SMEs, Productivity and Management: A Research Agenda for ICT and
Business Clusters
[Paper presented at the 2004 Conference on Information Science,
Technology and Management (CISTM), Alexandria, Egypt, 8-10 July 2004]
Jonathan Liebenau
London School
of Economics, University of London
liebenau@lse.ac.uk
G. Harindranath
Royal
Holloway, University of London
g.harindranath@rhul.ac.uk
Gul Berna Özcan
Royal
Holloway, University of London
g.ozcan@rhul.ac.uk
Abstract
In this paper we present a
research agenda for ICT and SME business clusters. Specifically, we reconsider
the effect upon business clusters of new ICT and suggest how worldwide best
practices might reduce the productivity gap. ICT-led productivity growth is a
function of the use of new technologies to improve business practices, enhance
organizational structures, and reach new markets and commercial partners.
Productivity varies among sectors and countries and our work-in-progress uses
comparisons of SME clusters in three leading EU economies (Britain, France and
Italy) and three accession states (Hungary, Romania and Turkey). Our
preliminary findings have a bearing on the general view of ICT-led productivity
growth potential. They imply that common European policies for regional
development and SME clusters may be misguided not because spatial
characteristics don’t matter but because management practices, especially those
related to information use, will always matter more. Virtual clusters and
sector-specific virtual networks are good in principle because they may easily
overcome out-of-region trade barriers (and especially international export
opportunities) both by making firms more aware of business opportunities and by
leading them into more effective competitive behaviours. However, the provision
of infrastructure such as broadband and the dissemination of leading business
software applications are insufficient to ameliorate the productivity gap where
information use behaviour is tied to traditional practices.
Keywords: SMEs, business clusters, ICT, productivity, policies
Introduction
The productivity of small
and medium sized enterprises [SMEs] varies internationally based on many
factors including the ease of access to sources of funding, the skills base in
the locale, and characteristics of regional markets. European Union [EU] policy makers and proponents of extensive use
of e-commerce and other applications of new information and communications
technologies [ICT] claim that productivity will rise when access and knowledge
about new technologies improves. In
this paper we set the foundation for an analysis of the relationship between
access to ICT and the productivity that results from its utilisation. We are
also interested in examining the argument that ICT will diminish the
significance of distance ('death of distance'). We propose to compare the
access that SMEs have to ICT in three EU countries [UK, France, Italy] and
three countries undergoing economic transition [Hungary, Romania, and Turkey]
to show that while ICT access may be closely correlated with a variety of
economic characteristics, it is poorly correlated with SME productivity. Effective use of ICT seems to be based on
specific qualities of information systems utilization, management and
capability, and especially business attitudes towards the value of information
and of networks.
The paper is organized to
introduce the key elements of the argument and explain the significance to
industrial policy of our study of business clusters. In the next section we introduce the problem. In the following we briefly compare the six
countries in terms of their economic and business characteristics as well as
their ICT infrastructures. This
highlights uneven development in Europe and indicates where most potential
gains might quickly be made. The final
section addresses the argument that virtual clusters and business networks undermine
local economies. We show why this is a
distraction from the importance of good practices for ICT use and not an
appropriate way of looking at the problem because certain kinds of local
activities enhance the ability of business network members to operate outside
their regions. The paper concludes with
a section that brings together management and business practice issues with
industrial and regional policy requirements.
Introducing the problem
Ten years ago it seemed as
if new ICT and globalisation were conspiring to undermine the very concept of
local economy. The evidence for that,
presumably, was seen in the ways in which SMEs in particular were newly enabled
to participate in networks of trade, labour markets and even customer
support. In their study of seven
localities around England, Curran and Blackburn (1994) describe the weakening
of local economic ties and describe what they call the “death of the local
economy”. Theorising along these lines proliferated
throughout the 1990s, based to a significant degree on 1980s studies and data
from business clusters in Silicon Valley, the Route 128 area near Boston,
Northeastern Italy, Sakaki in Japan, and latterly studies of British
localities. Some of these studies
influenced policies to promote science parks, free trade zones and industrial
districts throughout the world. In Britain and elsewhere in the European Union
local development agencies, armed with fiscal incentives, training budgets,
advising services, and significant political clout emerged, often creating
major impact upon social and economic structures.
But the whole movement is
based on what now might seem a paradox.
If the embeddedness of local economies is increasingly irrelevant
because of the effects of globalisation, then why should public policy, or
indeed strategic firm behaviour, act as though local context and action were
paramount? Indeed, the promotion of
broadband services, efficient uses of internet for business purposes, and other
ICT applications have been based on the notion that abilities to participate
globally should in some way be harnessed to enhance local economies (Acs,
1992). Some of this might be explained as a basic confusion, originating from
desire to emulate Silicon Valley while conscious of the inability to replicate
the confluence of all the distinctive features that account for its success
(including luck) (Jorgenson, 2001).
Some of it comes from the reasonable hope by local, regional and
national governments, and indeed of the EU (Keeble & Weaver, 1986), that
SME-led growth will continue to solve local problems of employment, quality of
life, and generally lift participation in e-society through, for example,
promoting the dissemination of mobile communications and broadband (Acs, 1992;
Keeble & Wilkinson, 1999a; 199b).
Our premise is that
localities remain important, and that good policies can affect their success
(Ozcan, 1995). We also believe that SMEs in Europe can make better use of
ICT. We should be able to see these
effects in the increasing flexibility, reduced risk aversion, information use,
and other features that better enable firms to compete. This paper poses the
question, what is the basis of the belief that even the most prosperous regions
of Europe continue to lag behind similar regions in the United States in
reaping the benefits of ICT-led productivity growth, despite similar access to
ICT, including broadband. Furthermore,
ten years of concerted efforts by the European Commission’s Information Society
Programme to promote best practices and with great sympathy for SMEs and
business clusters may not have reduced the productivity gap significantly.
Our work is influenced by
“New Economic Geographers”, “New Industrial Geographers”, science policy
analysts and macroeconomists assessing the productivity effects of ICT (Breschi
& Lissoni, 2001; Bresnahan & Trajtenberg, 1995; Utterback & Suarez,
1993). There have been very few recent British studies of ICT and business
clusters, of the type seen in, for example, the Netherlands (Stroeken, 2001)
and Italy (Bagella & Becchetti, 2002).
Furthermore, there have been no studies that have focussed on the key
issue: what are the ways in which ICT-led productivity growth differ
internationally among SMEs, and which industrial policies and business
strategies might be most effective in delivering similar benefits across
regions. Based on data from the OECD
and other sources, we can now describe many of the differences between
countries and sectors as regards use of ICT.
We can also use these differences to comment on a variety of features,
including the relationship between ICT use by SMEs and the overall productivity
of the economy, the character of managerial use of ICT and by extension the
likely characteristics of markets for new ICT among small businesses,
especially broadband.
SMEs, ICT and economic context
How can we compare the six
countries in terms of their economic and business characteristics as well as
their ICT infrastructures? Table 1
below, showing the distribution of enterprises as well as Internet usage and
GDP per capita in each of our comparison country and the US, highlights uneven
development in Europe and indicates where most potential gains might quickly be
made.
The role of SMEs in the
comparison countries varies although for all over 92% of companies employ under
50 persons and almost all firms employ fewer than 100, except in the UK where
4.2% employ over 99. This contrasts
with the situation in the United States where small firms make up only just
over 70% of all companies and 6.7% employ over 99.
Table
1: National Context: Distribution of enterprises in the economy
|
Country |
0-9 |
10-49 |
50-99 |
100-499 |
500+ |
Internet users in
thousands (per 10,000 in parentheses)[3] |
GDP/C in US$ (IMF figures
for 2003) [4] |
|
United Kingdom |
72.0 |
20.5 |
3.3 |
3.5 |
0.7 |
24,000 (4,062) |
29,642 |
|
France |
82.4 |
13.5 |
2.0 |
1.1 |
0.1 |
18,716 (3,138) |
28,279 |
|
Italy |
83.7 |
14.3 |
1.1 |
0.8 |
0.1 |
17,000 (3,011) |
24,998 |
|
Hungary |
96.3 |
3 |
0.6[5] |
0.1 |
0.1 |
1,600 (1,576) |
8,378 |
|
Romania |
92.8 |
5.8 |
1.4 |
0 |
0 |
1,800 (806) |
2,342 |
|
Turkey |
95.0 |
3.2 |
0.8 |
0.9 |
0.2 |
4,900 (728) |
3,533 |
|
United States |
56.8 |
15.8 |
20.7 |
5.2 |
1.5 |
155,000 (5,375) |
37,312 |
(Source: data
collated from oecd.org; itu.int; imf.org)
South-Eastern
England
The South-Eastern region
of England is not only the most prosperous; it also has attracted the largest
concentration of highly innovative companies in the country. By the same token
it has been the most attractive region to conduct experiments and tests of
plans for improvements associated with introducing state-of-the-art managerial
practices and improvements. Nevertheless,
productivity has not been able to accelerate at the rate necessary to close the
gap with the most productive comparable regions in the United States. While some of this can be explained
structurally, in relation to financial fundamentals, regulation and labour
market characteristics, some of it can be attributed to the differences in
usage of ICT.
South-Western France
The region around Toulouse
is one of the fastest growing areas of France, fuelled by high technology
investments. The stimulus for this has
most conspicuously included the extensive Airbus operations in the region and
the wide use of nearby specialist suppliers.
However, it has also been affected by the presence of some excellent
departments in Toulouse University and the increasing popularity of the style
of life afforded in the area, which has made it attractive for settlement to
many highly skilled professionals. While some aspects of industrial development
in the region are underdeveloped, those associated with the aeronautics and
avionics businesses are highly productive and include leading users of state of
the art ICT. What makes the region
attractive for study is the manner in which networks have been formed around
the identity of the region as a high technology cluster distinct from the Paris
metropolitan area. The major products
of the SMEs are also associated with export industries and so the
competitiveness of clusters is marked by true international comparisons. One
other factor of special interest for the Toulouse region is the way in which EU
cluster policies have had strong influence.
French regional policies have been closely in tune with EU
recommendations (indeed they have influenced the character of such policies)
and we expect that the behaviour of clusters is consonant with EU expectations.
North-Eastern Italy
The regions of Emilia
Romana and surrounding districts (including the towns of Bologna, Modena, and
the Milan area) were the first ones to attract the attention of those studying
new regional policies, SME clusters and the use of advanced management
techniques and ICT. The productivity of
the region remains very high by European standard, and recent experience has
shown that in terms of capital investment and innovative management techniques,
there is still much good practice and a high level of competence. The region
has enjoyed a sustained period of investment in high quality infrastructure of
all kinds and boasts the best educational institutions and other organizations
that contribute to technology transfer.
There have been numerous experiments with electronic means of
coordination, from dedicated local business networks to advanced e-government
applications locally, to experiments with virtual networks.
Romania:
Bucharest and Ilfov
The Bucharest and Ilfov
region is taken because it accounts for just about 20% of SMEs in the economy,
and half of all companies with foreign investment. Despite the cluster of businesses established in this region,
much of Romania is undeveloped both in economic terms and especially in terms
of the application of EU regional policies. SME specific policies are only
beginning to be applied following on from EU directed initiatives and the
requirements to complete preparations for EU accession.
Hungary: Budapest and North-Western Hungary
Hungary, widely seen as
the most competitive Central European economy, has undergone rapid economic and
political transition since 1989/90. During the very first year of transition,
Hungary lost nearly 70% of its export markets in Eastern and Central Europe, a
phenomenon that affected Hungarian SMEs most severely. Much of the country's
regulatory and economic framework has been thus guided by the principle of EU
membership. The Hungarian economy is characterised by acute regional differences
with a per capita GDP of 83.5% of the EU 25 average for Budapest and Central
Hungary while regions outside the capital lagging well behind (Financial Times,
2003). There are concentrations of SMEs in and around Budapest and the central
Hungarian cities of Szekesfehervar (of which a large proportion act as
suppliers to multinational manufacturers) and Gyor. The eastern city of
Szombathely also hosts many SMEs again associated with multinational
manufacturers.
Central
Aegean Region of Turkey
The recent export-oriented
growth fuelled many Turkish SMEs, which traditionally relied on local markets
and narrow product specifications.
Along with trade liberalisation and economic reforms, Turkey’s
integration into world markets and close EU ties helped to create new growth
poles and SME clusters over the last two decades (Ozcan, 1995). While old traditional craft based firms have
been gradually disappearing, new SME clusters have been quick to transform
local economies across the country. The
Central Aegean region has such fast growing regional clusters of SMEs spreading
from an old port, Izmir, and stretching to newly industrialising towns of
Manisa, Denizli, and Aydin. Most of the SMEs in this region concentrate in
traditional manufacturing sectors such as textiles, food processing and
agro-industries. High-tech firms are
promoted through governmental agencies and educational investments in the
region. The region has great potential
for economic growth due to its high business start ups, rich agricultural base
and tourism. But, this potential is not
fully realised due to regional as well as macro economic and political
constraints.
Comparing the rich EU
countries with major SME clusters in the three transition states shows that low
productivity poses one of the major problems for innovative SMEs. At the same
time, comparisons of EU and accession states with US data shows poor European
productivity levels in general. Lack of managerial and labour skills, poor
technology use/adaptation, and pervasive illegal economy in the case of some of
the accession states may be major reasons behind this productivity gap. SMEs
rely on strong family and social ties with weak business networks for their
survival. Local proximity and social network ties appear to be crucial for the
survival and growth of businesses with low technology base, particularly in
Turkey, Romania and Hungary. These latter countries have been trying to
replicate EU policies and best practices.
Incubator houses, industrial parks, enterprise zones and other
instruments of regional and rural development have been applied in these
countries, but with little effect upon the rate of productivity growth. In Romania, Hungary and Turkey, most SMEs
are young start-ups and are very susceptible to economic instability. Regulatory systems are not SME friendly in
these countries because of their emphasis on attracting FDI. The diffusion of
good use of ICT is often overlooked in the policy framework as well. There is
need for a total institutional shake-up and long term policy development in
order to close Europe's productivity gap with the US.
ICT, SMEs and Business clusters
According to the OECD
(1998), there is a positive correlation between adoption of ICT and firm
size. However, there are substantial
variations even within SMEs. For
instance, OECD research based on 2881 companies in 12 European countries
shows that the Internet is used by 41 per cent of companies with 50‑99
employees, by 30 per cent of companies with 10‑49 employees, and by
only 16 per cent of companies with 1‑9 employees. An EU-Databank
survey of ICT use by European SMEs found that SMEs use the Internet for e-mail
(64 per cent), to maintain contact with a wide range of organisations (60 per
cent), to access a wider range of information sources (56 per cent), to improve
co-ordination of their main line of business within and outside the
organisation (29 per cent), to experiment (26 per cent), to conduct on-line
sales/orders (14 per cent) and on-line purchase of goods and services (12 per
cent). The same study indicated that
SMEs were motivated to create web sites to raise visibility and advertise (67.2
per cent) and to widen the range of target customers (40 per cent), but that
on-line sales and orders were less important (11 per cent). However, broad
conclusions about SMEs are at best indicative and at worst of limited value for
policy makers. For electronic commerce
(as indeed for most applications and the diffusion of ICT), the heterogeneity
of SMEs (in terms of market structure, aversion to risk, location, sector,
organisational structure, innovation climate, etc.) affects the scale and rate
of implementation of electronic commerce and its very appropriateness. This has implications for the rate of adoption
as well as efficacy of policies and programmes to encourage SMEs to adopt
electronic commerce.
The translation of
awareness of electronic commerce into adoption, investment and use also
varies. For example, an SME can operate
in a highly competitive market with low innovation rates (e.g. low end of the
retail sector) or in a highly dynamic industry with high innovation rates (e.g.
software). The former may be
indifferent to electronic commerce or perceive it only as a way to reduce
costs; the latter may see it as a tool for new process, product or service
innovations or for creating new markets (La Rovere, 1996). Other SMEs may
operate in clusters for production or within networks dominated by large
companies (such as SME suppliers to single large manufacturers in the car
industry). In fact, one would expect
that large companies would drive the adoption of business-to-business
electronic commerce. In these cases,
there is likely to be a highly innovative environment and electronic commerce
would be used for productive processes and supply line relations (OECD, 1998).
SMEs in all sectors in the
prosperous regions of the UK, France and Italy use ICT, even if only in a
rudimentary fashion (Baptista, 2000).
Common uses these days were regarded as exotic only a few years ago, but
customer databases, websites, routine business uses of mobile telephones and
e-mail are now the norm. Some of these,
especially mobile telephones and e-mail, are equally a part of domestic life
and so have come to characterise e-society generally. Indeed, it can no longer
be regarded as a distinguishing characteristic or a contribution to business
strategy merely to utilise ICT. As was
long predicted, ICT has now become much more like a commodity than a
specialised product (Carr, 2003).
However, good practices applied to business are still rare (Brown &
Hagel, 2003; Knol & Stroeken, 2001).
Indeed, a very high variability in productivity can be seen among
companies (Stoneman & Toivanen, 1997), and indeed clusters of companies
(Curran & Blackburn, 1994), which have similar access to hardware and
communications channels, including broadband.
The other differentiating
feature is that some sector norms promote participation in e-society. Exactly what these norms are is not clear
(Tracy & Clark, 2003), but we suppose that they include high standards of
website design and use, effective data compatibility, and good decision making
about what to control internally and what to outsource (Sadowski, Maitland
& van Dongen, 2002; Santarelli & D'Altri, 2003). We wish to see if these are characteristics
of sectors, and if so whether these are generally common of those sectors
internationally or more locally of how those businesses within the region
respond (Audretsch, 1998; Bagella & Becchetti, 2002).
Clusters that evidence
effective use of ICT are able to foster participation in e-society (Caldeira
& Ward, 2002) by disseminating skills, standards, and good practices of
information use (Elfring & Hulsink, 2003).
We propose that this, as well as the general efficacy in promoting
effective use of ICT, differ among different kinds of business groupings such
that proximity clusters differ in their ability to promote good use from
virtual clusters (Beaudry, 2001; Beaudry & Breschi, 2003).
ICT use is a function of
management practices, not possession of hardware, software and communications
technologies or services (La Rovere, 1996). Clusters that have a critical mass
of good practices for the use of ICT (DTI (UK), 1998) are more likely to be
productive (Brynjolfsson & Hitt, 1996; 2000) than those that do not, but we
still see that local productivity advantages from ICT vary widely among
clusters in the UK (DTI (UK), 2001; Naylor & William, 1994; Robertson, Swan
& Newell, 1996). It is more likely that norm-building activities within
clusters are more successful than service agencies in boosting participation in
e-society and benchmarking is an important such norm-building practice (Cragg,
2002).
Conclusions
We believe that current EU
and national policies with respect to ICT are inadequately attuned to business
practices that foster effective ICT use by SMEs. Some studies have argued that
business networks and virtual clusters have the capacity to undermine local
economies (for instance, Curran & Blackburn, 1994). However, we believe
that certain kinds of local activities enhance the ability of business network
members to operate outside their regions and that the local economy continues
to matter. This might also imply that common European policies for regional
development and SME clusters may be misguided not because spatial
characteristics don’t matter but because management practices, especially those
related to information use, will always matter more. Virtual clusters and
sector-specific virtual networks are good in principle because they may easily
overcome out-of-region trade barriers (and especially international export
opportunities) both by making firms more aware of business opportunities and by
leading them into more effective competitive behaviours. However, the provision
of infrastructure such as broadband and the dissemination of leading business
software applications are insufficient to ameliorate the productivity gap where
information use behaviour is tied to traditional practices.
In its recommendations for
SME-related policies, the OECD (1998) clearly recognised that government
actions to facilitate the use of electronic commerce by SMEs continued to
remain fragmented and tentative. OECD also suggested the need to conduct
extensive cross-country studies to analyse the opportunities, benefits and
barriers for the take-up of electronic commerce by SMEs as well as best
practice in policy responses. Our call for a new research agenda presented in
this paper aims to contribute to this effort but by asking different sets of
questions in relation to much of the current research agenda in this area. We
are interested in understanding the impacts and influences of virtual versus
proximity clusters of SMEs on ICT use and practices within the six EU/accession
states discussed above. We believe that this will have a direct bearing on
EU-wide policy initiatives for ICT use within SMEs and for benchmarking
ICT-related practices within European SMEs in general.
References
Acs, Z.J. 1992. Small business economics: A global perspective. Challenge, Nov-Dec.
Audretsch, D.B. 1998. Agglomeration and the location of innovative activity. Oxford Review of Economic Policy, 14(2): 18-29
Bagella, M. & Becchetti, L. 2002. The ‘geographical agglomeration-private R&D expenditure’ effect: empirical evidence on Italian data. Economics of Innovation and New Technology, 11(3): 233-247
Baptista, R. 2000. Do innovations diffuse faster within geographical clusters? International Journal of Industrial organisation, 18: 515-535
Beaudry, C. 2001. Entry, growth and patenting in industrial clusters: a study of the aerospace industry in the UK. International Journal of the Economics of Business, 8(3): 405-436
Beaudry, C. & Breschi, S. 2003. Are firms in clusters really more innovative? Economics of Innovation & New Technology, 12(4): 325-343
Breschi, S. & Lissoni, F. 2001. Localised knowledge spillovers vs innovative milieux: Knowledge "tacitness" reconsidered. Papers in Regional Science, 80: 255-273.
Bresnahan, T.F. & Trajtenberg, M. 1995. General purpose technologies - 'engines of growth'? Journal of Econometrics, 65: 83-108
Brown, J.S. & Hagel, J. III. 2003. Does IT matter? Harvard Business Review, June: 109-112
Brynjolfsson, E & Hitt, L.M. 1996. Paradox lost? Firm-level evidence on the returns to information systems spending. Management Science, 42(4) 541-558
Brynjolfsson, E. & Hitt, L.M. 2000. Beyond computation: information technology, organizational transformation and business performance. Journal of Economic Perspectives, 14 (4): 24-48.
Calderia, M.M. & Ward, J.M. 2002. Understanding the successful adoption and us of IS/IT in SMEs: an explanation from Portuguese manufacturing industries. Information Systems Journal, 12: 121-152.
Carr, N.G. 2003. IT doesn't matter. Harvard Business Review, May: 41-49.
Cragg, P.B. 2002. Benchmarking information technology practices in small firms. European Journal of Information Systems, 11:267-282
Curran, J. & Blackburn, R. 1994. Small Firms and Local Economics Networks. Paul Chapman.
Elfring, T. & Hulsink, W. 2003. Networks in entrepreneurship: the case of high-technology firms. Small Business Economics, 21:409-422
Financial Times. 2003. Financial Times Special Report - Hungary. Financial Times, 27 May: I-VI.
Jorgenson, D. W. 2001. Information technology and the US economy. American Economic Review, 91:1-32
Keeble, D. & Weaver, E. (Eds.). 1986. New firms and regional development in Europe. Beckenham: Croom Helm.
Keeble, D. & Wilkinson, F. 1999a. Collective learning and knowledge development in the evolution of regional clusters of high technology SMEs in Europe. Regional Studies, 33(4):295-303
Keeble, D. & Wilkinson, F. 1999b. Collective learning processes, networking and ‘institutional thickness’ in the Cambridge Region. Regional Studies, 33(4):319-332
Knol, W.H.C. & Stroeken, J.H.M. 2001. The diffusion and adoption of information technology in small- and medium-sized enterprises through IT scenarios. Technology Analysis & Strategic Management, 13(2): 227-246
La Rovere, R.L. 1996. IT diffusion in small and medium-sized enterprises: elements for policy definition. Information Technology for Development, 7(4): 169-182
Naylor, J. & William, J. 1994. The successful use of IT in SMEs on Merseyside. European Journal of Information Systems, 3(1), 48-56.
OECD. 1998. SMEs and electronic commerce. Paris: OECD.
Ozcan, G. 1995. Small firms and local economic development. Avebury: Aldershot.
Robertson, M., Swan, J. & Newell, S. 1996. The role of networks in the diffusion of technological innovation. Journal of Management Studies, 33(3), 333-359.
Sadowski, B.M., Maitland, C. & van Dongen, J. 2002. Strategic use of the Internet by small-and medium-sized companies: an exploratory study. Information Economy and Policy 14: 75-93
Santarelli, E. & D'Altri, S. 2003. The diffusion of e-commerce among SMEs: Theoretical implications and empirical evidence. Small Business Economics, 21: 273-283.
Stoneman, P. & Toivanen O. 1997. The diffusion of multiple technologies: An empirical study. Economics of Innovation and New Technology, 5: 1-18
Stroeken, Jan H.M. 2001. The adoption of IT by SMEs: the Dutch case. Journal of Enterprising Culture, 9(1): 129-152
Tracy, P. & Clark, G.L. 2003. Alliances, networks and competitive strategy: rethinking clusters of innovation. Growth & Change, 34(1).
Utterback, J. & Suarez, E. 1993. Innovation, competition and industry structure. Research Policy, 22, 1-21.