Content uploaded by Robert V. Levine
Author content
All content in this area was uploaded by Robert V. Levine on Oct 23, 2014
Content may be subject to copyright.
JOURNALOFCROSS-CULTURAL PSYCHOLOGY
Levine,Norenzayan/ PACE OF LIFE
This study compared the pace of life in large cities from 31 countries around the world.
Three indicators of pace of life were observed: average walking speed in downtown
locations, the speed with which postal clerks completed a simple request (work speed),
and the accuracy of public clocks. Overall, pace of life was fastest in Japan and the coun-
tries of Western Europe and was slowest in economically undeveloped countries. The
pace was significantly faster in colder climates, economically productive countries, and
in individualistic cultures. Faster places also tended to have higher rates of death from
coronary heart disease, higher smoking rates, and greater subjective well-being. Discus-
sion focuses on how the pace of life is intertwined with the social-psychological and
community characteristics of a culture, and the central role of pace of life in defining the
personality of a place and its people.
THE PACE OF LIFE IN 31 COUNTRIES
ROBERT V. LEVINE
California State University, Fresno
ARA NORENZAYAN
University of Michigan
Beyond doubt, the most salient characteristic of life in this latter por-
tionofthe19thcenturyisitsSPEED,—whatwemaycallitshurry,the
rate at which we move, the high-pressure at which we work;—and the
question to be considered is, first, whether this rapid rate is in itself a
good; and, next, whether it is worth the price we pay for it—a price
reckoned up, and not very easy thoroughly to ascertain.
—W. R. Greg (1877) on “Life at High Pressure” (p. 263)
The pace of life has been defined as the rate (Lauer, 1981); speed (Amato,
1983); and “relative rapidity or density of experiences, meanings, percep-
tions and activities” (Werner, Altman, & Oxley, 1985, p. 14). There is evi-
dencethat a city’spaceoflife is relativelystableacross measures. Levine and
Bartlett (1984), for example, found correlations ranging from .52 to .82
across 12 cities from six different countries between three diverse measures
178
AUTHORS’NOTE: We wish to thank Karen Bassoni, Adrian Bivalarou, Gary Brase, Stephen Buggie, Andy
Chuang, Holly Clark, John Evans, Robert Lautner, Royce Lee, Andy Levine, Todd Martinez, Michiko Mori-
yama, Carlos Navarette, Julie Parravano, Aroldo Rodrigues, Sanjay Sinha, David Tan, Jyoto Verma, Karen
Villerama, Sachiko Watanabe, and Lori Yancura for their assistance gathering field data; Andy Chuang,
KrisEyssell,Sheri Osborn,andJ. Rachelle Solesfor theirassistancein dataanalysis;and EdwardDienerand
JOURNAL OF CROSS-CULTURAL PSYCHOLOGY, Vol. 30 No. 2, March 1999 178-205
© 1999 Western Washington University
of the pace of life—walking speed, work speed among postal clerks, and
accuracy of bank clocks.
Severalstudieshavedemonstratedstrongandconsistentdifferencesinthe
overall pace of life of cities both within and between countries (e.g., Amato,
1983; Bornstein, 1979; Bornstein & Bornstein, 1976; Buggie, 1993, 1994;
Levine & Bartlett, 1984; Levine, West, & Reis, 1980; Lowin, Hottes, Sandler, &
Bornstein, 1971; Walmsley & Lewis, 1989; Wirtz & Ries, 1992; Wright,
1961). Fewer studies have attempted to explain these intercity differences. In
thosestudiesthathave,virtuallyallhavefocusedonthesinglefactorofpopu-
lation size. For example, Bornstein and Bornstein (1976) observed a strong
positive relationship between population size and the average walking speed
of pedestrians across a sample of 15 cities and towns in six countries in
Europe,Asia, and North America; Bornstein (1979) later replicated thisfind-
ing in a sample of six more cities in Europe and the United States. Amato
(1983) found that population size was significantly related to both walking
speed and a measure of work speed (the speed of betel nut transactions in
open market places) in a large city compared to small towns in New Guinea.
Levine and Bartlett (1984) found generally faster walking speeds and more
accurate timepieces in large versus medium-sized cities in six different
countries.Overall,then,thereissubstantialevidencethatbiggercitiestendto
have faster tempos.
Population size, however, is only one of many qualities that define the
character of a city and that distinguish cities from each other (e.g., Cutter,
1985; Krupat & Guild, 1980; Levine, Miyake, & Lee, 1989). Strauss has
argued that “the entire complex of urban life can be thought of as a person
rather than a distinctive place, and the city can be endowed with a personal-
ity of its own” (cited in Krupat & Guild, 1980, p. 21). Explaining why some
cities move faster than others requires an understanding of some of the com-
plexity of cities’“personalities” beyond simply the size of their populations.
Afewpreviouspace-of-lifestudies (most notably,Bornstein,1979)havedis-
cussedmultipleexplanationsforthepaceofplaces.Withthesingleexception
ofLevine,Lynch,Miyake,andLucia’s(1990)investigationofthepaceoflife
in 36 U.S. cities, however, these studies have ignored the larger constellation
of community characteristics that are related to its pace of life. What factors,
other than population size, might predict differences in the pace of life?
Levine, Norenzayan / PACE OF LIFE 179
Harry Triandis for generously sharing their data. We thank Michael Bond, Connie Jones, Richard Nisbett,
Aroldo Rodrigues, Lynnette Zelezny, and anonymous reviewers for their comments on earlier drafts of this
article. This project was partially supported by a grant from the School of Natural Sciences, California State
University, Fresno. Reprint requests should be sent to Robert V. Levine, c/o Department of Psychology, Cali-
fornia State University, Fresno, CA 93740-0011; e-mail: robertle@csufresno.edu.
Furthermore,what are the consequences of a community’s pace of life for the
well-being of its residents?
In large part, the lack of systematic assessment of other (than population
size) explanations of community-level differences in pace of life may be a
by-product of sampling shortcomings. All but Levine et al.’s (1990) 36-city
study and (marginally, at least) Bornstein and Bornstein’s (1976) 16-city
study have been limited to convenience samples of a few readily testable
large and small communities. To compare multiple predictors of tempo at
the city level, however, it is necessary to treat each city as the unit of analysis
in a correlational-type design. By treating each city as a single n, their pace-
of-life scores can then be correlated with other available community char-
acteristics, for example, community-level data reflecting population attrib-
utes, economics, and social-psychological qualities. These correlational
analyses, however, require testing considerably more “subjects” (i.e., cities)
than have most previous studies.
Furthermore, the two earlier studies that had tested large numbers of cities
were focused on six countries in Europe in one case (Bornstein & Bornstein,
1976) and on cities solely in the United States in the other (Levine et al.,
1990). Because there is relative homogeneity in the economic and social
makeup of these cities—compared to the larger range of international
differences—these did not present strong tests of the validity of multiple
predictors.Certainly,theydonotallowforstrongtestsoftherelationshipof
cultural values to the pace of life.
THE PRESENT STUDY
The present study compared three indicators of the pace of life in a sam-
ple of large cities in 31 countries. The study had three main goals: first, to
investigatedifferencesinthepace of life across large cities in a wide rangeof
countries; second, to examine which community characteristics best predict
these differences; and third, to explore the consequences of the pace of life
for the well-being of individuals and their communities.
The study focused on a large city—in most cases, the largest—in each
country. There were several reasons for this choice. First, for practical
reasons, it would have been difficult to carry out the field experiments in
small places in some countries. Second, although no single city represents
the entirety of a country, it seemed that the largest cities in each country
would be best matched for purposes of making meaningful cross-national
comparisons. Third, there is a strong population movement to large cities on
an international scale. It is estimated that more than half of the world’s popu-
lation will be living in urbanized areas within the next decade (Gottdiener,
180 JOURNAL OF CROSS-CULTURAL PSYCHOLOGY
1994). Fourth, many more statistics reflecting community characteristics are
available for large cities in each country than there are for nonmetropolitan
areas.
Littletheoretical attention has been paid to thequestionofhowthepaceof
life is related to other community characteristics. Given the lack of a strong
theory, the focus of the present study was to explore relationships between
pace and a number of other social variables not previously studied. Rather
than testing a single model or theory, our goal was to discover clusters of
regularitiesinthesocioeconomic patterns of cultures that relate to the paceof
life.Asabeginning,thepresentstudydrewonHoch’s(1976)theoryconcern-
ing the relationship between economic factors and the pace of life. Hoch
argues that the population density of large cities drives up the prices of land
and other goods. These economic demands require that people use their time
more efficiently so that greater economic value is assigned to people’s time,
which, in turn, leads to a faster pace of life. Conversely, of course, a faster
pace of life will in itself tend to produce more vital economies.
PREDICTORS OF PACE
Using Hoch’s economic model as a starting point, a series of hypotheses
concerning the relationship of the pace of life to other community variables
were developed. The first four of these hypotheses concerned the community
characteristics that are predictors of the pace of life.
Hypothesis 1: Economic vitality: The more vital a city’s economy, the
faster its pace of life.Totesttheeconomichypothesis,threeindicatorsofeco-
nomic vitality were examined. First, as a measure of the economic vitality of
the country as a whole, we looked at the relationship of pace of life to each
country’s gross domestic product (GDP) per capita. Next, as an indicator of
the economic well-being that is experienced by the average citizen, we
examined what is known as purchasing power parity (PPP). This is an esti-
mate of how muchtheaverageincomeearnedineachcountryiscapable of
purchasing.Finally, as an estimate of how well people are able to fulfill their
minimum physiological needs, the average caloric intake of each nation was
examined.
Hypothesis 2: Climate: Hotter places are slower. The popular, but mostly
untested, stereotype that the pace of life is slower in warmer climates may be
predicted from two perspectives. First, according to economic theory, if the
paceof life is drivenbyaneed for consumer goods, it may be argued that peo-
pleinwarmerclimates,whorequirefewerandlesscostlybelongings—fewer
Levine, Norenzayan / PACE OF LIFE 181
clothes, simpler homes—should have less need for making every moment
productive,leadingtolessvalueplacedonpeople’stime,resultingina slower
paceof life. Second, a relationship between climate and the paceof life might
alsobe predicted from an ergonomic perspective.A rapid pace of life requires
moreenergy,andpeople in warmer climates have less physical energyingen-
eral (Bornstein, 1979; Levine, 1988). There is, in fact, some evidence that
people within the same city walk faster when the temperature is cooler (Hoel,
1968; Rotton, 1985). No studies, however, have determined whether the pace
of life differs between cities with different climates.
Hypothesis 3: Cultural values: Individualistic cultures are faster than
those emphasizing collectivism. It was predicted that a culture’s basic value
system would be related to its pace of life. We focused on what many cross-
cultural psychologists believe to be the most important construct for explain-
ing the social patterns of cultures: individualism versus collectivism.
Roughly, individualistic societies are characterized by an orientation to the
individual and their nuclear family, whereas collectivistic societies give
highest priority to the welfare of one or more larger collectives (Triandis,
1995). (For a discussion of subtypes of individualism-collectivism, see Sin-
gelis, Triandis, Bhawuk, & Gelfand, 1995.)
Triandis (1995) argues that a major antecedent of individualism is affluence,
whichwehaveargued is both a precursor to, and consequence of, a rapid pace
of life. Compared with collectivist cultures, individualistic ones put more
emphasis on individual achievement than on affiliation. Because an empha-
sis on individual achievement requires a greater concern with time than one
that focuses on social affiliation, we predicted that individualistic countries
would have a faster pace of life than those that place greater emphasis on the
well-being of the collective.
Hypothesis 4: Population size: Bigger cities have faster paces.This fourth
predictor of the pace of life, which follows directly from Hoch’s economic
theory, was only minimally retested in the present study. As discussed above,
this hypothesis has received substantial empirical support in previous studies.
Most of the studies that have looked at the relationship between population
size and pace have had a large range of city sizes, ranging from under 29,000
to up to more than 1 million. In our study, in contrast, 23 of the 28 cities
(population sizes were unavailable for three cities in our sample) had popula-
tions of more than 1 million. Our sample, therefore, does not allow an ade-
quate test of the population-pace hypothesis as construed by past research on
this topic. The present data did, however, allow us to examine to what extent
this relationship is constrained by a critical cutoff point in population size.
182 JOURNAL OF CROSS-CULTURAL PSYCHOLOGY
PACE AND WELL-BEING
The next three hypotheses focused on the consequences of the pace of life
for physical and psychological well-being.
Hypothesis 5: On the whole, faster places have higher rates of death from
coronary heart disease. Perhaps no single physical disease has been so asso-
ciated with the temporal demands of industrialization as coronary heart
disease. More than a century ago, Greg (1877) commented that
the anxiety to be on time, the hurrying pace—often the running to catch trains
(which are punctual in starting, whatever they may be in arriving)—cause a
daily wear and tear, as well as accelerated action of the heart, of which, in a few
months or years, most of us become unpleasantly conscious, and which, as we
all know, sometimes have a fatal and sudden termination. (pp. 267-268)
The association between economic stressors, time urgency, and coronary
heartdiseasehasbeen perhaps best described by the TypeA behavior pattern.
The original concept of the Type A personality proposed a global pattern of
behaviors that predispose individuals to coronary heart disease (CHD). The
core elements of this Type A pattern include extremes of aggressiveness,
easily aroused hostility, competitive achievement striving, and a chronic
sense of time urgency—the perpetual struggle to achieve a great many goals
in a short period of time (e.g., Friedman & Rosenman, 1974). From this, it
might be hypothesized that faster places will have higher rates of CHD.
Thereis considerable controversy,however, about the relationship of time
urgency to CHD. A number of subsequent researchers have presented evi-
dence that “hostility-anger” may be the single toxic element of the global
Type A pattern. They argue that the other Type A behaviors, including time
urgency, are unrelated to CHD (see, e.g., reviews by Booth-Kewley &
Friedman [1987] and Matthews [1988]). Nonetheless, recent studies have
found that scores on time urgency are just as strong or even stronger predic-
tors of CHD as are hostility-anger scores taken alone. Friedman and
Ghandour (1993), for example, found that the symptoms, traits, and psycho-
motor signs of time urgency were a better predictor of CHD in a sample of
patients than were hostility scores; in addition, time urgency scores were
almost as efficient in predicting CHD as were total scores on a global Type A
measure. (See Friedman, Fleischman, & Price, 1996, for a review of articles
suggesting that time urgency may be more important than hostility for pre-
dicting CHD.)
VirtuallyallTypeA-CHDstudieshavemeasured time urgencyby individ-
ual self-reports, either with questionnaires or structured interviews.
Levine, Norenzayan / PACE OF LIFE 183
Examining the relationship between the pace of life and CHD at the level of
the city offers a different perspective to the Type A debate. The single previ-
ousstudytodo this wasLevineet al.’s(1990) 36-city U.S. study,which found
a significant relationship (r= .51) between a city’s overall pace of life and its
rate of death from CHD. The present study was designed to see if this rela-
tionship would replicate at the more heterogeneous international level.
Hypothesis 6: Smoking rates are higher in faster places. From the eco-
nomic hypothesis, we might also predict that people in faster places will have
a tendency to engage in unhealthy behaviors. Levine et al. (1990) theorized
that the higher CHD rates in faster places that emerged in their 36-city study
may result from a tendency for people in time-urgent environments, who are
presumablyundergreaterstresstoachieve,to engage in unhealthy behaviors.
They hypothesized that one of these stress-related behaviors is cigarette
smoking,whichisawell-establishedriskfactorfor the developmentof CHD.
Although statistics concerning smoking rates were not available for all of the
36 cities in their sample, Levine et al. found preliminary support for the
smoking hypothesis when comparing regions: Smoking rates and CHD rates
were highest and the pace of life was fastest in the northeast, followed in
order by cities in the north central, south, and western regions. The present
study offered a further test of this hypothesis, at the international level.
Hypothesis 7: People in fast places have greater subjective well-being.
This hypothesis concerns a direct measure of psychological well-being:
ratings of how happy and/or satisfied people are with their lives. To assess
subjective well-being (SWB), we drew on previous large-scale surveys of
happiness and life satisfaction that have been conducted within a number of
countries. These survey data were available for 15 of the 31 countries in our
study, which allowed at least a preliminary test of the happiness question.
Two diametrically opposed predictions about the relationship of pace to
SWB may be derived from the economic hypothesis. On one hand, if eco-
nomicvitality leads to temporal stress, should not thesame stressors that lead
to physically destructive behaviors like cigarette smoking also detract from
general psychological well-being? This would lead to the prediction that a
slow pace of life makes for happier people.
On the other hand, the opposite prediction follows from the well-
established empirical finding that economic productivity is highly related to
SWB,bothat the levelof nations andof individuals. A recent 55-nation study
by Diener, Diener, and Diener (1995), for example, found high correlations
between national life satisfaction and a wide range of national economic
indicators, including per capita GDP, purchasing power, and the fulfillment
184 JOURNAL OF CROSS-CULTURAL PSYCHOLOGY
of basic needs. Because our basic hypothesis is that the pace of life is posi-
tively related to economic vitality, this economic argument would predict a
positive relationship between the pace of life and SWB.
METHOD
OVERVIEW
Three measures of the pace of life were sampled in large cities in 31
countries. These measures were correlated with statistics reflecting
climate, economic well-being, individualism-collectivism, CHD, cigarette
smoking, and life satisfaction for each city (when available) or country.
SUBJECTS (CITIES)
The “subjects” in this study were large cities in each of 31 countries. The
selection of countries was aimed at the widest possible sampling of the
regions and cultures of the world. For practical reasons, however, subject
selection was driven by convenience. Data collection at the various interna-
tional sites was conducted by (a) interested, responsible students who were
either traveling to foreign countries or returning to their home countries for
the summer or (b) cross-cultural psychologists and their students in other
countries who were willing to assist the authors. Although this method of
subject selection was clearly less than random,the final list of countries did
encompass a wide range of locales in North and South America, Europe, and
Asia.
In most countries, data were collected in either the largest city or a rival
major city: Amsterdam (the Netherlands), Athens (Greece), Bucharest
(Romania), Budapest (Hungary), Dublin (Ireland), Frankfurt (Germany),
Guanzhou (China), Hong Kong (Hong Kong),1Jakarta (Indonesia), London
(England), Mexico City (Mexico), Nairobi (Kenya), New York City (United
States), Paris (France), Prague (Czech Republic), Rio de Janeiro (Brazil),
Rome (Italy), San Jose (Costa Rica), San Salvador (El Salvador), Seoul (South
Korea), Singapore (Singapore), Sofia (Bulgaria), Stockholm (Sweden),
Taipei (Taiwan), Tokyo (Japan), Toronto (Canada), and Vienna (Austria). In
four other countries, again for reasons of convenience, the observations
were made in more than one city: In Poland, data were collected in
Wroclaw, Lodz, Poznan, Lublin, and Warsaw. In Switzerland, measures
were taken in both Bern and Zurich. In Syria and Jordan, most observa-
tions were made in the capital cities of Damascus and Amman, but some
Levine, Norenzayan / PACE OF LIFE 185
were done in secondary population centers. In each of these cases, data from
the different cities were combined for that country, after testing for cross-city
differences.
PROCEDURE
Three indicators of the pace of life were measured in each country. These
measures demonstrated minimal experimenter effects in two previous stud-
ies (Levine & Bartlett, 1984; Levine et al., 1990). All measures were taken
during a warm summer month between 1992 and 1995. The three measures
were the following:
Walking speed. Male and female walking speed over a distance of 60 feet
was measured in at least two locations in main downtown areas in each city.
Measurementsweretakenduringmain businesshourson clear summer days.
All locations were flat, unobstructed, had broad sidewalks, and were suffi-
ciently uncrowded to allow pedestrians to move at potentially maximum
speeds. To control for the effects of socializing, only pedestrians walking
alone were used. Children, individuals with obvious physical handicaps, and
window-shopperswere not timed. Thirty-fivemen and 35 women were timed
in most cities.
Postal speed. As a sample of work speed, the time it tookpostal workers to
complete a standard request for stamps was measured in each country. Postal
clerks at randomly selected post offices in each city were handed a note (to
minimize experimenter effects) in the native language, written by a native
speaker, requesting one stamp of a commonly used small denomination.
Along with this note, the clerk was given a denomination of paper currency
that required change in both coins and paper. In the United States, for
example,theclerk was handed a $5 bill with a request for one 32-cent stamp.
Theexperimenterin each city was a neatly dressed native or native-appearing
man or woman. The dependent measure was the time elapsed between the
passing of the note and completion of the request. A minimum of eight postal
clerks were approached in each city.
Clock accuracy. As a sample of concern with clock time, the accuracy of
15 clocks, in randomly selected downtown banks, were checked in each
country. The criterion for the correct time was that reported by the telephone
company.
186 JOURNAL OF CROSS-CULTURAL PSYCHOLOGY
EXPERIMENTERS
A total of 19 experimenters (11 male, 8 female), most often working indi-
vidually, each collected data in one or more of the cities. All experimenters
were college age, from the native country, or able to “pass” as natives, and
dressed neatly and casually.
Several steps were taken to ensure standardization and to minimize
experimenter effects. First, all experimenters received both a detailed
instructionsheetand on-site fieldtrainingin procedures for subject selection,
observation, and timing. The first author oversaw all such training. Second,
two of the measures were unobtrusive (walking speed and clock accuracy).
For postal clerk speed, steps were taken to ensure maximum standardization
of experimenters’behaviors; in particular, to eliminate the effects of experi-
menters’ talking speed and style, transactions were conducted with notes.
Third, techniques were used that ensured that experimenters could not prese-
lect participants (“blind” selection technique). Given the fact that most
experimenters were responsible for data collection in only one or two coun-
tries, it was difficult to conclusively analyze for experimenter effects. As
indicated earlier, however, our earlier studies with these measures indicated
that, when provided the same experimenter training, experimenter effects
were minimal. Finally, given that the data for each country in the present
study were mostly gathered by different experimenters, the clear trends for
consistent regional differences that are described in the Results section
would argue that any experimenter differences were not a dominant factor in
the overall results.
COMMUNITY VARIABLES
Climate.Climatewasmeasured by the average annual high temperature in
each city (or, where more than one city was tested in a county, the average of
their temperatures) (Conway & Liston, 1974; Wright, 1993). (Data for
humidity and temperature-humidity indexes, which may be a more sensitive
measure of climate, are not available for many international cities.)
Economic indicators. As described earlier, three economic measures,
each measuring a different aspect of economic vitality, were used. First, as a
measure of the economic health of the country as a whole, we examined each
country’sper capita GDP for 1993 (WorldBank, 1994).Second, as an indica-
tor of the economic well-being that is experienced by the average citizen,
Levine, Norenzayan / PACE OF LIFE 187
estimates of what is known as PPP—how much the average income earned in
each country is capable of purchasing—were extracted for the year 1993
(World Bank, 1994). Third, as an estimate of how well people are able to ful-
fill their minimum physiological needs, the average caloric intake of each
nation for 1989 (the most recent available data) was examined (Wright,
1993).
Individualism/collectivism. Triandis rated each of the countries ona1to10
scale, with 1 being the most collectivistic and 10 the most individualistic.
Some of these ratings were reported by Diener et al. (1995); the remainder were
obtained directly from Triandis (personal communications, February 19,
1994; January 9, 1995; and September 19, 1995). Triandis was blind to the
data and hypotheses in this study. Although these ratings were developed
somewhat subjectively, it should be noted that Diener et al. (1995) found a
correlation of .83 between Triandis’s collectivism ratings and Hofstede’s
(1980) empirically derived collectivism ratings. (Hofstede’s ratings were not
available for many of the countries in the present study.)
Population size. Although not a central hypothesis in this study (see
above), the population of the metropolitan area for each city, for the year
1992, was taken from U.S. Bureau of the Census (1993) or, when not avail-
able from that source, from the United Nations (1992).
CHD. Rates of death from ischemic heart disease (the sum of deaths for
acute myocardial infarction plus other ischemic heart diseases) for the latest
available year were taken from the United Nations (1992). Data were avail-
able for 26 countries in the sample.
Smoking rates. Smoking statistics were taken from Nicolaides-Bouman,
Wald,Forey,andLee’s(1993) compilation of cigarette-smoking statistics for
121 countries. Data reflect the percentage of manufactured cigarette con-
sumptionper adult (aged 15+), for men and women combined, in1982-1983.
SWB. Veenhoven (1993) has compiled and integrated the results of a large
number of previous national surveys of happiness and well-being. Many of
these surveys varied in the phrasing of questions and/or response options.
One survey, for example, asked respondents to rate how happy they were,
offering three response options ranging from very happy to not too happy.
Another survey asked people how satisfied they were with their lives, offer-
ing 11 response options varying from very satisfied to not satisfied. To cali-
brate the different surveys, Veenhoven asked a sample of expert researchers
188 JOURNAL OF CROSS-CULTURAL PSYCHOLOGY
inthe area of SWB (persons working on the WorldDatabase of Happiness) to
assign a value of 0 to 10 to each response option in each survey. This weight-
ing process produceda0to10Thurstone value for the SWB of each country,
based on the last (or only) survey conducted in that nation. These survey data
wereavailable for slightly less than one half(n=15)ofthenations in the pres-
ent sample.
RESULTS
Although multiple measurements were taken for each field measure in
each city, it should be noted that for purposes of analysis, the 31 countries
weretreatedastheunitofanalysis. Each data point represents the mean of the
sample for a given country.
OVERALL PACE-OF-LIFE INDEX
For each country, the three measures of pace were converted to z-scores,
which were then combined to produce an overall pace-of-life index. This
overall index significantly correlated with each of its components (walking
speed, r= .76; post office speed, r= .80; clock accuracy, r= .69; df = 29, p<
.001usingaone-tailed test in all cases). Walkingspeed and post officespeeds
weremorecloselyrelatedtoeachotherthaneitherwasto clock accuracy.The
intercorrelations between the three pace measures were as follows: walking
speedwithpostofficespeed: r= .48, p< .01; walking speed with bank clocks:
r= .25, p< .09; bank clocks with post office speed: r= .32, p< .04 (df =29in
all cases). Although it may be questioned whether correlations of this magni-
tude warrant creation of an overall index, such an index appeared helpful for
exploring some of the hypotheses and is included in the analyses.
RANKS
The ranks and means for each country on each of the pace variables are
shown in Table 1. It may be seen that Japan and the Western European coun-
tries (more precisely, the non-ex-Soviet bloc countries of Western Europe)
hadthefastestoverallpace of life scores. Other than Japan,the 9 fastest coun-
tries were from Western Europe. Put another way, the 9 Western European
countries all scored among the fastest 11 countries. Switzerland was the fast-
est country, based on consistently high rankings on each individual measure:
The combined scores from Zurich and Bern ranked third in walking speed,
second in postal times, and first in clock accuracy. The middle of the list was
Levine, Norenzayan / PACE OF LIFE 189
dominatedbyex-Sovietbloc European countries; newly industrialized Asian
countries; and, squarely at the median, the United States. The slowest coun-
tries on the list were all nonindustrialized countries—from the Middle East,
Latin America, and Asia.
190 JOURNAL OF CROSS-CULTURAL PSYCHOLOGY
TABLE 1
Meansaand Ranks on Pace Measures by Country
Overall Clock
Pace Index Walking Speed Postal Speed Accuracy
Country MRank MRank MRank MRank
Switzerland –3.43 1 11.80 3 16.91 2 19.29 1
Ireland –3.02 2 11.13 1 17.49 3 51.42 11
Germany –3.00 3 12.01 5 13.46 1 43.00 8
Japan –2.68 4 12.11 7 18.61 4 35.00 6
Italy –2.13 5 12.75 10 23.00 12 24.17 2
England –2.09 6 12.00 4 20.78 9 53.72 13
Sweden –1.96 7 12.92 13 19.10 5 40.20 7
Austria –1.43 8 14.08 23 20.60 8 25.00 3
Netherlands –1.43 9 11.45 2 24.42 14 82.33 25
Hong Kong –1.39 10 13.10 14 20.10 6 54.83 14
France –1.36 11 12.34 8 27.84 18 49.00 10
Poland –1.32 12 12.90 12 25.83 15 43.00 8
Costa Rica –1.13 13 13.33 16 21.13 10 55.38 15
Taiwan –0.73 14 13.58 18 20.22 7 68.00 21
Singapore –0.65 15 14.75 25 22.42 11 32.00 4
United States –0.30 16 12.03 6 36.99 23 67.87 20
Canada –0.26 17 12.86 11 30.50 21 70.00 22
S. Korea –0.02 18 13.76 20 29.75 20 58.00 16
Hungary 0.01 19 13.75 19 28.45 19 64.17 18
Czech Republic 0.28 20 13.80 21 27.73 17 76.07 23
Greece 0.54 21 13.10 14 24.33 13 117.0 29
Kenya 0.78 22 12.58 9 42.50 30 77.14 24
China 1.03 23 14.26 24 39.63 25 51.82 12
Bulgaria 1.59 24 15.57 27 33.67 22 60.00 17
Romania 2.42 25 16.72 30 42.25 29 32.46 5
Jordan 2.44 26 15.79 28 39.92 27 66.16 19
Syria 3.26 27 15.95 29 40.02 28 94.52 27
El Salvador 3.63 28 14.04 22 25.88 16 210.0 31
Brazil 3.98 29 16.76 31 38.17 24 108.0 28
Indonesia 4.14 30 14.82 26 39.64 26 161.5 30
Mexico 4.23 31 13.56 17 70.00 31 92.31 26
a. Overall pace index means are the average of the z-scores for each measure. For the other
measures, smaller numbers represent faster walking speeds, faster postal times, and smaller
clock deviations (all in seconds).
RELATIONSHIP OF COMMUNITY
VARIABLES TO PACE OF LIFE
First-order correlations. Table 2 presents scores for each country on the
community variables. Table 3 presents first-order correlations between the
ninecommunityvariablesand each of the pace-of-life measures. Note, again,
that these correlation analyses treat each of the countries as a single subject.
Given the large number of predictor variables for a sample of 31 countries,
these results must be treated cautiously.
Given that unidirectional predictions were made concerning the relation-
ship of each of the community variables to pace of life, the significance of
first-order correlations was evaluated with one-tailed tests. Using this liberal
standard, it may be seen in Table 3 that, as predicted, every community char-
acteristic other than population size was significantly correlated, in the
expected direction, with the overall pace-of-life index. The highest correla-
tions were for the “antecedent” variables (see Hypotheses 1 through 4). The
highest correlations were for two economic predictors—per capita GDP
(r= .74) and PPP (r= .72). Looking at the pace measures individually, it may
be seen that the strongest predictors of walking speed were GDP (r= .61),
individualism/collectivism(r= –.60), andPPP (r= .59); the strongest predic-
tors for post office speed were GDP (r= .55) and PPP (r= .53); and the
strongest predictors of clock accuracy were climate (r= –.53), GDP (r= .48),
and PPP (r= .48). In summary, places with a faster pace of life were signifi-
cantly more likely to have colder climates, have healthier economies, and to
emphasize individualism.
The correlations for the variables reflecting well-being with the pace of
life were also all in the predicted direction but were mostly of lower mag-
nitude than those for the “predictor” variables. For overall pace of life, there
were significant positive correlations with CHD (r= .35), percentage
smokers (r= .52), and SWB (r= .59). Looking at the individual pace meas-
ures, the highest correlations for walking speed (r= .40) and postal speed
(r= .54) were with SWB; the highest correlation for clock deviations was
with percentage smokers (r= .40). In summary, faster places had higher rates
of death from CHD, higher smoking rates, and higher SWB.
To test whether differences in well-being were consequences of pace of
life or of economic factors, we conducted partial correlations between an
index of overall economic productivity (see below for a description of this
index) and the three well-being measures, controlling for pace of life. The
resulting correlations between the economic index and psychological well-
being (r= .35, df = 10) and CHD (r= .28, df = 19) were nonsignificant, indi-
cating that the significant correlations between pace of life and these two
Levine, Norenzayan / PACE OF LIFE 191
192
TABLE 2
Values of Pace of Life Index and Other Community Variables
Country Pace Index Climate GDP PPP Calories Collectivism Population CHD Smoking SWB
Switzerland –3.43 55.0 364.10 236.2 35.6 9 298.9 1.5 9 12.9
Ireland –3.02 56.0 125.80 118.5 37.8 5 921.0 2.4 7 —
Germany –3.00 57.0 235.60 209.8 34.4 8 644.9 2.2 7 —
Japan –2.68 66.0 314.50 210.9 29.6 4 27,530.0 0.4 9 –5.0
Italy –2.13 71.0 196.20 180.7 35.0 6 3,028.0 1.2 7 –3.4
England –2.09 48.0 179.70 177.5 31.5 9 9,168.0 2.9 7 2.3
Sweden –1.96 48.0 248.30 175.6 29.6 8 1,471.2 3.0 5 12.0
Austria –1.43 55.0 231.20 188.0 35.0 8 2,392.0 2.1 7 —
Netherlands –1.43 54.0 207.10 180.5 31.5 9 1,053.4 1.5 5 8.8
Hong Kong –1.39 77.0 178.60 216.7 — 4 5,762.0 0.5 5 —
France –1.36 59.0 223.60 194.4 34.7 7 8,589.0 0.9 5 –3.4
Poland –1.32 53.0 22.70 50.1 — 5 — 1.1 9 –9.1
Costa Rica –1.13 77.0 21.60 55.8 28.1 5 395.4 0.6 3 —
Taiwan –0.73 79.0 106.00 — 29.7 5 6,924.0 — 7 —
Singapore –0.65 87.0 193.10 204.7 32.0 5 2,743.0 0.9 7 1.5
United States –0.30 62.0 247.50 247.5 36.7 10 14,628.0 2.1 9 6.4
Canada –0.26 54.0 206.70 204.1 34.8 9 3,182.0 1.7 9 10.4
South Korea –0.02 61.0 76.70 98.1 28.5 3 17,334.0 0.1 7 –1.7
193
Hungary 0.01 60.0 33.30 62.6 36.4 6 2,304.0 2.7 9 –3.4
Czech Republic 0.28 54.0 27.30 77.0 36.3 6 — 3.2 7 —
Greece 0.54 71.0 73.90 83.6 38.3 7 — 1.2 9 –7.4
Kenya 0.78 74.0 2.70 13.1 21.6 3 1,162.2 — 1 —
China 1.03 79.0 4.90 21.2 26.4 3 3,314.0 0.3 3 —
Bulgaria 1.59 60.0 11.60 37.3 — 4 1,221.0 2.3 7 —
Romania 2.42 62.0 11.20 29.1 31.6 4 2,175.0 1.8 5 —
Jordan 2.44 74.0 11.90 40.1 26.3 3 9,363.0 — 5 —
Syria 3.26 76.0 — — 30.0 3 1,444.3 — 5 —
El Salvador 3.63 90.0 13.20 23.6 23.2 3 497.6 0.2 3 —
Brazil 3.98 79.0 30.20 54.7 27.5 4 3,182.0 0.5 5 —
Indonesia 4.14 86.0 7.30 31.4 27.5 2 10,185.0 — 3 —
Mexico 4.23 72.0 37.50 71.0 30.5 5 21,615.0 0.2 3 –17.2
NOTE: The pace index represents the combined z-scores of the three pace measures; lower numbers signify a faster pace. Climate is the average annual maxi-
mumtemperature (in Fahrenheit). GDP is per capita grossdomestic product. PPPrefers to purchasing power.Calories refer to the averagecaloric intake.Collec-
tivismrefers to Triandis’s individualism-collectivismratings;higher numbers indicategreater individualism. Population isthe populationsize ofthe metropoli-
tan area for the city or the average when more than one city was measured (in hundreds). CHD refers to the rate of death from coronary heart disease. Smoking
refersto percapita cigarettesmoking, on a scale of 1 (low)to9(high). SWB refers to subjective well-being.See textfor furtherdescription of these variables.
well-being indicators could not be accounted for by the economic measures.
However, the resulting correlation between the economic index and cigarette
smoking (r= .68, df = 22) was significant (p< .001), indicating that the sig-
nificant correlation between pace of life and cigarette smoking could be
accounted for by the economic measures. It should be noted, however, that
because of missing data on different variables, these partial correlations were
not ideal tests of the question of the relative validity of the measures of pace
of life versus economic productivity for predicting well-being.
Although not a major focus of this study, we also examined the correla-
tions between walking speed and the community variables for men and
women separately; ttests indicated that the correlations of men versus those
194 JOURNAL OF CROSS-CULTURAL PSYCHOLOGY
TABLE 3
Correlationsa(and sample sizes)bBetween the Pace of Life
and Other Selected Community Characteristics
Pace Measure
Clock
Community Characteristic Overall Pace Walking Speed Postal Speed Deviation
Climate (temperature) –.58** –.47** –.30* –.53**
(31) (31) (31) (31)
Gross domestic
product per capita .74** .61** .55** .48**
(30) (30) (30) (30)
Purchasing power parity .72** .59** .53** .48**
(29) (29) (29) (29)
Daily caloric intake .51** .39* .35* .40*
(28) (28) (28) (28)
Collectivism –.59** –.60** –.39** –.36*
(31) (31) (31) (31)
Population size –.07 .15 –.31 .00
(28) (28) (28) (28)
Coronary heart disease .35* .19 .29 .28
(26) (26) (26) (26)
Percentage smokers .52** .29* .48** .40**
(31) (31) (31) (31)
Subjective well-being .59* .40 .54* .31
(15) (15) (15) (15)
a. Higher correlations indicate that the community characteristic is positively related to a faster
pace of life (faster walking speeds, faster postal speeds, and smaller clock deviations).
b. Note that statistics for some community characteristics were not available for some countries,
resulting in smaller sample sizes for those analyses.
*p< .05. **p< .01 (one-tailed significance test).
for women did not significantly differ for any of the nine community charac-
teristics. However, although the differences between the rs for men and for
women were not significant, for two community variables the correlations for
women were significant (p< .05), whereas those for men were not significant:
percentage smokers: r(women) = .36, r(men) = .20; SWB: r(women) = .50,
r(men) = .28.
Multiple regression analyses. To compare the validity of each of the “pre-
dictor” variables, they were simultaneously entered in a series of multiple
regression analyses in which one of the four pace-of-life variables served as
the criterion variable. Given the large number of predictor variables for the
sample size, the three economic predictors were combined, after first con-
verting each to zscores, into a single economic index. This overall index was
significantly correlated with each of its components (GDP: r= .92; PPP: r=
.95; average calorie intake: r= .78, df = 24 [reflecting missing data for five
cities], p< .001 using one-tailed test in all cases). The intercorrelations
between the three economic measures were as follows: GDP with PPP: r=
.96,p< .01; GDP with caloric intake: r= .48, p< .01; PPPwith caloric intake:
r= .57, p< .01 (df = 24 in all cases). The overall alpha for the three-item eco-
nomic index was .73. The resulting alpha if any single item were deleted
would have been .07 (PPP), .12 (GDP), and .95 (caloric intake). Even though
including caloric intake reduced the overall alpha, it measures an important
aspect of economic well-being at the individual level and thus seemed war-
ranted for inclusion in the index.
As can be seen in Table 4, when the three predictor variables (climate,
economic index, and individualism-collectivism) were optimally com-
bined, the resulting uncorrected multiple Rs were generally high: R(overall
pace index) = .81 (p< .001); R(walking speed) = .63 (p< .01); R(post office
speed)=.57(p< .05); R(clock accuracy)=.67(p<.01);df =3,22 in all cases.
The adjusted R2values, which take into account the sample size and number
of predictors, were as follows: overall pace index = .60; walking speed = .32;
post office speed = .23; clock accuracy = .38. It may be seen that economic
well-being was the strongest single predictor of all four dependent measures.
The economic index accounted for significant variance for postal speed,
clock accuracy, and the overall pace index in these multiple regression analy-
ses. Individualism-collectivism also accounted for significant variance on
clock accuracy. Thus, the unique contributions of climate for explaining the
variance in each of the four pace-of-life measures, and the unique contribu-
tions of individualism-collectivism for three of the pace-of-life measures,
become nonsignificant after partialing out for the effects of economic well-
being. First-order intercorrelations between the three predictor variables
Levine, Norenzayan / PACE OF LIFE 195
supported the finding that they are highly interrelated: Economic well-being
was negatively correlated with climate (average temperature) (r= –.63) and
positively correlated with individualism (r= .82); individualism was nega-
tively correlated with climate (r= – .68); p< .001 in all cases.
DISCUSSION
COUNTRY RANKS
The fastest pace of life on the present measures was found in Japan and in
the countries of Western Europe. This strong overall trend for Western
Europe was particularly remarkable: Eight of the 9 overall fastest countries
were from this regional category, and the 9 countries in our sample falling
into this category all scored among the fastest 11 countries. (Greece, which
might be marginally classified as “Western,” was 21st.) Only Japan (4th) and
Hong Kong (10th) intruded on this Western European monopoly.
The United States, Canada, and four economic-growth countries of Asia
(Hong Kong, Taiwan, Singapore, and South Korea) dominated the middle of
the rankings; these six countries scored between 10th and 18th overall.
Poland (12th) and Costa Rica (13th) also fell into this middle group. Other
than Poland, the four remaining ex-Soviet countries on the list tended to fall
justbelowthis middle group: Hungary and the Czech Republic were 19th and
20th and Bulgaria and Romania were 24th and 25th, respectively. The
remainder of the slowest third of the list was composed of relatively nonin-
dustrializedcountries from Africa (Kenya,22nd),Asia (China, 23rd), the Mid-
dle East (Jordan and Syria, 26th and 27th), and Latin America (El Salvador,
Brazil, Indonesia, and Mexico, 28th through 31st, respectively).
196 JOURNAL OF CROSS-CULTURAL PSYCHOLOGY
TABLE 4
Results of the Multiple Regression Analyses of Pace of Life
on the Predictor Variables (standardized betas)
Overall Pace- Clock
Predictor Variable of-Life Index Walking Speed Postal Speed Accuracy
Climate .36 .17 –.10 .52*
Economic index –.79** –.37 –.66* –.70*
Individualism-collectivism .30 –.15 .21 .58
NOTE: N= 26 for all variables. Adjusted R2: overall pace index = .60; walking speed = .32; post
office speed = .23; clock accuracy =.38.
*p< .05. **p< .01.
Perhaps the most surprising rankings were the very average times of the
U.S. and Canadian cities—compared, at least, to the industrialized countries
of Western Europe. It might be noted that in a later replication, a second
experimenter conducted the same three measurements in New York City and
obtained virtually identical times (Norenzayan, 1994). It is interesting to
speculate whether New York’s modal overall pace reflects the fact that the
city moves at slower speeds than its popular stereotype would suggest or
whether the pace of life in workday Western Europe has simply accelerated.
Because two Western European countries—Italy and England—and the
United States were included in our 1980 study (Levine & Bartlett, 1980), it
was possible to partially answer this question by comparing these countries’
times in the present study with those from the early study. Compared to 1980,
New Yorkers were faster on walking speed (12.03 seconds compared with a
prorated 13.65 seconds) but slower on postal times (36.99 vs. 28.00 seconds)
and less accurate on clock time (67.87 vs. 42.0 seconds). Compared to 1980,
the Romans were faster on all three measures: 12.75 seconds in the current
study versus 14.14 (prorated) seconds in 1980, 23.0 seconds versus 52.13
seconds on postal clerk speed, and 24.17 seconds versus 130.20 seconds for
average clock deviations. Compared to 1980, Londoners were virtually iden-
tical on walking speed (12.00 seconds compared with a prorated 12.03 seconds)
but faster on postal clerk times (20.78 vs. 35.63 seconds) and more accurate
on clock time (53.72 vs. 72.00 seconds). In other words, the two Western
European cities were clearly faster overall in the current study compared to
1980, whereas the scores from New York showed mixed changes. These pre-
liminary data would suggest that the relatively slow ranks for New York City
reflect increased speed in Western European cities rather than slow speeds in
New York.
The overall regional and economic trends were striking. Given that, in
most cases, data were gathered by different experimenters in each of these
countries, it is unlikely that these trends were the result of experimenter
effects. Rather, it appears that at this moment in time, the fastest pace of life,
during main business hours, on our measures, is in Japan and Western
Europe. Consistent with the popular stereotype, the slowest speeds were in
the nonindustrialized Third World. The very slowest were in three countries
popularly associated with a relaxed pace of life: Brazil (where the stereotype
of “amanha” [literally, “tomorrow”] holds that, whenever it is conceivably
possible, people will put off the business of today until tomorrow); Indonesia
(where the hour on the clock is often addressed as “jam kerat” [“rubber
time”]); and, slowest of all, the archetypical land of a manana, Mexico.
Of course, there are many aspects of the pace of life that the present meas-
ures do not address. The three indicators of pace focus solely on the tempo
Levine, Norenzayan / PACE OF LIFE 197
during working hours. Cities and countries may, however, also differ in the
tempo of their “downtime” and, perhaps even more significantly, on the bal-
ance between these two times. It has been suggested, for example, that West-
ern Europeans are more skilled at slowing down at the end of the workday
than are their cohorts in, for example, the United States and Japan. (A pair of
New York Times articles captured this stereotype with the headline “Why La
Dolce Vita is Easy for Europeans....AsJapanese Work Even Harder to
Relax” [Riding, 1991; Sanger, 1991]). There is, in fact, some evidence for
this stereotype. Annual work hours, for example, are longer in Japan and the
United States than they are in Western Europe. One recent estimate indicates
that the average annual paid working hours are 2,159 in Japan, compared
with1,957intheUnited States, 1,646 in France, and 1,638 in the former West
Germany. Only 27% of the Japanese labor force works as little as a 5-day
week, compared with 85.1% in the United States and 91.7% in France (Japan
External Trade Organization, 1992). Western Europeans also take signifi-
cantly more vacation days than do workers in the United States, who take
more than those in Japan. Whereas U.S. workers typically take 2 weeks of
summer vacation time, every country in Europe has collective bargaining
agreements guaranteeing minimum paid vacations ranging from 4 to 5
12
weeks (Schor, 1991). In Japan, recent data indicate that workers take an aver-
age of 8.2 vacation days, although an average of 15.5 days are authorized
(Japan External Trade Organization, 1992). Perhaps an even more dramatic
indicator of the single-minded commitment to work in Japan is the fact that
the Labor Ministry has had to undertake a formal campaign to encourage
workerstotakemore vacationtime, using slogans such as “Totake a vacation
is proof of your competence” (Sanger, 1991).
Each of the three pace measures may also be criticized for other short-
comings.Ourmeasureofworkspeed, forexample,measuresthetimeittakes
to complete a transaction but does not take into account the waiting time
before the transaction occurs. The postal times in Rome, for example, were a
respectable 12th fastest in our sample, which is contrary to the popular
stereotype of inefficient Italian post offices. However, a recent Italian
“Reporton the Social Situation of the Country,”compiled by the authoritative
Censis foundation, indicates widespread consumer discontent with the wait-
ing lines in Italian post offices, a variable that our measure does not address.
Perhaps this combination of relatively rapid transactions and long lines is
partly explained by the fact that post offices are only open for about 5 hours
perdayinItaly—compared,forexample,with 11 hours in France (Gasparini,
1995).
198 JOURNAL OF CROSS-CULTURAL PSYCHOLOGY
Despite these shortcomings, however, the present data suggest a number
of generally consistent economic patterns and geographic trends in the gen-
eral pace of life—at least the pace of life in main downtown areas during the
workday. The present data do not warrant inferences about the pace of life of
any single city or country, just as generalizations about any entire sample do
not justify inferences about individual participants within that sample. How-
ever, taken as a whole, the present data suggest a number of interesting pat-
terns for the pace of life.
PREDICTORS OF THE PACE OF LIFE
The present study provided strong support for each of the three previously
untested hypotheses concerning the predictors (or correlates) of the pace of
life: Hotter cities were slower than cooler ones, places with more vital econo-
mies were faster, and individualistic cultures were faster than those empha-
sizing collectivism. Although each of these community characteristics was
significantly related to the overall pace of life, multiple regression analyses
indicated that the variance in overall pace was best accounted for by the eco-
nomic hypothesis.
It should be noted that the present measure of individualism-collectivism,
which consisted of an expert but subjective rating, was not ideal. It would be
helpful, in a future study, to test the relationship of this variable to pace of life
by developing systematic, objective indicators of individualism-collectivism
for the countries in our study.
A fourth hypothesis, that more populous places are faster, was not sup-
ported by the present results. As indicated earlier, however, the present study,
which focused only on large cities, provided a very limited test of this
hypothesis. It may be that there is a critical point beyond which population
size does not relate to the pace of life. Bornstein’s (1979) and Bornstein and
Bornstein’s (1976) earlier studies, which provided the strongest demonstra-
tion of the link between population size and pace, were based on cities with a
wide range of population sizes, beginning at under 29,000. Only 3 of the 21
cities in their two studies combined had populations of more than 1 million.
In the present study, 23 of the 28 cities for which population size data were
available had populations of more than 1 million. It might be interesting in a
future study to systematically select cities to test whether there is, in fact, a
particular range, or critical cutoff point, for which population size is an
important determinant of pace of life. The present study, however, does not
address population size differences across the broad spectrum of city sizes
Levine, Norenzayan / PACE OF LIFE 199
that has been tested in previous studies. As such, it does not adequately chal-
lenge the results of previous studies that strongly demonstrate a positive rela-
tionship between population size and pace of life.
WELL-BEING AND THE PACE OF LIFE
The present study also offered evidence that the pace of life is related to
the physical and psychological well-being of communities: Faster places
tended to have higher rates of death from CHD, smoking rates were higher in
faster places, and people in faster places tended to report somewhat greater
SWB. These relationships were not generally as strong as those between the
predictorvariablesand the pace measures. Theywere,however,all in the pre-
dicted directions.
The finding that people in faster places tended to have higher CHD death
rates but to also have higher SWB was particularly interesting. If a fast pace
of life creates the stress that leads to CHD, should not this same stress make
people less happy? The key to this seeming paradox may lie in the central
role that the pace of life plays in the broader web of community characteris-
tics in which these findings are embedded. Although the present data are
correlational, we suggest that some of the same variables that successfully
predict the pace of life are themselves the product of the pace of life that they
create. Perhaps the two best examples of this are economic vitality and
individualism-collectivism. Economic needs are primary forces in creating a
sense of time urgency, and that sense of time urgency, in turn, leads to a pro-
ductive economy. Similarly, a focus on individualism thrives on a rapid pace
of life, which, in turn, creates pressure for further individualism. We argue
that these forces—economic vitality and individualism—have both positive
and negative consequences. On one hand, the focus on making every minute
count and being productive creates the stressors that lead to cigarette
smoking and CHD. On the other hand, they provide material comforts and a
general standard of living that enhance the quality of life. Productivity and
individualism—which in themselves are very difficult to separate from one
another—have double-edged consequences.
It is interesting to note that the mixed consequences of these community
variables have also been found on other measures. Diener (personal commu-
nication,March23,1995), for example,hasobservedthat although divorce is
much higher in individualistic nations, marital satisfaction is also often
higher—theUnited States being a primary case in point. His research has also
found that suicide and psychological well-being are both higher in individu-
alistic cultures than in collective ones.
200 JOURNAL OF CROSS-CULTURAL PSYCHOLOGY
CONCLUSIONS
On the whole, the present results strongly support the hypotheses. Taken
together, they suggest that the pace of life is central to a wide range of com-
munitycharacteristicsthatdefinethe personality of a city or culture. Thegoal
of the present study was to investigate clusters of regularities in the socioeco-
nomic patterns of cultures that may relate to the pace of life, in the hope of
leading to an eventual inclusive theory. Figure 1 summarizes the hypotheses
and major findings of the present study and, extrapolating from the present
correlational data, offers some speculations as a beginning toward building
such a theory.
We propose that the relationships between pace of life and other commu-
nity characteristics are frequently mutually reinforcing. Factors such as
population size, economic conditions, and cultural values may raise or lower
the characteristic pace of life. These norms may, in turn, directly alter the
same population and cultural characteristics that produced them. For
example, the economic opportunities in fast-paced places will tend to
attract potential migrants, resulting in even larger population size and den-
sity. Similarly, the fast-paced norms generated by individualism exert pres-
sure for even greater individualistic activity. Although not tested in this
study, it is also suggested that migration patterns will be selective: People
attracted to faster places may be the very ones who are likely to support a fast
pace of life, and vice versa.
This model assigns a key role to economic vitality, which emerged as the
strongest predictor in our study. Faster paced places will tend to be more eco-
nomically productive—which then raises the value of time and, subsequently,
the pace of life. The consequences of this economic vitality, as the present
findings suggest, are mixed. On one hand, the very time stressors that are
responsible for success lead to unhealthy behaviors (e.g., smoking) and
stress-related physical problems (notably CHD). On the other hand, the tan-
gible rewards of economic success raise the general level of psychological
well-being. People in fast-paced places tend to be happier.
One further consequence of the pace of life that warrants future explora-
tion is its relationship to social behavior. Cognitive processing theories pre-
dict that people who move quickly are less likely to find time for social
responsibilities, particularly when those responsibilities involve strangers.
Milgram’s(1970) system overloadtheory,in particular,argues that one of the
consequences of a rapid pace of life is inattention to the needs of strangers.
Milgram hypothesizes that people who are confronted with more sensory
inputs than they are able to process experience psychological overload. This
Levine, Norenzayan / PACE OF LIFE 201
202 JOURNAL OF CROSS-CULTURAL PSYCHOLOGY
Figure 1: A Community’s Pace of Life: Some Predictors and Consequences
is frequently the case in modern cities—the larger the city, the greater the
overload. One way that the overloaded urbanite adapts to this predicament is
by screening out nonessential stimuli. In essence, the city dweller focuses on
his or her goals and moves directly toward them as quickly as possible.
Because the needs of strangers are usually low on the urbanite’s hierarchy of
necessities, attention to these needs becomes a frequent casualty of the
screening process. Indeed, there is evidence that people who are pressed
for time are less likely to help a stranger than people who are not in a hurry
(Darley & Batson, 1973).
It would be interesting to test this model in a future study using causal
modeling analyses—to test which variables are, in fact, predictors and/or con-
sequences of the pace of life, and which variables are mutually reinforcing.
The small Nin the current study did not permit the application of causal mod-
eling to the present data.
This model speaks in generalizations. The course of any single culture
may be buffered or altered by many factors. What the model does attempt to
demonstrate is how broadly the pace of life is intertwined with other cultural
characteristics and how central it is to the personality of a place and its
people.
NOTE
1. For simplicity, Hong Kong is referred to here as a country. Technically, however—at the
time this study was conducted—it was a colony.
REFERENCES
Amato, P. R. (1983). The effects of urbanization on interpersonal behavior. Journal of Cross-
Cultural Psychology,14, 353-367.
Booth-Kewley, S., & Friedman, H. (1987). Psychological predictors of heart disease: A quanti-
tative review. Psychological Bulletin,101, 343-362.
Bornstein, M. H. (1979). The pace of life: Revisited. International Journal of Psychology,14,
83-90.
Bornstein, M. H., & Bornstein, H. (1976). The pace of life. Nature,259, 557-559.
Buggie, S. (1993). Time-urgency in transitional Poland. Polish Psychological Bulletin,24, 167-
180.
Buggie, S. (1994, March). Time urgency in the Arab Middle East. Paper presented at the annual
meeting of the Southeastern Psychological Association, New Orleans, LA.
Conway,H. M., & Liston, L.L. (1974).The weather handbook. Atlanta, GA: Conway Research.
Cutter, S. L. (1985). Ratings places: A geographic view on the quality of life. Washington, DC:
Association of American Geographers.
Levine, Norenzayan / PACE OF LIFE 203
Darley, J. M., & Batson, C. D. (1973). From Jerusalem to Jericho: A study of situational and dis-
positional variables in helping behavior. Journal of Personality & Social Psychology,27,
100-108.
Diener, E., Diener, M., & Diener, C. (1995). Factors predicting the subjective well-being of
nations. Journal of Personality & Social Psychology,69, 851-864.
Friedman,A. P.,& Rosenman,R. H. (1974). TypeA behavior and your heart.NewYork:Knopf.
Friedman, M., Fleischman, N., & Price, V. (1996). Diagnosis of Type A behavior pattern. In
R. Allan & S. Scheidt (Eds.), Heart & mind: The practice of cardiac psychology (pp. 179-
195). Washington, DC: American Psychological Association.
Friedman, M., & Ghandour, G. (1993). Medical diagnosis of Type A behavior. American Heart
Journal,126, 607-618.
Gasparini, G. (1995). On waiting. Time & Society,4, 29-45.
Gottdiener, M. (1994). The new urban sociology. New York: McGraw-Hill.
Greg, W. R. (1877). Literary and social judgments. London: Trubner.
Hoch, I. (1976). City size effects, trends and policies. Science,193, 856-863.
Hoel, L. A. (1968). Pedestrian travel rates in central business districts. Traffic Engineering. 38,
10-13.
Hofstede, G. (1980). Culture’s consequences. Beverly Hills, CA: Sage.
Japan External Trade Organization. (1992). Nippon 1992 business facts & figures. Tokyo:
Author.
Krupat, E., & Guild, W. (1980). Defining the city: The use of objective and subjective measures
for community description. Journal of Social Issues,36(3), 9-28.
Lauer,R.H. (1981). Temporalman: The meaning and uses ofsocial time.New York:Praeger.
Levine,R. (1988). The pace of life across cultures. In J. E. McGrath (Ed.), The social psychology
of time: New perspectives (pp. 39-62). Newbury Park, CA: Sage.
Levine, R., & Bartlett K. (1984). Pace of life, punctuality and coronary heart disease in six
countries. Journal of Cross-Cultural Psychology,15, 233-255.
Levine, R., Lynch, K., Miyake,K., & Lucia, M. (1990). The Type A city: Coronary heart disease
and the pace of life. Journal of Behavioral Medicine,12, 509-524.
Levine, R., Miyake, K., & Lee M. (1989). Places rated revisited: Psycho-social pathology in
metropolitan areas. Environment and Behavior,21, 531-553.
Levine, R., West, L., & Reis, H. (1980). Perceptions of time and punctuality in the United States
and Brazil. Journal of Personality & Social Psychology,38, 541-550.
Lowin,A., Hottes,J., Sandler,B., &Bornstein, M. (1971). The paceof life and sensitivity to time
inurban andrural settings: A preliminary study.Journal ofSocial Psychology,83,247-253.
Matthews,K. (1988).Coronary heart disease and TypeA behaviors: Update on andalternative to
the Booth-Kewley and Friedman (1987) quantitative review. Psychological Bulletin,104,
373-380.
Milgram, S. (1970). The experience of living in cities. Science,167, 1461-1468.
Nicolaides-Bouman, A., Wald, N., Forey, B., & Lee, P. (1993). International smoking statistics:
A collection of historical data from 22 economically developed countries. London: Wolfson
Institute of Preventitive Medicine.
Norenzayan, A. (1994). [The pace of life in New York City]. Unpublished raw data.
Riding, A. (1991, July 7). Why la dolce vita is easy for Europeans....The New York Times: The
Week in Review,p.2.
Rotton, J. (1985, August). Pedestrian movement in warm and cool settings: Three quasi-
experiments. Paper presented at the annual meeting of the American Psychological Associa-
tion, Los Angeles.
Sanger, D. (1991, July 7)....AsJapanese work ever harder to relax. The New York Times: The
Week in Review,p.2.
Schor, J. B. (1991). The overworked American: The unexpected decline of leisure. New York:
Basic Books.
Singelis, T., Triandis, H., Bhawuk, D., & Gelfand, M. (1995). Horizontal and vertical dimen-
sions of individualism and collectivism: A theoretical and measurement refinement. Cross-
Cultural Research,29, 240-275.
204 JOURNAL OF CROSS-CULTURAL PSYCHOLOGY
Triandis, H. (1995). Individualism & collectivism. Boulder, CO: Westview.
United Nations. Department of Economic and Social Development. (1992). Demographic
yearbook 1991. New York: Author.
U.S. Bureau of the Census. (1993). Statistical abstracts of the United States: 1993 (113th ed.).
Washington, DC: Author.
Veenhoven, R. (1993). Happiness in nations. Rotterdam, the Netherlands: Risbo.
Walmsley, D., & Lewis, G. (1989). The pace of pedestrian flows in cities. Environment and
Behavior,21, 123-150.
Werner, C. M., Altman, I., & Oxley, D. (1985). Temporal aspects of homes: A transactional
perspective. In I. Altman & C. M. Werner (Eds.), Home environments: Vol. 8. Human behav-
ior and environment: Advances in theory and research (pp. 1-32). New York: Plenum.
Wirtz, P., & Ries, G. (1992). The pace of life reanalysed: Why does walking speed of pedestrians
correlate with city size? Behaviour,123, 78-83.
World Bank. (1994). The World Bank atlas. Washington, DC: Author.
Wright, H. F. (1961). The city-town project: A study of children in communities differing in size.
Unpublished grant report.
Wright, J. W. (1993). The universal almanac 1994. New York: Andrews & McMeel.
RobertV.Levine is a professorof psychology at California State University, Fresno.He is
the author of A Geographyof Time(Basic Books, 1997). His main research interests are
in the areas of cross-cultural differences in the psychology of time, the psychology of
helping behavior and the psychology of persuasion.
Ara Norenzayan did his undergraduate work in psychology at California State
University, Fresno. He is currentlycompleting a doctoral degree in social psychology at
the University of Michigan. His current research interests include cutlural and historical
influences on thinking, and the social psychology of cultural transmission.
Levine, Norenzayan / PACE OF LIFE 205