Floud R., Johnson P. The Cambridge Economic History of Modern Britain, Volume 1: Industrialisation, 1700-1860

Подождите немного. Документ загружается.

Education and skill of the British labour force 341

theabsence of any centrally administered or ﬁnanced schooling network,

one can surmise that by 1700 a substantial minority of English children

would have spent at least some time in school. The 1851 Census of Edu-

cation indicates that, by that date, 60 per cent of boys and 56 per cent of

girls aged 5 to 9 were scholars. Relatively sluggish growth of literacy and

schools in the century and a half prior to this would suggest that the

level of school attendance in the early eighteenth century would have

been between perhaps 20 and 40 per cent of this age group. Therefore,

attending school in early eighteenth-century England was not an uncom-

mon experience, even for many working-class children (Jones 1938; Hans

1951;Simon 1968; Lawson and Silver 1973; O’Day 1982; Mitch 1993). For

this period it is estimated that between 40 and 50 per cent of adults

could sign their name (Cressy 1980). Although basic reading and writing

skills were often acquired outside of formal schooling during this period,

these ﬁgures would suggest widespread familiarity with what was done in

school. Schools were quite heterogeneous in their ﬁnancing and in their

educational aims (Lawson and Silver 1973; O’Day 1982). The majority of

students were probably in private-venture schools, i.e. single-teacher oper-

ations (commonly neighbourhood women or dames or elderly men lack-

ing alternative employment) run for proﬁt. The curriculum would have

focused on basic literacy and whatever other odd subjects the teacher

wanted to throw in. Schools with religious afﬁliation and support were

widespread as well, though religious motives behind the propagation of

education did not attain the same level of intensity as in other Protes-

tant regions such as Sweden and Germany (Houston 1988b: 40–8). Philan-

thropists also supported schools without any formal religious afﬁliation

(Lawson and Silver 1973: 181–2).

Achild reaching adolescence around 1700 would have been relatively

unlikely to attend school, given the rise in his or her opportunity costs

(foregone earnings) with maturation. Traditional grammar schools, to-

gether with academies run by and for religious dissenters, did pro-

vide some secondary instruction (O’Day 1982: ch. 11). The Universities

of Oxford and Cambridge offered higher education, and institutions

for morespecialised professional training such as the Inns of Court

were present in London. Such post-primary education would have been

pursued by well under 5 per cent of the relevant age group in early

eighteenth-century England. Many children would have made their sub-

sequent way in life through the school of hard knocks: ﬁnding employ-

ment as adolescents in order to take advantage of their maturing earning

power or, in the case of females, helping out at home (Ben-Amos 1994).

Because of its shorter length and the younger age when it would have

occurred, with consequent lower opportunity cost, the typical personal

investment in formal schooling would have cost considerably less than

thetypical apprenticeship. For example, three years of schooling between

theagesof8and 10, perhaps roughly sufﬁcient to acquire basic literacy,

Cambridge Histories Online © Cambridge University Press, 2008

342 David Mitch

can be costed as having direct fees of 6 pence per week and opportu-

nity costs per week usually exceeding no more than 2 shillings. At forty

weeks of attendance per year, this implies a total cost per person of £15.

Assume that 40 per cent of an estimated 100,000 English 10-year-olds in

1700 had been in school for three years and that the average working

life for those reaching the age of 10 was a further thirty-ﬁve years.

This

yields a further cumulative human capital stock in literacy and related

primary schooling for males and females of £21 million, about half of

theestimated cumulative investment in male apprenticeship of £52 mil-

lion. This combined human capital stock in both literacy and primary

schooling for males and females and in apprenticeship for males in 1700

of £73 million is still well under half of Allen’s (1994) estimates of the

capital stock in agriculture in 1700 of £183 million.

The beneﬁts from apprenticeship, in terms of a stream of enhanced

earnings once qualiﬁed, were potentially large. This is suggested by

Gregory King’s assignment of an annual income per family of £40 for

those in artisan and handicraft occupations compared with £15 for

labouring people and outservants (Mathias 1979: 186–7). However, the

percentage rate of return for schooling may have been higher, relative to

its markedly lower opportunity and total costs.

Apprenticeship and formal schooling were by no means incompatible.

Indeed in a range of contexts, pursuit of apprenticeship was based on

an earlier foundation of formal schooling. This would have occurred in

professions such as law and medicine, in which some instruction in basic

languages and other disciplines would generally be expected before en-

tering the apprenticeship upon which professional training in the early

eighteenth century was generally based. One important activity, for which

formal schooling was the foundation for apprenticeship and related on-

the-job experience, was that of the merchant. Pollard (1965) has suggested

afﬁnities between the education and training of the merchant and that of

subsequent industrialists in the early industrial revolution. The merchant

commonly mastered both general skills such as basic literacy and more

speciﬁc skills suitable for commerce such as accounting and mastery of

certain modern languages used in trade. Malachy Postlethwayt (1774) pro-

posed in the mid-eighteenth century that a formal commercial academy

be set up for this purpose. Although nothing came of this proposal,

dissenting academies set up a similar curriculum (O’Day 1982: 208–15,

270).

Aspiring merchants commonly started as agents in distant loca-

tions as preparation for acquiring more senior status in merchant houses

The thirty-ﬁve-year expected service life ﬁgure for a 10-year-old comes from the admittedly

quite crude adjustment of simply adding ten years to the previous rough estimate of a

twenty-ﬁve-year service life for a 21-year-old.

Alecture given by Leone Levi (1868) in 1868 indicates that at that time, King’s College,

London was offering classes in physical geography, commerce and commercial law aimed

at the aspiring merchant. However, Levi’s remarks also suggest that few students were

actually taking these classes.

Cambridge Histories Online © Cambridge University Press, 2008

Education and skill of the British labour force 343

(Ben-Amos 1994: 124–7). Given the creative nature of entrepreneurial ac-

tivity, it is not clear that training for entrepreneurship can be provided in

asystematic way. It was probably at least as critical in guaranteeing a sup-

ply of entrepreneurs that relatively few barriers and disincentives were

created for those wishing to pursue their entrepreneurial proclivities.

TRENDS IN HUMAN CAPITAL

ACCUMULATION, 1700–1860

The century and a half that is surveyed in this volume is commonly seen

as a period of profound change and transformation in the English econ-

omy. One might expect corresponding evidence of fundamental change in

theeducation of the English labour force. Yet educational histories of this

period tend to suggest a pattern more of stagnation than of improvement.

However, evidence on schooling and other forms of investment in human

capital during this period is fragmentary. This sets the tasks of examin-

ing the evidence for educational stagnation and how the relationship

between educational trends and economic change can be understood.

Literacy rates as measured by signature rates at marriage provide the

most easily quantiﬁable indicator of trends in education between 1700

and 1860. Signature rates are not without difﬁculty as measures of read-

ing and writing ability, but nevertheless long-term trends in signature

ratesare likely to reﬂect trends in the latter (Schoﬁeld 1973; Houston

1988b; Reay 1991). Cressy (1980: 176), having examined a range of sources

such as legal documents which indicate signature rates, has suggested

that, at the accession of George I in 1714, approximately 45 per cent of

men and 25 per cent of women could sign their names (see Figure 12.1).

Cressy’s assessment is that this reﬂects continuous improvement since

1500, at which date perhaps at best 10 per cent of men, and an even

smaller per centage of women, could sign their names. By this point, as

Table 12.2 shows, other countries already had an educational lead over

England. North America and Sweden stand out in this regard (Graff 1987:

ch. 6). It is also commonly believed that Scotland had higher literacy rates

than England, though recent research has qualiﬁed this view (Houston

1985). England seems to have been in a similar position to France, and

wasperhaps behind Germany and Holland, but almost surely ahead of

areas in southern and eastern Europe such as Spain, Italy and Russia

(Graff 1987: ch. 6). As already suggested, while literacy was not acquired

exclusively in schools, these literacy ﬁgures probably in some degree re-

ﬂected the widespread network of primary schools which were present

throughout England at this time.

Asimilar intermediate position in higher levels of educational at-

tainment characterises the international standing of England’s universi-

ties and scientiﬁc establishment in 1700 (Stone 1974; O’Day 1982: 273–5;

Cambridge Histories Online © Cambridge University Press, 2008

Edited by Foxit Reader

For Evaluation Only.

344 David Mitch

Figure 12.1

Estimated illiteracy of

men and women in

England, 1500–1900

Source: Cressy 1980:

177.

Houston 1988b: ch. 4; Inkster 1991). Scotland and the Netherlands had far

more vibrant activity than England in such ﬁelds as law and medicine.

But despite recurrent disdainful comments about Oxford and Cambridge,

these universities did have some scholars making respectable if not

exceptionally brilliant intellectual contributions.

Schoﬁeld (1973) has made national-level estimates of trends in signa-

ture rates for brides and grooms at marriage for England between 1750

and 1840 (see Figure 12.2). Some improvement in ability of brides to sign

their name isevident, with the proportion rising from about a third in

the1750s to about half in the 1830s. However, for grooms, any rise was

more modest, increasing from perhaps 60 to 70 per cent over this pe-

riod. Moreover, regional studies focusing on Lancashire, a focal point of

thegrowth of cotton textile manufacture, have found deteriorating sig-

nature rates during industrialisation (Sanderson 1972a). This assessment

has been subject to some debate. Laqueur (1974) has argued that literacy

Table 12.2 Literacy for selected areas in Europe and North

America c. 1700

Region Male literacy rate Female literacy rate

England 40% 25%

France 29% 14%

Amsterdam 70% 44%

Moklinta, Sweden 89% 89%

(reading ability)

Iceland Almost 50% Almost 50%

New England 70%

Note: Literacy rates for England, France and Amsterdam based on signature

rates at marriage.

Sources: Cressy 1980; Graff 1987.

rateswere actually rising modestly in

industrialising Lancashire. And Nicholas

and Nicholas (1992) have constructed na-

tional literacy estimates by occupational

grouping, indicating that rates were ac-

tually rising during a ﬁrst phase of

industrialisation through the late eigh-

teenth century, followed by a decline

which they interpret as reﬂecting a

deskilling phase of industrialisation.

More generally, considerable variation in

local and regional literacy trends has

been noted (Schoﬁeld 1973; Stephens

1987). Nevertheless, the preponderance

Cambridge Histories Online © Cambridge University Press, 2008

Education and skill of the British labour force 345

Figure 12.2

Estimated annual

percentage of males

and females unable to

sign at marriage,

1754–1840 (from a

random sample of 274

English parish

registers)

Source: Schoﬁeld

1973: 445.

of evidence suggests little overall upward trend in literacy rates before

the second quarter of the nineteenth century.

Evidence on corresponding schooling trends is much more patchy.

In considering likely developments, it is useful to distinguish between

theprivate-venture, for-proﬁt schooling market and schools that were ﬁ-

nanced at least in part by subsidies. There is no reason to think there were

any major changes between 1700 and 1850 in the private-venture school-

ing market. It is likely to have expanded and contracted with working-

class demand. There would seem to have been a ready and elastic supply

of those willing to sell instructional services at prevailing market rates

(Gardner 1984). There is little reason to think there was an expansion

in working-class demand during the industrial revolution period, in so

farasneither living standards nor labour market demands for schooled

workers were increasing.

Cambridge Histories Online © Cambridge University Press, 2008

346 David Mitch

Table 12.3 Proportion of population under 20 in England and

Wales who were scholars, 1818–71

1818 1833 1851 1861 1871

0.128 0.179 0.267 0.346 0.355

Source: Mitch 1982: 10.

Turning to subsidised schools, there

does seem to have been an extensive

charity school network in the early eigh-

teenth century. However, there is lit-

tle evidence to suggest clear revival or

growth of this network (Simon 1968).

Beginning in the early nineteenth century, various religiously afﬁliated

societies do seem to have been active in promoting the growth of school-

ing provision. The national surveys sponsored by parliament in 1818 and

1833 are themselves evidence of awakened national interest in schooling

activity. And in 1833 parliament began to fund the building of schools

and subsequently developed apparatus for supporting running costs as

well as for the development of curriculum standards and the provision

of teacher training and certiﬁcation. The impact of this was revealed in

rising enrolment rates in day schools between 1818 and 1851. Yet private-

venture enrolments appear to have kept up with subsidised enrolments

throughout this period. The proportion of all scholars who were enrolled

in private venture schools was more or less the same, at 60 per cent, in

both the 1818 and the 1833 national educational surveys.

Sunday schools experienced the clearest surge of activity in the late

eighteenth and early nineteenth centuries. They were virtually non-

existent before the late eighteenth century. By the second quarter of the

nineteenth century, Sunday school enrolments exceeded those in day

schools. While Sunday schools commonly had religious afﬁliations and

sponsorships, they also appear to have offered some secular instruction

in basic literacy skills. Historians have differed over the extent to which

their expansion was driven by working-class demand for secular educa-

tion or by the supply of sponsoring religious organisations (Laqueur 1976;

Dick 1980). Sunday schools were commonly used as an alternative to day

school instruction. Sunday school enrolment rates were positively corre-

lated with child labour force participation and negatively correlated with

literacy rates (Laqueur 1976; Mitch 1992; Snell 1999).

The period 1700 through 1860 does not appear to have been one of

general expansion or improvement for secondary and university educa-

tion. Secondary education during the eighteenth century was subject

to conﬂicting forces. Inﬂation had eroded the value of many grammar

school endowments but students seeking commercial and related types

of practical education contributed to the growth of dissenting academies

(Parker 1969; O’Day 1982). Enrolments at Oxford and Cambridge reached a

nadir in the mid-eighteenth century and university enrolments relative to

population continued to decline in England through the mid-nineteenth

century (Sanderson 1972b: 3; Stone 1974). In any case, no more than a

small percentage of the population would have enrolled in secondary

or higher education during this period. There does appear to have been

an upsurge in more informal intellectual institutions such as literary,

Cambridge Histories Online © Cambridge University Press, 2008

Education and skill of the British labour force 347

philosophical and scientiﬁc societies and in free-lance lecture series with

quite diverse aims and clientele (Musson and Robinson 1969; O’Day 1982:

210; Inkster 1991) but the magnitude and signiﬁcance of this activity are

difﬁcult to gauge.

Turning to trends in other forms of human capital than schooling,

formal apprenticeship is commonly seen as going into decline during

theeighteenth century, culminating in repeal of the Statute of Artiﬁcers

in 1814. Dunlop’s (1912) assessment is that compulsory apprenticeship

under the terms of the Statute of Artiﬁcers was in decline as early as the

late seventeenth century. According to her, over the course of the eigh-

teenth century, capitalist master employers who increasingly demanded

less skilled labour than that implied by a seven-year apprenticeship term

were able to evade the conditions of the Statute. In trades in which guilds

were able to maintain their control over entry or in which an extensive

apprenticeship period was genuinely required for mastery of the skills

involved, apprenticeship, albeit voluntary, persisted into the nineteenth

century. Snell (1985), using settlement documents for artisans from south-

ern England, found that the length of apprenticeship terms declined

appreciably in the second half of the eighteenth century compared with

theﬁrst half. He argued that this occurred because apprentices perceived

shrinking opportunities for utilising the skills acquired and hence less

value in completing their apprenticeships. This decline was in evidence in

theapprenticeship documents that Snell examined for women as well.

However, Berg (1987: 75) has suggested that the purpose of female ap-

prenticeships was more often to provide support and socialisation while

awaiting marriage than mastery of speciﬁc crafts. Furthermore, with the

decline in the agricultural sector and shift of labour out of agriculture,

both the more structured institutions such as service and the more in-

formal exposure to agricultural work that served as training would have

been on the decline as well.

Some have argued that the eighteenth-century decline of apprentice-

ship was followed by further deskilling tendencies in the early nineteenth

century, with the rise of outwork or so-called sweated trades and the in-

creasing employment of women in unskilled manufacturing occupations,

especially textiles and metal manufactures (Berg 1987). According to this

view, the ready supply of low-wage labour, especially female labour, made

it proﬁtable to divide up manufacturing processes in clothing, textiles

and metal working into specialised steps, each requiring dexterity that

it was thought female labour could provide but which did not entail a

long period of skill acquisition.

On the other hand more informal sorts of skill acquisition arrange-

ments arose as alternatives to apprenticeship. Boot (1995) has argued

that the effective period of training in cotton textile factories could be at

least as long as traditional formal apprenticeship, and perhaps as long as

ten years,implyingcomparable magnitudes of human capital investment

Cambridge Histories Online © Cambridge University Press, 2008

348 David Mitch

and well in excess of what characterised typical patterns of formal school-

ing. It can be argued that such informal arrangements were better suited

than traditional apprenticeships to the development and acquisition of

the range of new empirically based skills that were important for the

newly emerging technologies fuelling industrialisation. Informal train-

ing could be adjusted in length to suit newly emerging and expanding

skills and occupations. In addition factory masters wanted to instil habits

of discipline and routine in their workforce that may have been incompat-

ible with guild and artisan traditions of independence, instead requiring

amore coercive approach (Pollard 1959).

Furthermore, Boot (1991) has documented the introduction of age-

related clerical salaries in the Bank of Scotland between 1780 and 1830

in contrast to previous apparently level age-earnings gradients for clerical

staff in the bank. Boot attributes this change to the effort of the bank to

create an internal labour market in response to changing conditions in

theclerical labour market. He suggests that this reﬂected an increasing

tendency over time for relevant skills in the bank to rise with age.

Nettrends in skill acquisition are difﬁcult to ascertain. Williamson

(1985) has argued that there was a rising premium for skill over the ﬁrst

half of the nineteenth century and that this indicates that the demand

for skill was rising relative to supply. However, others have questioned his

estimates, and available evidence would not seem to support a uniform

widening premium for skill over the industrial revolution period (see

Mitch 1999: 268–70 for more detail). The redeployment of perhaps a quar-

terofthe male labour force from agriculture into manufacturing (based

on Crafts 1994: 45) between 1700 and 1840 and the arguably greater poten-

tial of manufacturing than agriculture for investing in skill (as suggested

by themuch ﬂatter proﬁle of earnings by age for agricultural labourers

than for cotton textile workers reported by Burnette (2000)) would seem

to imply a net increase in skill acquisition over this period. Baxter’s 1867

allocations of the male manual labour force according to skill level indi-

cate that over 60 per cent were in lower or higher skilled occupations (see

Table 12.4). This is arguably higher than would have been the case in 1700

as crudely suggested by the 17 per cent of those in the overall categories

of general labour, trade, distribution and manufactures, as reported in

King’s 1688 estimates (Mathias 1979: 189).

Lindert and Williamson (1982, 1983b) have proposed adjustments to both King’s 1688 ﬁgures

and Baxter’s 1867 ﬁgures. Using Lindert and Williamson’s readjustment of Baxter’s 1867

ﬁgures yields 68 per cent of the manual classes in higher and lower skilled occupations

(their ﬁgure is for all types of labour, not just male labour). Using their readjustment of

King’s 1688 ﬁgures yields 52 per cent of the manual classes in higher and lower skilled oc-

cupations. Thus, while Lindert and Williamson’s revised estimates still suggest an increase

between 1688 and 1867 in the percentage of the manual classes who were skilled, this in-

crease is considerably less than implied by King’s ‘unadjusted’ estimates. Thus there would

seem to be a good deal of uncertainty as to the extent of any increase in the percentage

of the manual labour force who were skilled during the eighteenth and early nineteenth

centuries.

Education and skill of the British labour force 349

Table 12.4 Distribution of men in the manual labour classes

across wage categories following Baxter’s 1867 classiﬁcation

Premium relative to Percentage of all

Wage category in unskilled wage of adult males in

shillings per week 10 shillings per week manual labour class

Higher skilled labour and manufactures

35s. 25s. 1.07%

28–30s. 19s. 20.3%

Lower skilled labour and manufactures

25s. 15s. 14.8%

21–3s. 12s. 26.1%

Unskilled labour and agriculture

15–20s. 7.5s. 6.62%

14s. 4s. 29.2%

14.5s. 4.5s. 1.9%

Source: Based on Baxter 1868: appendix IV.

Trends in physical capital accumula-

tion appear to have been much more

rapid than those in human capital ac-

cumulation. Investment ﬂows in physical

capital as a share of national income

were substantially larger than invest-

ment ﬂows in schooling. West’s (1970)

estimates of direct educational expendi-

tures of £3 million in 1833 for England

is just under 10 per cent of Feinstein’s

(1978: 76) estimate of annual ﬁxed cap-

ital formation for Britain averaged over

theperiod 1831–40 (pro-rated to Eng-

land’s population share). Allowance for

opportunity costs could increase school-

ing investments to as high as a quarter of

ﬁxed capital formation. Once allowance

is made for workplace training investments as with Boot’s estimates, the

gap would appear to be considerably smaller. Nevertheless, there is no

clear evidence in favour of a marked increase in the rate of human capital

investment during the eighteenth and early nineteenth centuries.

This is suggested by considering the impact of a similar annihilation by

plague followed by Eskimo repopulation in 1700 as that considered above

for1860. One can use Gregory King’s ﬁgures for 1688 to estimate total na-

tional income at £44.7 million. Deducting income to nobility and rentiers,

to landed agriculture, and to trade and distribution other than seamen,

leaves an income of £17.3 million that can be attributed to labour income.

Alternative estimates per family for unskilled labour income would be

the£15 per family for labouring people and outservants and the £6.5

per family for cottagers and paupers. King estimates a total of 1,360,000

families. Using £15 per family yields an income to unskilled labour of

£20.4 million, which is greater than the total labour ﬁgure and obviously

problematic. Using the alternative lower ﬁgure yields an income to un-

skilled labour of £8.8 million. This leaves a residual for human capital of

£17.3 million − £8.8 million = £8.5 million. This in turn produces an es-

timate of the share of human capital in national income of 8.5/44.7 = 19

per cent for 1700 compared with the 15.41 per cent estimated above for

1860. However, employing the upper estimates above for 1860 human

capital shares in the range of 25 to 35 per cent implies substantially

more scope for expansion of the role of human capital during the 150

years inquestion.This underscores the degree of uncertainty in current

knowledge about the quantitative extent of skill formation during this

period.

How is one to explain the apparently limited role of human capi-

tal in the industrial revolution? One explanation, following Galor and

Moav (2000), is that Britain was still at a relatively early stage in the

350 David Mitch

accumulation of physical capital and that returns to physical capital were

substantially in excess of those to human capital. Evaluating this possibil-

ity requires deciding on the relevant margin for expansion and coming

up with economy-wide estimates for the marginal return to each type of

capital. On some margins it would appear that returns to human capital

were in excess of those on physical capital, thus challenging this expla-

nation. Estimates of both nominal and real interest rates for government

bonds and other types of capital investments tend to average in the range

of 3 to 7 per cent for the eighteenth and ﬁrst half of the nineteenth cen-

turies. Estimates of the rate of return to basic literacy for the 1840s are

well in excess of 20 per cent (Mitch 1984), and Elbaum (1989) for the

early twentieth century estimates a rate of return to apprenticeship of

just over 20 per cent. A second explanation is that technical change was

not skill using during this phase but was to become so in later periods.

This can be explained by the increasing role of science-based technical

change later in the nineteenth century and the corresponding increasing

importance of formal over informal education. A third explanation is that

market failures led to underinvestment in human capital, as suggested

by its high relative returns as just noted above.

UNDERINVESTMENT IN HUMAN CAPITAL?

It has been suggested that England was underinvesting in schooling by

themid-nineteenth century, not because of evidence of the high rate of

return to schooling but on indications of higher levels of literacy and

primary school enrolments in other countries such as the United States

and Prussia (see Table 12.5A).

It is difﬁcult to address claims of underinvestment directly, because it

is difﬁcult to determine what the optimal level of investment in educa-

tion is for any economy. Since it is now commonly presumed in modern

societies that both the ﬁnancing and the provision of primary school-

ing should be subject to government controls, the lag of England behind

other countries in establishing such provision has often been seen as

reﬂecting some deﬁciency (Green 1990).

Closer examination of England’s public provision of primary school-

ing in the eighteenth and early nineteenth centuries mitigates the ex-

tent to which there was any underinvestment due to the lack of gov-

ernment in education. There was extensive ﬁnancing and provision of

non-proﬁt schooling from non-governmental sources even prior to the

initial government funding of elementary schooling in 1833. The funds

came from a diverse mix of organisations and individuals and with di-

verse aims, including the propagation of religious doctrine, philanthropy

and a sense of obligation to the local community. As early as 1818, some

40 per cent of the scholars recorded in a national parochial survey were