Bureau of Labor Statistics, U.S.

Department of Labor

INDUSTRIAL INJURIES AND THE BUSINESS CYCLE

Author(s): Max D. Kossoris
Source: Monthly Labor Review, Vol. 46, No. 3 (MARCH 1938), pp. 579-594
Published by: Bureau of Labor Statistics, U.S. Department of Labor
Stable URL: http://www.jstor.org/stable/41815400
Accessed: 13-03-2016 05:01 UTC

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at http://www.jstor.org/page/
info/about/policies/terms.jsp

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content
in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship.
For more information about JSTOR, please contact [email protected].

Bureau of Labor Statistics, U.S. Department of Labor is collaborating with JSTOR to digitize, preserve and extend access to
Monthly Labor Review.

http://www.jstor.org

This content downloaded from 132.174.254.159 on Sun, 13 Mar 2016 05:01:00 UTC
All use subject to JSTOR Terms and Conditions
 INDUSTRIAL INJURIES AND THE BUSINESS CYCLE

By Max D. Kossoris, ř/. S. Bureau of Labor Statistics

WHAT was the course of industrial injuries during the recent depres-
sion and subsequent recovery? An analysis made by the Bureau of
Labor Statistics of more than 426,000 disabling injuries for the years
1929 through 1935 reveals (1) that the trend of the frequency rate of
industrial injuries generally followed the trend of industrial employ-
ment, and (2) that the trend of the percentage of reported injuries
with a disability duration of 1 week or less followed the trend of in-
dustrial employment very closely. In other words, as employment
decreased, the frequency rate as well as the percentage of so-called
"1-week" cases generally decreased; and as employment increased,
both the frequency rate and percentage of 1-week cases generally
increased.
Trend of Industrial Injuries

For the composite group of 29 manufacturing industries, the trend

of the frequency rate of disabling industrial injuries moved in general
conformity with the trend of employment.1 (See table 1 and chart 1.)

Table 1. - Trends of Disabling Industrial Injuries and Employment in 29 Manu-

facturing Industries , 1929-35

Item 1929 1930 1931 1932 1933 1934 1935

Total number of disabling injuries studied i.. 82,734 62,561 55,142 36,104 54,494 67,168 68,182
Industrial injury frequency rate
Index of frequency rate (1929-35=100)

Index of employment (1929-35=100)

1 These data are from a number of States whose workmen's compensation laws require the reporting of
all injuries for which disability exceeds the day on which the injury occurred.

As employment dropped from an index of 132.4 in 1929 to 111.4 in

1930 and then to 90.7 in 1931 (using 1929-35 = 100), the frequency
rate dropped from an index of 110.7 to 87.0 But whereas the employ-
ment index continued to drop to a low of 75.7 for 1932, the frequency-
rate index rose slightly to 90.2 for that year. For the next 2 years
the two trends moved in the same direction - upward. The fre-
quency rate rose more sharply than employment in 1933, whereas
* The composite frequency rates given in table 1 were arrived at by weighting the frequency rate for each
industry according to the average number of wage earners, using census data where available and calculating
data for intercensus years by means of the indexes of employment computed by the Bureau of Labor Sta-
tistics. The employment indexes were calculated similarly, with the average for the period 192^-25= 100.
Because no frequency rates were available for this base period, the two series, i. e., frequency rates and em-
ployment, were recomputed with the annual average 1929-35= 100, thus making the two series strictly com-
parable. It is these two series which are shown on chart 1.
Ö7y

This content downloaded from 132.174.254.159 on Sun, 13 Mar 2016 05:01:00 UTC
All use subject to JSTOR Terms and Conditions
580 Monthly Labor Review - March 1938

employment rose more sharply during 1934. In 1935, the two trends
moved in opposite directions - the employment index continued to
increase, from 99.6 in 1934 to 105.4 in 1935, while the frequency-rate
index decreased from 104.4 to 99.0, with a decrease of more than one
disabling injury per million hours worked. On the whole, as is
clearly apparent from chart 1, the two trends moved in the same
general direction, with 1935 the only notable exception.

Chart I

TRENDS OF INDUSTRIAL INJURY FREQUENCY RATES AND

EMPLOYMENT IN 29 MANUFACTURING INDUSTRIES, 1929-1935

Why should there be this relationship between the frequency rate

and employment? The frequency rate measures the average number
of disabling injuries per million man-hours, and therefore has no
direct relation to the volume of employment. Theoretically, the
frequency rate could remain constant regardless of whether the volume
of employment moved up or down. All that is required to reach
this result is that there shall be the same average number of dis-

This content downloaded from 132.174.254.159 on Sun, 13 Mar 2016 05:01:00 UTC
All use subject to JSTOR Terms and Conditions
 Industrial Injuries and Business Cycle 581

abling injuries per million man-hours worked. This relationship

need not be affected by any fluctuations in the volume of employment.
The reasons for fluctuations in the frequency rate generally, and
the close correlation of its trend with that of the employment index,
must be sought in the economic factors which accompanied the
depression. It certainly cannot be said that the downward trend
during the depression years was due to greater managerial concern
for safety during this period, for in many instances safety activities
were discontinued or severely curtailed as economy measures. Con-
sequently, it would be in order to anticipate rising frequency rates
rather than falling rates during depression periods. Why, then, did
the rate fall so sharply from the level of 1929 to the levels of 1931
and 1932? Four reasons are suggested here, the first three of which
are generally accepted by safety engineers, whereas the fourth follows
from an analysis of the data:
1. As the depression deepened, labor forces were curtailed, with
those most recently added laid off first. This generally left employed
workers with long years of service and, usually, those of the skilled
or semiskilled types which management wanted to retain as a nucleus
for subsequent expansion. Such workers generally were thoroughly
familiar with job hazards and had developed safety habits which
were carried from job to job. As a result, there were fewer disabling
injuries per million hours worked, and the frequency rate decreased.
2. In the early stages of depression, lay-offs tended to lag behind
reductions in operation, with the result that the total number of man-
hours worked exceeded those which would ordinarily have been
required. Coupled with decreased numbers of injuries, attributable to
a general slowing down in operating tempo, the swollen man-hours
total resulted in a lowering of the frequency rate (which is the average
number of disabling injuries per million man-hours).2
3. As business conditions became worse, management shifted toward
the use of the most efficient equipment, which generally meant the
most modern equipment. Such equipment, as a general rule, was also
the safest. This factor in itself would be sufficient to reduce the
frequency rate. Coupled with a more skilled and safety-conscious
labor force, it was still more effective.
4. As the depression deepened, the number of reported disabilities
not exceeding 1 week decreased out of proportion to the general de-
crease in industrial injuries generally. As will be pointed out later,
there is a strong probability that many minor injuries, which under
more normal conditions would have caused disabilities ranging from
• In connection with man-hours, one other point which may affect the frequency rate must be noted:
There is a possibility that during the depression man-hours were overestimated by reporting employers
who may not have allowed sufficiently for part-time operations. The reporting requirements instituted
under N. R. A. codes may have overcome this difficulty. The overestimating of man-hours worked during
the depression years, if it occurred, would have operated to depress the frequency rate.

This content downloaded from 132.174.254.159 on Sun, 13 Mar 2016 05:01:00 UTC
All use subject to JSTOR Terms and Conditions
 582 Monthly Labor Review - March 1938

1 to 3 days, were not reported by workers. As a result, the number of

disabling injuries which entered into the calculation of the frequency
rate was depressed.3
From these explanations it can also be readily understood why the
frequency rate should have risen as employment increased. With the
relatively small increase in employment from an index of 75.7 in 1932
to 84.7 in 1933, the frequency-rate index rose from 90.2 to 102.3 -
which, as a composite rate for the entire group of 29 industries, is a
very sharp increase. The obvious reason for this rise is that increases
in employment meant hiring of workers not accustomed to the hazards
of their new jobs, or workers whose safety habits had been blunted
through lengthy lay-offs and financial worries, and who, perhaps,
were too eager to make a favorable showing.
Evidence that newly added employees, as a group, were the greatest
source of disabling injuries is furnished by the petroleum industry.4
The following data are cited from a recent study by a large oil
company:

Table 2. - Relation of Accidental Injuries to Length of Employment

Percent of
Percent of total dis-
Length of employment total em- abling
ployees accidental
injuries

Less than 1 year

1 and under 2 years

2 and under 3 years

3 and under 5 years

Over 5 years

While employees with less than 1 year's service, i. e., newly added
workers, formed only 15.1 percent of the total number of employees,
they had 33.0 percent of all disabling accidental injuries. Workers
with 2 and 3 years of employment also had a heavier percentage of
injuries than their proportionate representation in the group as a
whole. Not until the group with 3 and under 5 years of employment
is reached does the percentage of total injuries fall below the per-
centage of total employment. Workers with less than 3 years' service,
1 It has been suggested that this decrease might be due, in part at least, to a practice on the part of bodies
administering workmen's compensation acts of discouraging the reporting of disabling accidents which do
not exceed the waiting period in disability and therefore were not compensable. The need for economy
and reduced budgets are cited as justification. No such practices have been found, however, and the
definite downward and subsequent upward trend of 1-week cases do not bear out any such assumptions.
It has therefore been disregarded.
« American Petroleum Institute. Accident Prevention Information, No. 67, April 15, 1937.
Similar data are given by the Portland Cement Association in its Accident Prevention Magazine of
August 1937, covering the 1936 experience of the industry. Workers. with less than 6 months' experience,
although only 3 percent of the total, had 20 percent of; all disabling injuries. Workers with less than 1
year's experience, only 5 percent of the total, had about 25 percent of all disabling injuries.

This content downloaded from 132.174.254.159 on Sun, 13 Mar 2016 05:01:00 UTC
All use subject to JSTOR Terms and Conditions
 Industrial Injuries and Business Cycle 583

comprising only 31.8 percent of the total working force, had 56.1
percent of all disabling accidents.
One objection which may logically be raised against the conclusion
that workers newly employed with a given concern have a higher
proportion of disabling accidents than employees with longer periods
of service is that many of such workers, while new to the particular
establishment, may not be new to the industry. Consequently these
workers should have some acquaintance with the accident hazards of
their new jobs. While this is undoubtedly true, it is also well recog-
nized that many of these workers had been out of work for a con-
siderable period of time, with the result that their safety habits may
have become blunted, and particularly so because the new environ-
ment may have presented somewhat different situations and may have
resulted in different psychological factors than they had to cope with
before. It is also a fairly common practice to shift workers with a
longer service record to less hazardous assignments, thus leaving the
more hazardous jobs to newly acquire'd workers.
From a low point of 75.7 in 1932, the volume of employment in the
29 manufacturing industries increased to an index of 99.6 in 1934
and 105.4 in 1935. With the volume of employment moving up
rather slowly in these 2 years, the frequency rate of injuries dropped
from 22.62 to 21.46, probably as a result of the accumulation of ex-
perience by workers taken on during 1933 and 1934. Data from other
sources indicate that 1936 will show a still lower frequency rate against
a continued increase in the index of employment.
If the experience described for the period from 1929 through 1935
may be taken as typical, it may be concluded that for the total of the
29 industries studied, and very likely for all manufacturing industries,
the course of the injury frequency rate during a business cycle is
about as follows:
1. As employment decreases, the frequency rate falls sharply. The
low points of the frequency rate and employment index tend to coincide
as to year.
2ě With the first decided increase in employment, the frequency
rate rises sharply, apparently because of the return to jobs of new or
rehired former employees.
3. Subsequent increases in employment are accompanied by less
decided increases in the frequency rate.
4. As employment tends to approach a plateau, the frequency rate
turns downward, apparently because of the increasing skill and de-
velopment of safety habits - and perhaps a greater feeling of security
in the job - of workers hired or rehired during the preceding year.
A corroboration of these conclusions, except for point 4, is furnished
by data for the State of Virginia. The following data start with 1931
46585 - 38 ■ 2

This content downloaded from 132.174.254.159 on Sun, 13 Mar 2016 05:01:00 UTC
All use subject to JSTOR Terms and Conditions
 584 Monthly Labor Review - March 1938

rather than 1929, but the general trends are very similar to those
already discussed.
Table 3. - Injury Frequency Record in Virginia Manufacturing Industries, 1931-35 1

Index 1

Year Number of Totha'™an" Disabling ~

employees w0°r^ injuries Q?atey Employ. Fre-
QrXy

1931

1932

1933

1934

1935

i Data furnished by Industrial Commission of Virginia.

1 Annual average 1931-35=100.

It is obvious from table 3 that, as in the national data, there is a

close relationship between the trends of employment and the accident
frequency rate. The employment index dropped from 94.0 in 1931 to
86.1 in 1932, and then rose year by year, reaching 118.9 in 1935. The
frequency-rate index dropped from 107.7 in 1931 to 90.7 in 1932, and
then rose year by year to 108.7 in 1935. The two trends moved
together, both downward and upward. For the year 1935, for which
it was possible to analyze more detailed figures, the following increases
were found in the second half of the year over the first half : Employ-
ment, 8 percent; man-hours worked, 13 percent; number of disabling •
injuries, 28 percent; and injury frequency rate, 13 percent. An
increase of 8 percent in employment, then, was accompanied by a 13
percent increase in the number of disabling accidents per million
hours worked.
Another corroboration of the conclusion that the trends of employ-
ment and injury frequency rates generally move together is given by
data published by the National Safety Council. Although there are
radical differences in the scope and composition of the data used by
the Council and the Bureau of Labor Statistics, it is significant that
the general direction of the trends of the series from the two sources
are the same. Reducing the frequency rates published by the
National Safety Council in its Accident Facts, 1936 edition, to an
index having the annual average of 1929-35 = 100, we arrive at the
following index numbers of frequency rates:
1929

1930

1931

1932

Although this pattern does not cover the same group of industries
and concerns itself with a more highly select group of firms than the
sample covered by the Bureau of Labor Statistics, it coincides in
direction of trend with that for the 29 manufacturing industries, even
to the downward trend from 1934 to 1935.

This content downloaded from 132.174.254.159 on Sun, 13 Mar 2016 05:01:00 UTC
All use subject to JSTOR Terms and Conditions
 Industrial Injuries and Business Cycle 585

In its most recent publication of frequency rates, the Council gives

a frequency rate of 13.57 for establishments reporting to it for 1936,
as against a rate of 14.02 for 1935, bearing out the thought already
indicated that with further increases in employment during 1936 there
would be a further decrease in the frequency rate.

Experiences of Individual Industries

As is often the case in the analysis of a large number of industries,

not all of them comply with the composite pattern for the entire
group. The trends in most of the 29 industries, however, are suffi-
ciently like that for the entire group to justify the conclusion that the
latter is representative, and that the general trends already discussed
are not due to the overbalancing of some industries moving in one
direction by others moving in the opposite direction. Table 4 gives
the individual industry data, and permits conclusions concerning the
trends of frequency rates and employment indexes. The last column
contains an evaluation of the degree of direct correlation of the two
trends for each industry. The table also shows the frequency rate for
each industry by type of disability - death, permanent injuries (includ-
ing permanent total and permanent partial disabilities), and temporary
disabilities.

Table 4. - Injury Frequency Rates and Employment Indexes for 29 Manufacturing

Industries , 1929-35
[1923-25=100]

Kelation of
total fre-
Item 1929 1930 1931 1932 1933 1934 1935 quency rate
to employ-
ment index

All industries:
Frequency rate
Deaths

Permanent disabilities
Temporary disabilities

Employment index

Agricultural implements:
Frequency rate

Deaths

Permanent disabilities
Temporary disabilities.

Employment index
Automobiles:
Frequency rate
Deaths

Permanent disabilities

Temporary disabilities

Employment index

Automobile tires and rubber goods:

Frequency rate

Deaths

Permanent disabilities
Temporary disabilities.

Employment index
Boots and shoes:
Frequency rate

Deaths

Permanent disabilities.
Temporary disabilities.

Employment index

This content downloaded from 132.174.254.159 on Sun, 13 Mar 2016 05:01:00 UTC
All use subject to JSTOR Terms and Conditions
 586 Monthly Labor Review - March 1938
Table 4. - Injury Frequency Rates and Employment Indexes for 29 Manufacturing
Industries , 1929-35 - Continued

Relation of
total fre-
Item 1929 1930 1931 1932 1933 1934 1935 quency rate
to employ-
ment index

Brick, tile, and terra cotta:

Frequency rate
Deaths

Permanent disabilities
Temporary disabilities
Employment index

Carpets and rugs:

Frequency rate
Deaths

Permanent disabilities

Temporary disabilities

Employment index
Chemicals:
Frequency rate
Deaths

Permanent disabilities

Temporary disabilities

Employment index

Cotton goods:
Frequency rate
Deaths

Permanent disabilities
Temporary disabilities

Employment index

Electrical machinery, apparatus, and

supplies:
Frequency rate

Deaths

Permanent disabilities

Temporary disabilities

Employment index

Fertilizer: rPnor- fr«-

Frequencyrate
Deaths

Permanent

Temporary disabUities

Employment index
Flour, feed, and other grain-mill prod-
ucts:
Frequency rate
Deaths

Permanent disabilities

Temporary disabUities

Employment index »

Foundry and machine-shop products:

Frequency rate
Deaths

Permanent disabilities

Temporary disabilities
Employment index
Furniture:
Frequency rate
Deaths

Permanent disabilities

Temporary disabUities

Employment index
Olass:
Frequency rate
Deaths

Permanent disabUities

Temporary disabUities

Employment index
Hudw&ro*
Frequency rate
Deaths

Permanent disabilities

Temporary disabilities
Employment index
Iron and steel:
Frequency rate
Deaths

Permanent disabilities

Temporary disabilities
Employment index 3

i These rates appear to be distorted, apparently due to erratic fluctuations in the reporting sample for 1931 .
* Employment index for flour.
s Employment index for blast furnaces, steel works, and roUing mUls.

This content downloaded from 132.174.254.159 on Sun, 13 Mar 2016 05:01:00 UTC
All use subject to JSTOR Terms and Conditions
 Industrial Injuries and Business Cycle 587
Table 4. - Injury Frequency Rates and Employment Indexes for 29 Manufacturing
Industries , 1929-35 - Continued
Relation of
total fre-
Item 1929 1930 1931 1932 1933 1934 1935 quency rate
to employ-
ment index

Leather:
Frequency rate
Deaths

Permanent disabilities

Temporary disabilities

Employment index

Lumber- logging and sawmills:

Frequency rate
Deaths

Permanent disabilities

Temporary disabilities

Employment index

Lumber- planing mills:

Frequency rate
Deaths

Permanent disabilities
Temporary disabilities

Employment index *

Machine tools*
Frequency rate
Deaths

Permanent disabilities

Temporary disabilities

Employment index

Paper and pulp:

Frequency rate
Deaths

Permanent disabilities

Temporary disabili ties

Employment index

Petroleum refining:
Frequency rate
Deaths

Permanent

Temporary disabilities

Employment index

Pottery:
Frequency rate
Deaths

Permanent disabilities..

Temporary disabilities

Employment index

Shipbuilding, steel and wood: rpwiv ni™«

Frequency rate
Deaths

Permanent

Temporary disabilities

Employment index

Slaughtering and meat packing:

Frequencyrate
Deaths

Permanent disabilities

Temporary disabilities

Employment index

Stamped and enameled ware:

Frequencyrate
Deaths

Permanent

Temporary disabilities

Employment index

Steam fittings, apparatus, and supplies:

Frequencyrate
Deaths

Permanent disabilities

Temporary disabilities

Employment index
Stoves:
Frequencyrate
Deaths

Permanent disabilities
Temporary disabilities

Employment index

Woolen goods:
Frequencyrate...

Deaths

Permanent

Temporary disabilities

Employment index

t¡ »¡These rates appear to be distorted, apparently due to erratic fluctuations in the reporting sample for 1931 .
^Employment Index for lumber and millwork.

This content downloaded from 132.174.254.159 on Sun, 13 Mar 2016 05:01:00 UTC
All use subject to JSTOR Terms and Conditions
588 Monthly Labor Review - March 1938

In 3 industries, the trends of frequency rates and the employment

index either moved in the same direction or closely followed the entire
group pattern. In 12 more, the degree of relationship was not so
marked, but nevertheless close. In 10 industries the relationship was
fairly close, but still definitely apparent. In 4 industries, however,
there appeared to be no relationship at all.
The movements of the frequency rates in a number of industries
require special comment. In the fertilizer industry, the trend of
employment followed the group trend until 1935, when the index for
the fertilizer industry dropped, while that for all manufacturing
industries continued upward. The frequency rate, however, in-
creased from 31.23 in 1930 to 48.54 in 1934, increasing each year over
the rate of the preceding year. In 1935, however, the rate decreased
to 46.73. In 1934, both the frequency rate and the employment index
reached their highest points since 1930, when the employment index
stood at 111.0. Just why there should have been this constant in-
crease in the frequency rate - an increase of about 55 percent - from
1930 to 1934 is not clear. Perhaps it is to be attributed to the high
seasonality of the industry, with employees hired, as a rule, only for
a relatively short busy season. Under such conditions increased
labor forces seem to result in increased numbers of injuries out of
proportion to the increases in employment.
The logging and sawmill industry had a very similar experience.
Regardless of the upward and downward fluctuations in employment,
the injury frequency rate moved steadily upward, from 46.79 in 1929
to 68.28 in 1935, an increase of about 45 percent, or 21 additional
disabling injuries for every million hours worked. Again, as in the
fertilizer industry, this is an instance in which a considerable portion
of the industry's activities are highly seasonal. It should be noted
also that the frequency rate of injuries was higher in this industry
than in any other industry listed.
Although the frequency rate of the boot and shoe industry was
generally low, it nevertheless indicates a fairly steady upward trend
during the period from 1930 to 1935. With only one exception, 1933,
the rate increased steadily from 6.95 in 1930 to 10.57 in 1934, de-
creasing slightly to 10.18 in 1935. As will be noted, the increases
occurred in permanent as well as temporary disabilities.
In the automobile industry, on the other hand, the trend in the
frequency rate was generally in a downward direction, decreasing
from 22.76 in 1929 to 10.77 in 1935, a decrease of more than half,
much of it occurring in 1935.
An exactly contrary experience was that of the petroleum-refining
industry. Except for a decrease in employment in 1935, the employ-
ment-index pattern resembles that of the composite manufacturing-

This content downloaded from 132.174.254.159 on Sun, 13 Mar 2016 05:01:00 UTC
All use subject to JSTOR Terms and Conditions
 Industrial Injuries and Business Cycle 589

industries group. But the frequency rate, which stood at 31.36 in

1930, decreased steadily to a low point of 10.46 in 1935 - a decrease
in the 6-year period of nearly 67 percent, or 21 fewer disabling injuries
for every million man-hours worked. There is a good reason to
attribute this highly satisfactory experience to careful, continuous,
and comprehensive safety work on the part of management in this
industry. Perhaps an additional reason lies in the fact that the
industry did not suffer the extreme swings of employment found in a
considerable number of the other industries.
The shipbuilding industry, for which the employment-index pattern
conforms fairly well to that of the entire group, also shows an almost
constant decline in the frequency rate from 1930 on. Dropping from
107.3 in 1930 to 56.8 in 1933, the employment index recovered to 76.5
in 1935. The frequency rates of industrial injuries for 1930 and 1933
were 29.66 and 19.32, respectively. But whereas the employment
index increased for the next 2 years, the frequency rate continued to
fall to 15.56 in 1935 - a drop of nearly 50 percent.
A third industry in this group with constantly decreasing frequency
rates is that producing automobile tires and rubber goods. Except
for a slight rise in 1933, the rate for this industry decreased steadily
from 25.19 in 1929 to 16.16 in 1935, a decrease of 9 injuries for every
million hours worked.

Trend of Minor Injuries During Depression

It has already been mentioned that as the depression deepened,
there was a consistent decrease in the percentage which disabilities of
less than 8 days' duration formed of all temporary total injuries.
There is a very strong probability that the decrease occurred most
heavily in accidental injuries which otherwise would be classed in the
1, 2, or 3 days' disability groups. The statistical data utilized, how-
ever, do not permit of a nicer analysis of reported injuries than " 1
week or less." As the data for 1929 are not available in sufficient
detail, the period covered is from 1930 to 1935. It was also necessary
to omit the iron and steel industry, as no detailed classification of
temporary total disabilities was possible for this group.
In table 5 is shown for the entire group, as well as for each of the 28
industries, the percentage which these 1-week cases were of the total
number of temporary total disabilities. The employment indexes
have been computed with the average for the period 1923-25 = 100.
To permit a direct comparison of the trend of 1-week cases with the
trend of employment, both types of data were reduced to index num-

This content downloaded from 132.174.254.159 on Sun, 13 Mar 2016 05:01:00 UTC
All use subject to JSTOR Terms and Conditions
590 Monthly Labor Review - March 1938

bers with the annual average for the period 1930-35=100. The two
series, shown graphically in chart 2, are as follows:
Index of Index of
1 -week cases employment
1930

1931

1932

1933

1934

1935

Chart 2

TRENDS OF PERCENTAGE OF ONE-WEEK CASES AND

EMPLOYMENT IN 28 MANUFACTURING INDUSTRIES, 19301935

This content downloaded from 132.174.254.159 on Sun, 13 Mar 2016 05:01:00 UTC
All use subject to JSTOR Terms and Conditions
 Industrial Injuries and Business Cycle 591
Table 5. - Percent of Temporary Total Disabilities in 28 Manufacturing Industries ,
Reported as of 1 Week or Less, 1930-35 1

Correlation of
Item 1930 1931 1932 1933 1934 1935 direction of
trend

All industries:
Cases of temporary total disability
Percent first week

Employment index 2

Agricultural Implements:
Cases of temporary total disability

Percent first week

Employment index
Automobiles:
Cases of temporary total disability

Percent first week

Employment index
Automobile tires and rubber goods:
Cases of temporary total disability
Percent first week

Employment index
Boots and shoes:
Cases of temporary total disability
Percent first week

Employment index..

Brick, tile, and terra cotta:

Cases of temporary total disability
Percent first week

Employment index

Carpets and rugs: ("Number of

Cases of temporary total disability

Percent first week

Employment index
I sion.
Chemicals:
Cases of temporary total disability

Percent first week

Employment index

Cotton poods:
Cases of temporary total disability

Percent first week

Employment index
Electrical machinery, apparatus, and sup-
plies:
Cases of temporary total disability

Percent first week

Employment index
Fertilizers:
Cases of temporary total disability
Percent first week

Employment index

Flour, feed, and other grain-mill products:

Cases of temporary total disability
Percent first week

Employment index

Foundry and machine-shop products:

Cases of temporary total disability
Percent first week

Employment index

Furniture:
Cases of temporary total disability

Percent first week

Employment index

Glass:
Cases of temporary total disability

Percent first week

Employment index.

Q&rdwdro *
Cases of temporary total disability

Percent first week

Employment index..

Leather:
Cases of temporary total disability
Percent first week

Employment index

Logging and sawmills:

Cases of temporary total disability
Percent first week

Employment index

1 1929 omitted because the necessary break-down is not available.

7 For all employment indexes, the average for the period 1923-25=100,

This content downloaded from 132.174.254.159 on Sun, 13 Mar 2016 05:01:00 UTC
All use subject to JSTOR Terms and Conditions
 592 Monthly Labor Review - March 1938
Table 5. - Percent of Temporary Total Disabilities in 28 Manufacturing Industries ,
Reported as of 1 Week or Less , 1930-35 - Continued

Correlation of
Item 1930 1931 1932 1933 1934 1935 direction of
trend

Planing mills:
Cases of temporary total disability
Percent first week.

Employment index »
Machine tools:
Cases of temporary total disability

Percent first week-

Employment index
Paper and pulp:
Cases of temporary total disability
Percent first week.
Employment index
Petroleum refining:
Cases of temporary total disability
Percent first week

Employment index

Pottery:
Cases of temporary total disability
Percent first week
Employment index
Shipbuilding:
Cases of temporary total disability
Percent first week

Employment index..

Slaughtering and meat packing:

Cases of temporary total disability
Percent first week

Employment index

Stamped and enameled ware:

Cases of temporary total disability
Percent first week

Employment index
Steam fittings, apparatus, and supplies:
Cases of temporary total disability
Percent first week

Employment index

Stoves:
Cases of temporary total disability

Percent first week

Employment index
Woolen goods:
Cases of temporary total disability
Percent first week

Employment index

• Employment index is for lumber and millwork.

The two series moved in the same downward and upward direction
in each of the 6 years. When the employment index went down,
the index of 1-week cases went down, and when the employment
index moved up, the index of 1-week cases moved up. When the
employment index was at its lowest point, the index of 1-week cases
was at its lowest point.
An analysis of table 5 shows that this relationship is found for a
very heavy percentage of the individual industries, indicating that
the phenomenon was general. There are decided differences between
the experiences of individual industries, as one would expect. In the
furniture industry, for instance, the proportion of 1-week cases was
high, indicating that about half of all temporary total disabilities
were due to minor injuries. In logging and sawmills, on the other
hand, the percentage of 1-week cases was low, indicating a prepond-
erance of more serious temporary total disabilities. In spite of all

This content downloaded from 132.174.254.159 on Sun, 13 Mar 2016 05:01:00 UTC
All use subject to JSTOR Terms and Conditions
 Industrial Injuries and Business Cycle 593

these differences, however, the analysis reveals that out of the 28

industries, 20 showed definite trend relationships between percent of
1-week cases and employment. In seven industries the relationship
was too erratic to permit any conclusion, and in one industry the
number of injuries were too few to justify a conclusion.
It therefore appears that the fluctuation of the percentage which
1-week disability cases were of the entire number of temporary total
disabilities was not at random, but that it followed a fairly definite
pattern, which closely resembles that of movements in employment
in these industries.
Is there any logical reason why there should have been a smaller
percentage of temporary disabilities of a minor-disability character
when employment was low, and a higher percentage when employ-
ment was up? Is there any logical reason why there should have
been the close relationship in direction and timing?
The answer to these questions must be in the negative. There
seems to be no logical reason why there should be this relationship.
Nor does there seem to be any logical reason for the definite down-
ward and upward movement of the percentage of 1-week cases. No
matter how many temporary injuries there are, no matter how the
number varies, the percentage of minor injuries should not vary
markedly from year to year, except for random fluctuations. Out
of a given number of temporary total injuries, the percentage of
1-week cases should remain approximately the same from year to
year. What variations there are should not follow any definite
pattern.
What, then, accounts for the trend in the percentage of 1-week
cases? One explanation is that there actually are fewer minor
injuries during the depression because the less experienced workers
have been laid off, and that there are proportionately more minor
injuries during periods of increased industrial activity due to the
accretion of inexperienced workers. This explanation, however, does
not appear to be feasible. True, there are less industrial injuries as
the less experienced help is laid off, and more as new help is added.
But the actual distribution of disabilities according to severity should
not be affected. Experienced help should have fewer accidental
injuries - but why should proportionally more of the injuries which
do result be of a more severe character?
The most logical solution to the problem appears to be that many
minor injuries are not reported by workers, or in any case are with-
held from reporting to State authorities. Failure to make such reports
are of course reflected in statistics of accidental injuries, for if
injuries are concealed by injured workers, they do not reach the
State authorities from whom the statistics under discussion are
obtained; nor, for that matter, do they all reach the proper officials
in private manufacturing establishments.

This content downloaded from 132.174.254.159 on Sun, 13 Mar 2016 05:01:00 UTC
All use subject to JSTOR Terms and Conditions
594 Monthly Labor Review - March 1938

Why should there have been this withholding of reports? The

answer to this question again goes back to the effects of the depression.
When jobs are scarce, and many employees are working only part
time and generally at reduced wage rates, there is every reason why
a worker with a minor injury should keep on working. One reason is
that he cannot afford to jeopardize his own job. Further, with
decreased earnings, resulting from decreased hourly rates or reduced
working hours or both, he needs every dollar he can earn. Finally,
very few States provide workmen's compensation benefits for dis-
abilities lasting less than a week, and when they do, the compensation
allowed is generally less than the earnings lost. The worker, there-
fore, either goes along as well as he can at his regular job, or, if he
has a sympathetic foreman, may be shifted to another job which he
may be able to perform in spite of his injury. But when the employ-
ment situation is easier, when hours and wage rates are more normal,
he is inclined to take time out if injured, even though it be only 1 or
2 days. At the same time, employees, particularly those recently
added, may not receive as sympathetic consideration from super-
visors and may be compelled to take time off until able to continue
the tasks for which they are employed so as to maintain the speed
of operations.
The understatement of disabling injuries in statistics covering the
depression years is estimated to have been at least 8 percent during
the worst depression year of 1932.6 The round figure of 10 percent
seems more nearly correct, and even so is believed to be a conservative
estimate. The following frequency rates, corrected for the given
percentage of 1-week cases and the constant of 39 percent, are given
in comparison with the percentages of such cases as computed directly
from the reported data:
Frequency rate
Computed Corrected
1930

1931

1932

« The figure of 8 percent was arrived at by assuming that, for the group as a whole, 39 percent of all tem-
porary total disabilities normally would be in the 1-week group. This is the figure for 1928, when the
employment index stood at 98.7, and occurs again for 1933, a year which combines low employment with
the first upward impulse under N. R. A.

This content downloaded from 132.174.254.159 on Sun, 13 Mar 2016 05:01:00 UTC
All use subject to JSTOR Terms and Conditions