productivity improvements do not propagate equally across sectors? .............................9Why are some firms more successful than others when applying IT innovation? ........................10

CONCLUSSION: DIFFUSION AND ADOPTION MATTER MOST.............................................................11

SOURCES................................................................................................................................................12


3/12


- 3 -

INTRODUCTION

Economic growth in the last 100 years has shown a surprisingly consistent behavior: an almost

constant rate of long run growth of GDP of 2,5-3% per year. This is traditionally known as the

technology frontier, implying that for a developed economy to continue growing it mustinnovate, but that this innovation is only translated into economic growth via increased

productivity, at the mentioned rate of 3% per year.

In contrast with this view of the growth rate of developed economies, we have the speed of

technological innovation that can be seen across multiple industries like semiconductors,

biotechnology, nanotechnology, genomics, IT, etc. In these and multiple other areas, the

innovation rate is much faster than a 3% per year. In fact, in many cases we can see

exponential growth rates with constant or even accelerating growth factors that multiply, not

merely add, to the previous year situation. For example we can see how the Moore Law predicts

double capacity of integrated circuits every 18 months (and this law has been happening for

the last 20 years).

So why does it happen that the very fast technological innovation that we see across multiple

industries is only translated into economic growth at a 3% per year?

We will try to analyze this phenomenon and answer the following critical questions:

1- Why does the economy grow only at a predictable 3% per year, when some

fundamental technologies show growth rates of x2 (100%)or more per year?

2- How do other processes, like the diffusion, adoption and leverage of innovation affect

the ultimate impact on the economy growth of the technological innovation growth?

3- How do external factors like regulation, demand and competition can accelerate

adoption and hence faster translate the technological innovation into economic

growth?

We will focus in one industry, IT. We will analyze its growth rates and how it is impacting

productivity and growth in the economy as a whole, but trying to understand which sectors

have benefited from higher productivity provided by their investment in IT and which sectors

have not seen this benefit. We will also explore differences across countries and identify how

external factors affect the impact on productivity.

IT is an excellent area to analyze because it has been extensively studied on the back of the

very special period that constituted the New Economy era, that is the 90s till 2001 with the

burst of the technology bubble. During that period, the US economy significantly improved its

productivity growth, moving from 1,4% during 1973-1995, to 2,4% from 1995-2000. How much ofthat growth was due to the impact of IT, and how much growth has IT brought to the US

economy after the burst of the bubble, are two crucial elements to understand the impact of IT

innovation in the economy.

In the end, technological innovation is at the center of economic growth. However, there seem

to be significant differences between the speed at which different firms, sectors and countries

adopt the same technologies. It seems that, for technological innovation to be translated into

real economic growth, it needs to be co-developed with other innovations in business processes

that can extract all the benefits from the technological innovation.


4/12


- 4 -

TECHNOLOGY IS FROM MARS, ECONOMY IS FROM VENUS

An analysis of the history of technology shows that technological change is exponential,

contrary to the common-sense "intuitive linear" view. So we won't experience 100 years of

progress in the 21st century -- it will be more like 20,000 years of progress (at today's rate). The"returns," such as chip speed and cost-effectiveness, also increase exponentially. There's even

exponential growth in the rate of exponential growth.

The paragraph above can be read in a famous article published in 2001 by Ray Kurzweil, one of

the best-known researchers of the evolution of technology. In his works Kurzweil concludes that

we are doubling the rate of progress every decade; in other words, we will see a century of

progress-at todays rate- in only 25 years. In different ways, on different timescales, and for a

wide variety of technologies ranging from electronic to biological, the acceleration of progress

and growth applies.

He enunciates the Law of Accelerating Returns in which he states that the rate of progress of

an evolutionary process increases exponentially over time the "returns" of an evolutionary

process (e.g., the speed, cost-effectiveness, or overall "power" of a process) increaseexponentially over time... as a particular evolutionary process (e.g., computation) becomes

more effective (e.g., cost effective), greater resources are deployed toward the further progress

of that process. This results in a second level of exponential growth (i.e., the rate of exponential

growth itself grows exponentially).

Kurzweils analysis of technological evolution also introduces the idea of paradigm shift,

explaining that a specific paradigm (a method or approach to solving a problem, e.g.,

shrinking transistors on an integrated circuit as an approach to making more powerful

computers) provides exponential growth until the method exhausts its potential. When this

happens, a paradigm shift (i.e., a fundamental change in the approach) occurs, which enables

exponential growth to continue.

The paradigm shift rate (i.e., the overall rate of technical progress) is currently doubling

(approximately) every decade; that is, paradigm shift times are halving every decade (and the

rate of acceleration is itself growing exponentially). So, the technological progress in the twenty-

first century will be equivalent to what would require (in the linear view) on the order of 200centuries. In contrast, the twentieth century saw only about 25 years of progress (again at

An example of the Law of

Acceleration Returns: "Moore's

Law."

Gordon Moore, then Chairman ofIntel, noted in the mid 1970s that

we could squeeze twice as many

transistors on an integrated circuit

every 24 months. Given that the

electrons have less distance to

travel, the circuits also run twice

as fast, providing an overall

quadrupling of computational

power


5/12


- 5 -

today's rate of progress) since we have been speeding up to current rates. So the twenty-first

century will see almost a thousand times greater technological change than its predecessor.

In these terms, technology is one manifestation (among many) of the exponential growth of the

evolutionary process. The exponential growth of computing is a marvelous quantitative example

of the exponentially growing returns from an evolutionary process. We can also express the

exponential growth of computing in terms of an accelerating pace: it took ninety years toachieve the first MIPS (million instructions per second) per thousand dollars, now we add one

MIPS per thousand dollars every day.

It is also important to note that in the evolution of technology we need to distinguish between

the "S" curve (an "S" stretched to the right, comprising very slow, virtually unnoticeable growth--

followed by very rapid growth--followed by a flattening out as the process approaches an

asymptote) that is characteristic of any specific technological paradigm and the continuing

exponential growth that is characteristic of the ongoing evolutionary process of technology.

Specific paradigms, such as Moore's Law, do ultimately reach levels at which exponential

growth is no longer feasible. Thus Moore's Law is an S curve. But the growth of computation is an

ongoing exponential. In accordance with the law of accelerating returns , paradigm shift, also

called innovation, turns the S curve of any specific paradigm into a continuing exponential. A

new paradigm (e.g., three-dimensional circuits) takes over when the old paradigm approachesits natural limit. This has already happened at least four times in the history of computation.

This "law of accelerating returns" applies to all of technology, indeed to any true evolutionary

process, and can be measured with remarkable precision in information-based technologies.

There are a great many examples of the exponential growth implied by the law of accelerating

returns in technologies as varied as DNA sequencing, communication speeds, electronics of all

kinds, and even in the rapidly shrinking size of technology

DNA sequencing costs:

When the human genome scan started in

1966, critics pointed out that given thespeed with which the genome could then

be scanned, it would take thousands of

years to finish the project. Yet the fifteen-

year project was nonetheless completed

slightly ahead of schedule.

(Growth Rate: DNA sequencing cost is now

doubling every 12 months)

Predictability of Internet:From the perspective of most observers,

nothing was happening until the mid 1990s

when seemingly out of nowhere, the world

wide web and email exploded into view. But

the emergence of the Internet into a

worldwide phenomenon was readily

predictable much earlier by examining the

exponential trend data.

(Growth Rate: Number of Internet hostsdoubles every 12 months)


6/12


- 6 -

We can then conclude that in multiple technology areas, we can see a constant growth factor.

That is, if plotted in a linear scale, we can see parabolic trends; if plotted in a logarithmic or

exponential scale we can see a line, which allows us to predict what the next stages of

evolution are going to be.

How does this technology progress rate driven by exponential growth and paradigm shifts,

compare with the progress in the economy?

Its a predictable growth after all!! Seems like, along years, no matter what historical events

occur, the growth rate of the economy is pretty constant and we can predict with astonishingaccuracy where we are going to be in 80-100 years! Just like we saw with technology, the

economic system shows a similar consistency and predictability in its growth rate!

However, there is one shocking difference: while the growth rate of the technological progress

frequently shows a multiplying factor of 2 (that is, it doubles or grows 100% in 12 months), in the

case of the economic growth, we only see a growth rate of ~3% for Real GDP and ~1,8 for Real

GDP per capita.

Leaving aside the surprising parallel behavior observed, we want to focus our attention in the

following issue: Why is it that an exponential growth rate in technology, when transferred into the

economy as a whole, looses intensity and can only reflect a constant linear growth rate of 3%?

In order to understand how innovation, and particularly technological innovation is translated

into economic growth, and why it does so at an apparent slow pace, we need to understand

how innovations are adopted and which elements favor or hamper the adoption of innovation.

In order to do so, we believe that the key factor to observe is the evolution of productivity.

However, looking at aggregated productivity levels in any given country can hide some

important sector-specific differences. Consequently, we are going to analyze how innovation in

IT has impacted productivity in the last years. We will do that by looking at (i) the productivity of

IT sector itself, (ii) productivity of sectors that use IT intensively, and (iii) other non-intensive IT

sectors. We will then observe how these differences impact the country-aggregated

productivity and extract some conclusions that will help understand why (or why not)

technological innovation is transmitted to economic growth via increased productivity.

Predictability of Economic

Growth:

Real GDP per capita grows

at 1,85% and seems to do it

with admirable consistency.

Looks like, not onlytechnology, but also

economic growth is

predictable with significantaccuracy!!


7/12


- 7 -

IT AND PRODUCTIVITY: ROBUST AND MIXED RELATION

The impact of IT on productivity has been extensively analyzed after the rise and fall of the so-

called New Economy. This is an economic period, normally considered to span between 1995-

2000, in which US productivity grew at levels of 2,4%, compared with historical levels of 1,4%during the period 1973-1995. It was assumed that faster productivity growth was fueled by

investments in information technology hardware and software. Interestingly, after the burst of

the Internet bubble in 2000, productivity in US has risen at 1,8% or higher, while IT investments

have languished or even slipped into negative territory after 2001.

In order to understand how IT impacts productivity, we observe the evolution of Average labor

productivity (ALP). ALP is defined as the ratio of output to hours worked. Under assumptions of

constant returns to scale and competitive factor markets, the growth of ALP can be

decomposed into three sources. The first is capital deepening, defined as the increase in capital

services per hour worked. The idea is that workers become more productive if they have more

or better capital (equipment, structures, or land) with which to worka faster computer for an

accountant, say, or a more sophisticated numerically controlled machine tool for amanufacturing worker. The second source of labor productivity growth is a gain in labor quality,

defined as an increase in labor input per hour worked. Labor quality reflects changes in the

composition of the workforce: as firms shift their hiring toward workers with more experience and

education, for example, average labor productivity rises. The third source is total factor

productivity (TFP) growth, which reflects all labor productivity growth that is not attributable to

capital deepening or labor quality gains. TFP growth is often associated with technological

progress but also reflects changes in utilization rates, reallocations of resources among sectors,

increasing returns to scale, and measurement error.

The table below shows the data for US in the period 1959-2003 and specifically compares the

period 1995-2003 when biggest growth in IT productivity occurred, with previous periods.

US productivity growth 1995-

2003: the role of IT.

60 percent of the increased

capital deepening in 1995

2003 was attributable to IT,

although information

processing equipment and

software accounted for only

about one-quarter of privatefixed investments in this

period.

IT production accounted for

more than 35 percent of the

increase in aggregate TFP, far

exceeding the 5 percent

share of IT goods in

aggregate output.


8/12


- 8 -

Of the 1.57 percentage point increase in ALP growth after 1995, 0.86 percentage point was due

to capital deepening and 0.80 percentage point due to faster TFP growth, with a small decline

in labor quality growth of 0.09 percentage point. The 35% contribution of IT to the increase in

aggregate TFP reflects the exceedingly high rates of technological progress in IT production and

is manifest in the 9.2 percent per year decline in the price of IT output in 1995-2003. Lying behind

this is the enormous fall in the quality-adjusted prices of IT since 1995, which has its roots intechnical progress in the semi-conductor industry. Rapid improvements in the power of semi-

conductors lead to big increases in productivity growth in the IT producing sectors. Moores Law

seemed to accelerate post 1994 (as predicted by Kurzweil) and this fall in the price of a key

input lowered prices across a whole range of products in the IT producing sectors. As the price

of IT products plunged, firms deepened their use of IT capital and this was naturally strongest in

sectors that intensively used ICT. Increasing usage of IT per worker hour, increased output per

hour tremendously.

Although IT progress is widely available and can be adopted by any firm, not all sectors were

capable of reaping the same benefits from IT. Moreover, the same sectors that did benefit from

IT in the US did not obtain the same benefit in Europe. While in the US, productivity effectively

boosted in the period 1995-2001, and mainly did so in the sectors that were heavy users of ICT(like retail, wholesale and finance), in Europe this did not happen. Productivity in the same

period did increase in ICT producing sectors (computing, semiconductors), like it did in the US,

with only a marginal 0,3% difference between US and. The graphic below, based on research

conducted at LSE illustrates the point:

Why IT productivity improvements do not propagate equally across countries?

First, lets take a look at the differences across countries. It is impossible to understand the

differences between the US and Europe, without investigating the external factors at work in

each country. Fewer external barriers to innovation and growth appear to have existed in the

US, which help explain the countrys stronger performance after 1995. When comparing

productivity increase in US and Europe, McKinsey finds three key differences: Regulatory,

Governance and Domestic Demand.

Regulatory restrictions help explain the speed at which innovations are diffused in the market.Higher competitive pressures would favor a faster adoption of innovations in order for firms can

The impact of IT on US and

European productivity:

In the US, productivity growth

accelerated by 3.5 percentage

points per annum in the ICT-

using sectors (from 1.2 per cent

p.a. pre-1995 to 4.7 per cent

p.a. post 1995). This did not

happen in Europe, which

remained at a constant 2%

growth rate in the same period.

Since IT is available throughout

the world at broadly similar

prices why were Europeanfirms not able to reap the same

benefits from IT as their US

counterparts?


9/12


- 9 -

remain competitive. Conversely, less competitive intensity due to regulatory protection, poses

no incentive to the adoption of innovations. For example, French hypermarket protection given

by zoning laws protects hypermarkets from innovative competitors; and in the German banking

sector small, state-owned and cooperative banks are, because of their ownership structure, not

exposed to shareholder pressure from capital markets. Also, the US mobile telecom market

experienced productivity growth during the 90s of 15%, while in Europe this was 25%. The

biggest factor explaining this difference is the regional license auctions in the US, where morethan 50 mobile providers serve fewer than 200.000 customers each. In France and Europe, 3 and

4 providers serve 10 million customers each.

The domestic demand differences lead to differences in productivity in sectors with a grid

network, in which higher demand leads to higher capacity utilization. For example the fixed-line

network in France and Germany is much less utilized than in the US and leads to a 40%

productivity disadvantage. In retail banking, the productivity of the network of branches and

ATM is also affected by utilization; in the US, bank customers conduct more transactions than

their German and French counterparts, leading to a 6% disadvantage in productivity in Europe.

In automotive, 11% of productivity gap between EU and US is explained by the demand of light

trucks in US that are easy to manufacture and deliver high value added per hour worked. In

Europe, the demand focuses on sophisticated vehicles that create less value added per hourworked.

Why IT productivity improvements do not propagate equally across sectors?

Second, lets look at the differences between sectors within the US. When analyzing the

evolution of different sectors, research from McKinsey concludes that in the period1995-2001, 43

of 58 US sectors representing 73% of GDP, experienced productivity gains. However, productivity

gains were not distributed evenly across alls sectors. In fact, 6 sectors of the economy,

comprising 32% of GDP, contributed 66% of the gross productivity gains experienced in the US

economy, and 76% of Net productivity growth. These 6 sectors were: semiconductors, wholesale,

securities, retail, computer assembly and telecom. The following table summarizes the findings:

Telecom, Semiconductors, Computer Manufacturing 8% of GDP 36% of US growth

Wholesale, securities and retail 24% of GDP 40% of US growth

The other 52 sectors 68% of GDP 24% of US growth

It is understandable that ICT producing sectors reap an immediate benefit from the intense

period of IT innovation. Moores Law directly impacts their productivity; innovations in the

telecom market, particularly in data transmission and mobility also have a direct impact in this

sector. But what was the technological innovation behind the other top 3 sectors?

In Retail Banking and securities, new technologies gave rise to further back-office automation

and new sales channels like online banking, online trading and call centers. For example, in the

US securities sector, online channels allowed to process explosive trading volumes without

adding traders. The shift to new forms of electronic payments also improved productivity.

In Wholesaling, distribution centers benefited form warehouse automation technology like

barcodes and scanners, and warehouse systems like inventory control and tracking. These

allowed the partial automation of the flow of goods, dramatically reducing the need for labor in

the picking, packing and shipping areas which counted for more than 40% of labor needs.

In essence, country and sector specific conditions have a huge impact on the velocity at which

innovations are adopted and hence translated into productivity growth.


10/12


- 10 -

Why are some firms more successful than others when applying IT innovation?

After analyzing the impact of IT innovation at a country and sector level, we want to understand

what makes the difference at the firm level. Successful IT innovations had the following

characteristics. First, they were tailored to sector specific characteristics and were linked to

performance levers. For example in Retail, warehouse management systems, transport

management systems and vendor coordination system brought improvements in merchandisevelocity. They reduced non-labor costs like inventory carrying and inventory costs in distribution

and logistics. In retail banking, credit scoring software and underwriting tools enabled

automation of manual processes associated with credit verification and authorization.

Second, they were deployed in a sequence that built capabilities over time. Highest productivity

gains occurred where IT and business skills were developed overtime in a process that allowed

firms to leverage previous investments in IT. Retailers first developed automated data capture

and storage and then used this data to develop enhanced decision-support capabilities. When

companies did the effort to deploy IT without the prerequisite infrastructure components, it

yielded little impact.

Finally, they coevolved with managerial and technical innovation. A close link between new

managerial models and new technical innovation was crucial in the adoption of imagingtechnologies for automating check processing and loan processing in retail banking .

Essentially, innovations occur in 2 different ways: in the form of new products or services (like

mobile telephony or web banking) or in the form of business processes (back office automation

in baking, optimization of supply chain). Innovative products and services help firms shift sales to

higher value added goods, while best practice business processes improved operational

performance. The graph below describes how this process works:

But to escalate from the firm level to the sector and economy level, we need to understand

something else. Innovation itself is only a partial explanation of economy wide productivity

increase. The Diffusion of Innovation (through replication by other firms and sectors) heightens

the impact of innovations in product, process or service done by individual firms. For example,

the innovative business processes adopted by Wal Mart in US (efficient logistical chain,

Business

And

Technology

Innovations

Drivers for productivity

performance

Increase output witha given input

Reduce input for

a given output

Consolidate to

better leverage

Improve operational

performance

Sell higher value

goods

Sell more goods to

increase

capacity utilization

Close gap to bestpractice operations

Find innovative

processes to

improve operations

Create innovative

High value added

Products & services

Shift to higher

Value goods withinProduct portfolio

Leverage

Leverage

Diffusion

Diffusion

Source: McKinsey Global Institute

Business

And

Technology

Innovations

Drivers for productivity

performance

Increase output witha given input

Reduce input for

a given output

Consolidate to

better leverage

Improve operational

performance

Sell higher value

goods

Sell more goods to

increase

capacity utilization

Close gap to bestpractice operations

Find innovative

processes to

improve operations

Create innovative

High value added

Products & services

Shift to higher

Value goods withinProduct portfolio

Leverage

Leverage

Diffusion

Diffusion

Source: McKinsey Global Institute


11/12


- 11 -

electronic data interchange,) pushed down margins and yielded productivity enhancing

efforts by competitive firms, encouraging the diffusion of best practices.

Even when diffusion is complete, the ability of firms to Leverage the innovation depends upon

the firms achieving sufficient scale. In some instances, firms leveraged innovations by

consolidating and achieving necessary scale. For example in retail banking, sector

consolidation dramatically reduced the need for clerical and administrative personnel. Anappropriate industry structure is critical to fully leverage the potential benefits determining a

slower or faster productivity growth. For example, the US mobile telecom industry benefited from

the additional spectrum being auctioned, which led to increased competition, price declines,

higher usage levels and improved performance levels.

As we can see, firm and sector specific characteristics can mutually reinforce each other to

facilitate or hinder the adoption of innovations. Companies are motivated to invest in IT to

increase their productivity, to gain an advantage over a competitor or to increase profitability.

IT investments are easy to replicate, but they were more likely to remain differentiating when

coupled with othercompetitive advantages like scale, significant changes in the business

process and associated learning effects. In highly competitive environments, IT investment

played a role in increasing productivity and, for some time, increasing profitability of the firm.

However, intense competition tended to devolve away from profits to consumer surplus in theform of lower prices and higher quality. At some moment in time, these IT investments simply

become a cost of doing business. The sector will benefit from increased productivity only in so

far as laggard firms were still catching up to the best practices of leading firms. This process is

crucial to understand how innovations start from a competitive pressure in firms, propagate

across the sector and end-up impacting the productivity of the economy as a whole.

CONCLUSSION: DIFFUSION AND ADOPTION MATTER MOST

Economic growth shows a surprisingly consistent behavior, with GDP growth levels in the region

of 3% per year over extended periods of time. At the same time, technological innovation shows

a growth rate of 100% (x2) or even more per year, which is explained by the Law of

Accelerating Returns. Why is technological innovation not being transplanted faster into

economic growth? The main conclusion of this paper is that Innovation creation grows at the

above-mentioned rates but it is the diffusion,adoption and leverage of innovations by firms that

ultimately impacts the economic growth of the economy. We concluded that sector and

country specific conditions have a direct (positive or negative) impact in the diffusion of

innovations and in the impact of innovations in economic growth.

We illustrate this conclusion by analyzing the impact of IT in productivity in the period 1995-2001.

IT has had an impact in US productivity in that period, making it move from 1,4% to 2,4% per

year; however the same is not true in Europe. We specifically looked at external factors like

regulatory environment, governance models and demand structure that directly affect thecountry specific productivity growth levels and help explain the differences between US and

Europe.

We also analyzed the differences across sectors and we found that innovations do not

propagate evenly across sectors. In fact, for the period 1995-2001, 6 sectors representing 34% of

GDP brought 66% of the productivity growth of the US economy. We have concluded that forIT

innovations to be effective at enhancing productivity there needs to be a connection between

IT and the business. IT innovations can have big impact in sector productivity if they are sector

specific, are deployed in a sequence and allow the co-evolution of IT and managerial systems.


12/12


- 12 -

SOURCES

- Americans do IT better: US Multinationals and the productivity miracle. Centre for

Economic Performance, London School of Economics. Paper 788, April, 2007.

- Productivity and ICT: a review of the evidence. Centre for Economic Performance,London School of Economics. Paper 749, August 2006.

- US productivity after the dot com burst. McKinsey Global Institute. December 2005.

- Information Technology and Productivity, it aint what you do, its the way you do IT.

EDS Innovation Research Program, London School of Economics. October 2005.

- The world technology frontier. Francesco Caselli, Wilbur John Coleman II. 2005

- Will the US productivity resurgence continue?. Current Issues in Economics and

Finance. Federal Reserve Bank of New York. December 2004.

- Whatever happened to the New Economy? McKinsey Global Institute. November

2002.- Information technology and Economic Growth in Canada and US. Harchaouni,

Tarkhani, Jackson, Armstrong. Monthly Labor Review, October 2002.

- Structural Change and Technology. A long view. Bart Verspagen. Eindhoven University

of Technology, May 2002.

- How IT enables productivity growth: The US experience across three sectors in the

1990s. McKinsey Global Institute, 2002.

- US productivity growth. Understanding the contribution of IT relative to other factors.

McKinsey Global Institute, 2002.

- The law of accelerating returns. Ray Kurzweil. November 2001.

Innovation and the Economic Frontier

Documents

Innovation and the Economic Frontier