The marketing-entrepreneurship paradox: A frequency-domain analysis
Abstract
The areas of overlap between the disciplines of marketing and entrepreneurship are substantial and they provide a wide variety of opportunities for multidisciplinary research. This paper lays out multidisciplinary foundations for the formal theoretical and practical treatment of the interaction between marketing, entrepreneurship and profitability in an organization. The focus of this research is on a company’s success as a function of organizational changes and the level of acceptable risk, measured by its profitability. The contribution to the literature on the relationship between entrepreneurship and marketing is reflected in a new approach that relies on the multi-scale (i.e., frequency-dependent) approach or the so-called “spiral of success”. In addition, this paper highlights the necessity for dynamic abilities and innovative character in an organization. More broadly, it explains an important theoretical paradox that organizations always face high risk, but, in order to survive in business, they need to enter new cycles of entrepreneurial activities (innovation and diversification) that involve even more risk. The novelty of this study lies in its application of the causality tests in the frequency domain for the bivariate system in order to demonstrate the marketing-entrepreneurship paradox. This is, to the authors’ best knowledge, the first paper that uses such a methodology in marketing and entrepreneurship. The paper’s principal hypothesis is tested on a well-diversified company (Amazon.com) where it is shown that marketing drives changes in net income at both medium and long horizons, but not vice-versa. The findings and related discussions can be useful to academics and practitioners, as well as to public policy-makers.
Full text
207
3, XX, 2017
Ma ke ing and T ade
DOI: 10.15240/ ul/001/2017-3-014
In oduc ion
The no ion o en ep eneu ship is no new, bu
en ep eneu ship is con inuously sea ching o
new ideas while inc easing hei applica ions
(Mo is & T o e , 1990; Mo is, Lewis, &
Sex on, 1994). Ma ke ing and en ep eneu ship
a e b oadening hei i eld o syne gic ac i i y,
bu some gaps in his in e ac ion s ill emain
(Bhuian, Menguc, & Bell, 2005). Nume ous
s udies ha e indica ed he link be ween
ma ke ing and en ep eneu ship (Mu ay,
1981; Mo is & Paul, 1987; He on, Sapienza,
& Smi h-Cook, 1992; Hills & LaFo ge, 1992;
Beche e & Mau e , 1998; Mo is, Schindehu e
& LaFo ge, 2002; K aus, Ha ms, & Fink, 2010;
Gilmo e, 2011, Hills & Hul man, 2011; Hul man
& Hills 2011; Ku gun, Bagi an, Oze en, &
Ma al, 2011; Mo ish, 2011; Buseni z, Plumme ,
Klo z, Shahzad, & Rhoads, 2014), bu
a numbe o esea ch ques ion ha e emained
unde explo ed.
This pape add esses he ela i e sca ci y
o schola ly wo k ha di ec ly es s he
causali y in he equency domain be ween
he en ep eneu ial ac i i ies and ma ke ing
in he unc ion o gaining and main aining
dis inc i e ma ke ad an age o a company.
Hence, he ocus o his esea ch is on
a company’s p o i abili y wi h special a en ion
on o ganiza ional changes and he measu e o
accep able isk in ma ke ing.
Nowadays business en i onmen is abundan
in unce ain y, ola ili y and comp omises, and
o such a eason isk eme ges as a majo
ac o in managemen (Al a ez & Ba ney, 2005;
C ame , Ha og, Jonke , & Van P aag, 2002),
whe eas en ep eneu ship e o s become basic
p e equisi es o success (Mo is, Schindehu e,
& LaFo ge, 2002). This implies ha in o de o
accep he changes and g ow en ep eneu ship
mus no only ha e dynamic abili ies (Menguc,
& Bell, 2005), bu also an inno a i e cha ac e
(S okes, 2000). Essen ially, en ep eneu ship
always includes isk ha may ange om
low o high le els (Mo is, Schindehu e, &
LaFo ge, 2002). This pape con ibu es o he
s udy o Da is, Mo is, and Allen (1991) on
he managemen capaci y o exposu e o isk
by aking in o accoun ha en ep eneu ial
ac i i ies en ail a h ea o e y expensi e ailu e.
Speci i cally, he con ibu ion o his pape lies
in combining ma ke ing wi h en ep eneu ial
ac i i ies in o de o manage o ganiza ional
changes aimed a assessing he measu e o
accep able isk and p o i abili y in he long un.
Schola ly con ibu ions o many au ho s (Mo is
& T o e , 1990; Mille , 1983; Mo is, Da is, &
Ewing, 1988; Knigh , 2000; Bhuian, Menguc,
& Bell, 2005; Naldi, No dq is , Sjöbe g, &
Wiklund, 2007) de i ne en ep eneu ship as
a p ocess whose main cha ac e is ics a e
inno a i eness, p oac i i y and abili y o ake
business isk, he eby making he link be ween
en ep eneu ship and ma ke ing e en s onge .
Ou esea ch mo i a ion is concen a ed on
examining he in eg al in e play be ween
hese wo disciplines. Ce ain pa icula i ies o
his phenomenon can be obse ed om he
company size, bu he analysis p esen ed in
his pape di e s om o he s udies because
i cla i i es he complemen a i y o he mu ual
in l uence o en ep eneu ial and ma ke ing
ac i i ies. Ou i ndings complemen and expand
he li e a u e on he ela ionship be ween
en ep eneu ship and ma ke ing by depic ing
ha syne gy h ough he mul i-scale (i.e.,
equency-dependen ) cycle o he so-called
“spi al o success”.
The main goal o his pape is o es and
analyze he causali y-d i en ela ionship be ween
en ep eneu ship and ma ke ing, i.e., he
ma ke ing-en ep eneu ship pa adox, wi h he
pa icula a en ion on he company’s ma ke ing
decisions in ega ds o p o i abili y. We ad oca e
THE MARKETING-ENTREPRENEURSHIP
PARADOX: A FREQUENCY-DOMAIN
ANALYSIS
Sla ka T. Nikolić, Nikola G adoje ić, Vladimi Đako ić,
Valen ina Mladeno ić, Jelena S anko ić
EM_3_2017.indd 207EM_3_2017.indd 207 7.9.2017 10:34:397.9.2017 10:34:39
208 2017, XX, 3
Ma ke ing a obchod
he iew ha ma ke ing and en ep eneu ship
ope a ions a e insepa able ac i i ies, om
which a wo d “ma ke p eneu ship” can be
de i ed. Based on he heo e ical concep s
and me hodology o causali y in he equency
domain (G adoje ic & Len o, 2015; G adoje ic
& Doba dzic, 2013) suppo ed by he case s udy,
we design ou p oposed esea ch amewo k
and d aw ou conclusions om he da a, as well
as om he g aphical ep esen a ion o he “spi al
o success”. Ou esul s show ha changes in
ma ke ing expense a e able o p edic changes
in ne income a bo h medium and long ho izons
(a business cycle equencies o 3-7 yea s).
In he con ex o ou “spi al o success”, we show
ha he companies may no only be success ul in
acing high isk when he o ganiza ional changes
we e in ense (in he sho o medium un), bu
also when he o ganiza ional changes we e no
in ense (in he long un).
We also poin o he pa adox in isky ma ke
condi ions, whe e he measu e o adequacy
o o ganiza ional esponse will de e mine he
measu e o business success. Mo e speci i cally,
we emphasize he exis ence o he ma ke ing-
en ep eneu ship pa adox caused by isk,
o ganiza ional changes and he le el o p oduc
di e si i ca ion. This esea ch is signi i can bo h
o he academic and p o essional communi ies
wi h pa icula impo ance o he op imal
business decision-making p ocesses, while
con ibu ing and expanding he cu en li e a u e
in he i eld o ma ke ing-en ep eneu ship.
To he au ho s’ bes knowledge, his pape is
he i s o pe o m he causali y es s in he
equency domain o he bi a ia e sys em ha
con ains changes in ne income and ma ke ing
expense.
In he nex sec ion, we p esen he
heo e ical concep s and in Sec ion 2 we speci y
ou p oposed esea ch amewo k. Sec ion
3 desc ibes ou esea ch me hodology o
wo king in he equency domain, while Sec ion
4 p esen s he da a cha ac e is ics. In Sec ion
5, we p esen he esul s and discuss hem in
he con ex o ou heo e ical conjec u es. The
i nal sec ion concludes he pape .
1. Theo e ical Backg ound
Conside ing he p oximi y o en ep eneu ship
and ma ke ing as scien i i c disciplines, i is
su p ising ha ela i ely li le “c oss- e iliza ion”
has occu ed (He on, Sapienza, & Smi h-
Cook, 1992, p. 6). Du ing he 1980’s, i was
sugges ed ha managemen explo e s should
be encou aged o make an e o o iden i y
whe he en ep eneu ship and ma ke ing
in e depend and in which way (Mo is & Paul,
1987). Almos a decade and a hal la e , a e
nume ous esea ches o many au ho s, some o
hem who we e op imis ic (e.g., Hul man & Hills,
2011) claimed ha he ela ionship be ween
en ep eneu ship and ma ke ing should s ill be
examined, while o he s (e.g., Miles, C ispin,
& Kasou , 2011) poin ed ou ha i is di i cul
o de i ne he en ep eneu ship-ma ke ing
link. The e exis signi i can a eas o o e lap
be ween he ma ke ing discipline and he s udy
o en ep eneu ship, and manage s should be
p epa ed o “bo ow boldly” om ma ke ing in
o de o ad ance he heo y o en ep eneu ship
(Hills & LaFo ge, 1992).
I is wo hwhile o no e ha he syne gy
be ween en ep eneu ship and ma ke ing is
so s ong ha many au ho s in eg a e hese
e ms in di e en ways, among which he one
used mos o en is en ep eneu ial ma ke ing
(K aus, Ha ms, & Fink, 2010; Lag osen &
S ensson, 2006; Hills & Hul man, 2011; Mo is,
Schindehu e, & LaFo ge, 2002; Ku gun,
Bagi an, Oze en, & Ma al, 2011; Mo ish,
2011; Gilmo e, 2011). This app oach, howe e ,
has a po en ial d awback in he ac ha he
majo i y o hese esea che s ha e conside ed
ha en ep eneu ship is ela ed o small o
middle-sized en e p ises (Knigh , 2000; Jones
& Rowley, 2011; Ku gun, Bagi an, Oze en,
& Ma al, 2011; Fai lie & Holle an, 2012;
An ónio & Mendes, 2013), while iden i ying he
en ep eneu ship wi h he en ep eneu , i.e.,
he owne o such an en e p ise. Ne e heless,
many pape s ha e in oduced he a gumen ha
en ep eneu ial o ien a ion can be embedded in
la ge companies as well as in small and middle-
sized companies (Mo is & Paul, 1987; Da is,
Mo is, & Allen, 1991; Miles & Da och, 2006;
Lag osen & S ensson, 2006; K aus, Ha ms,
& Fink, 2010). Some au ho s such as Mo is
& T o e (1990) and Sebo a & Thee apa ong
(2010) go e en u he by using he ph ase
co po a i e en ep eneu ship in he con ex o
he abili y o s imula e he a ibu es o small
en ep eneu ial companies in la ge and mo e
de eloped companies (Ku a ko, Ho nsby,
Na zige , & Mon agno, 1993), and, mo e
gene ally, unde s anding i as a comp ehensi e
e o o an o ganiza ion o sp ead in e nally
(Mo is, Lewis, & Sex on, 1994).
EM_3_2017.indd 208EM_3_2017.indd 208 7.9.2017 10:34:397.9.2017 10:34:39
209
3, XX, 2017
Ma ke ing and T ade
The p esen wo ld o consump ion is
cha ac e ized by he synd ome o ins an ism
and his phenomenon is go e ned by speed,
anxie y and ola ili y. This in u n ein o ces
he no ion o inno a i eness ha is necessa y
o ma ke en y (Nikolic, Tumbas, Kos es,
& S anko ic, 2013). We iew inno a i eness
om he company pe spec i e (Danneels &
Kleinschmid , 2001) as a c ea i e p ocess
o making new p oduc s, echniques o
echnologies (Knigh , 2000) and as a s a ing
poin o he o ganiza ional inno a i eness
(Amabile, Con i, Coon, Lazenby, & He on,
1996). Consequen ly, inno a i eness is he
key mechanism o he o ganiza ional g ow h
and i has a cen al signi i cance in he e o s o
achie ing compe i i eness (Lawson & Samson,
2001). Mo eo e , new ma ke ing ends o e an
oppo uni y o inc easing ma ke ing in l uence
on en e p ise compe i i eness (Pilik, 2008).
I is impo an o no e ha inno a ion
gene a es changes and g ow h; ha is why
en ep eneu ship in ol es a o m o “con inuous
c isis”. Ko le , Wong, Saunde s and A ms ong
(2004) a gued ha day- o-day ac i i ies o
economic agen s a e ma ked by unce ain y,
discon inui y, dis u bance and h ea . Th ough
o ganiza ional inno a i eness, en ep eneu ship
acks he equency o changes, i inc eases he
business isk, bu i also po en ially imp o es
he p o i abili y o an o ganiza ion (Mo is,
Schindehu e, & LaFo ge, 2002). Ma ke ing, on
he o he hand, by he means o communica ion
bo h in e nally and ex e nally in an o ganiza ion,
ep esen s he p ocess o wo king on di e ences
ha s eng hens he ma ke posi ion and
ep esen s a con ol mechanism o i nding he
ue measu e o inno a i eness. Hence, he
con empo aneous o ien a ion o a company
owa ds en ep eneu ship and ma ke ing
ep esen s a no el pa adigm ha guides
a company o i s op imal business pe o mance
(A uahene-Gima & Ko, 2001). The combina ion
o hese wo o ien a ions also ep esen s
a s a egic esponse o he u bulen en i onmen
ha companies equen ly ace (Mo is & Paul,
1987; Bhuian, Menguc, & Bell, 2005; A uahene-
Gima & Ko, 2001). The key ing edien o he
symbiosis be ween en ep eneu ship and
ma ke ing is inno a ion (Da is, Mo is, & Allen,
1991; Lag osen & S ensson, 2006).
I can be a gued ha ma ke ing is uniquely
capable o analyze he ( )e olu ion in he business
en i onmen and an icipa e new possibili ies
o en ep eneu ial ac i i ies (e.g., Pilik,
2008; Buseni z, Plumme , Klo z, Shahzad, &
Rhoads, 2014). Mo eo e , ma ke ing shel e s
he en ep eneu ial p ocess and p o ides i
wi h di ec ions o u u e business expansion
(Mu ay, 1981). Also, ma ke ing is in cha ge
o in e ac ion wi h he business en i onmen
(Da is, Mo is, & Allen, 1991). None heless,
one should no only ocus on he global iew o
ma ke ing, bu also on i s empo al dynamics,
by which ma ke ing pushes an o ganiza ion
along he “spi al o success”. Essen ially, i is
necessa y o unde s and he business cycle
pa e ns o he ac i i ies ha lead o business
success o ailu e (Newman, 2007). “The apid
pace o change means ha he i m’s abili y o
change will become a compe i i e ad an age”
(Ko le , Wong, Saunde s & A ms ong, 2004).
This means ha a company mus y o
ha monize i s en ep eneu ial o ien a ion wi h
he ma ke esponse (A uahene-Gima & Ko,
2001). I becomes an impe a i e o manage
a business sys em h ough he cycles o new
de elopmen al unde akings, by mo ing along
he “spi al o success” and a oiding he pa h o
o ganiza ional ailu e.
2. The Pa adox o Ma ke ing-
En ep eneu ship: “Spi al
o Success”
Ou model is based on he p emise ha being
awa e o he change is necessa y o business
de elopmen , whe e en ep eneu ship manages
he de elopmen , while ma ke ing manages
isk. In o he wo ds, ma ke ing de e mines he
le el o accep able isk which acili a es climbing
he spi al o o ganiza ional success. Ma ke ing
as an o ganiza ional unc ion is signi i can ly
a ec ed by he deg ee o en ep eneu ial
o ien a ion o a company (K aus, Ha ms, &
Fink, 2010). Howe e , he long- e m ou look o
he en ep eneu ial o ien a ion is de e mined
by he quali y o ma ke ing ac i i ies (Ko le ,
Wong, Saunde s, & A ms ong, 2004). Mo e
impo an ly, i should be s essed ha ma ke s
ha a e cons an ly changing a e no a h ea ,
bu ini ia o s o ma ke ing (Buseni z, Plumme ,
Klo z, Shahzad, & Rhoads, 2014). Such ma ke s
c ea e oppo uni ies ha a e no common o he
pe iods o ma ke s abili y.
The “spi al o success” o an o ganiza ion ha
shows a dynamic link be ween en ep eneu ship
and ma ke ing in ol es wo pa ame e s ha
de i ne i s empo al cha ac e is ics (Fig. 1).
EM_3_2017.indd 209EM_3_2017.indd 209 7.9.2017 10:34:397.9.2017 10:34:39
210 2017, XX, 3
Ma ke ing a obchod
The diame e o he spi al de i nes a measu e
o di e si i ca ion, while he s ep o he spi al
ep esen s a measu e o equency o ins ances
when an o ganiza ion en e s new de elopmen al
cycles. The “spi al o success” o a well-
de eloped o ganiza ion is cha ac e ized by
sho e s ep sizes and a wide diame e , which
can be in e p e ed as equen de elopmen al
cycles wi h highe le el o di e si i ca ion (i.e.,
new i elds o business ope a ions). To su i e in
a compe i i e en i onmen , high equency and
di e si y a e a necessi y.
In pa icula , an inc ease in changes, as he
basis o en ep eneu ial ac i i ies (En↗) leads o
an inc ease in isk (R↗) ha a business aces.
The inc ease in isk demands s ong ma ke ing
ac i i ies ha should esol e he p oblem
by wo king on di e ences (WoD), i.e., by
c ea ing a compe i i e ad an age ha causes
a dec ease in he isk (R↓) and inc ease in he
dynamic s abili y o he sys em. Bu , in he long-
un, he lack o o ganiza ional changes (Ch0)
may lead o an inc ease in isk (R↗) because i
may dec ease he compe i i e ad an age o he
o ganiza ion and he eby s eng hen i s ma ke
compe i ion. We emphasize he pa adox: when
he o ganiza ional changes a e in ense, bu also
when hey a e no , he o ganiza ion aces high
isk (Mo is, Schindehu e, & LaFo ge, 2002).
Such ad e se de elopmen s equi e en y in
a new cycle o en ep eneu ial ac i i ies by he
way o inno a ion ha no only inc eases isk
(R↗), bu also c ea es a ma ke ad an age om
ano he WoD cycle.
In gene al, ma ke ing and en ep eneu ship
a e di ec ly ela ed o he i m’s p o i abili y
(Beche e & Mau e , 1998). A success ul
de elopmen o new p oduc s and se ices is
able o sus ain g ow h and main ain p o i abili y
o heal hy companies in he long un (Ko le ,
Wong, Saunde s, & A ms ong, 2004). This
means ha ma ke ing pe o ms he ole o
a condui ha p o ides an en ep eneu ial
momen um (Mo is & Paul, 1987). Taken
join ly, en ep eneu ship and ma ke ing o e
a possibili y o de elop unique solu ions ailo ed
o mee he equi emen s o buye s in he so-
called ma ke p eneu ship-o ien ed businesses
(McAuley, 2011).
3. Me hodology: Causali y in he
F equency Domain
The es o causali y in he equency domain
by B ei ung and Candelon (2006) o igina es
Fig. 1: The “Spi al o Success”
Sou ce: own
EM_3_2017.indd 210EM_3_2017.indd 210 7.9.2017 10:34:397.9.2017 10:34:39
211
3, XX, 2017
Ma ke ing and T ade
om Geweke (1982) and Hosoya (1991). Le
z = [x , y ]′ be a wo-dimensional ime se ies
ec o wi h = 1,...,T. I is assumed ha z has
a i ni e-o de VAR ep esen a ion
() ,
Lz
(1)
whe e (L) = I – 1L – … – pLp is a 2×2 lag
polynomial wi h Lk z = z -k. I is assumed ha
he ec o is whi e noise wi h E( ) = 0 and
E( ′) = , whe e is a posi i e de i ni e ma ix.
Nex , le G be he lowe iangula ma ix o he
Cholesky decomposi ion G′G = Σ-1, such ha
E( ′) = I and = G . The sys em is assumed
o be s a iona y, implying he ollowing MA
ep esen a ion:
(2)
whe e 1
() ()LL
and 11
() ()
LLG
.
Using his ep esen a ion, he spec al densi y
o x can be exp essed as
22
11 12
1
() () ()
2
ii
x
ee
(3)
The measu e o causali y sugges ed by
Geweke (1982) and Hosoya (1991) is de i ned
as
2
11
2()
() log
()
x
yx i
M
e
(4)
2
12
2
11
()
log 1
()
i
i
e
e
(5)
This measu e is ze o i in
which case i is said ha y does no cause x a
equency . To es he hypo hesis ha y does
no cause x a equency he ollowing null
hypo hesis is used:
0)(
xy
M (6)
Yao and Hosoya (2000) es ima e
0)(
xy
M by eplacing and
om Equa ion (5) wi h es ima es
ob ained om he i ed VAR. Howe e , his
app oach is no app op ia e since
is a complica ed nonlinea unc ion o he VAR
pa ame e s. Yao and Hosoya (2000) p opose
a nume ical es ima ion p ocedu e ins ead
o an exac analy ical exp ession. B ei ung
and Candelon (2006) esol e his p oblem by
showing ha he null hypo hesis 0)(
xy
M
is equi alen o a linea es ic ion on he VAR
coe i cien s. Fi s , hey use 11
() ()LLG
and
22
12
12
()
() ()
gL
LL
(whe e 22
g is he
lowe diagonal elemen o 1
G and ()L is
he de e minan o ()L) o exp ess he null
hypo hesis as
(7)
whe e 12,k
is he (1,2)-elemen o k. Thus,
a necessa y and su i cien se o condi ions o
12 ()0
i
e
is
(8)
(9)
The no a ion can be simpli i ed by le ing
aj = θ11,j and βj = θ12,j. Then, he VAR equa ion
o x can be w i en as
...
11 p p xaxax
.... 111 p p yy
(10)
The hypo hesis 0)(
xy
M is equi alen
o he linea es ic ion
,0)(:
0
RH (11)
EM_3_2017.indd 211EM_3_2017.indd 211 7.9.2017 10:34:407.9.2017 10:34:40
212 2017, XX, 3
Ma ke ing a obchod
whe e 1,..., p
and
)sin(...)2sin()sin(
)cos(...)2cos()cos(
)(
p
p
R
The o dina y F s a is ic o (11) is
app oxima ely dis ibu ed as F(2,T – 2p) o
ω
(0,π). As in B ei ung and Candelon (2006),
o assess he s a is ical signi i cance o he
causal ela ionship be ween exchange a e
e u ns and o de l ows, he causali y measu e
o ω
(0,π) is compa ed o he 5% c i ical
alue o a X2-dis ibu ion wi h 2 deg ees o
eedom (5.99). B ei ung and Candelon (2006)
s udy he local powe o he es when he
equency being es ed con e ges o he ue
equency and show ha he Wald s a is ic is
asymp o ically dis ibu ed as non-cen al X2.
4. Da a Desc ip ion
The da a se is o Amazon.com (AMZN),
sampled a a qua e ly equency and was
ob ained om Bloombe g. The da e o he
IPO o AMZN was May 14, 1997. Thus,
he da a o ne income (NI) and ma ke ing
expense (MKT) span he pe iod be ween
June 30, 1997 and No embe 30, 2014, o
he o al o 70 obse a ions. The uni s o bo h
a iables a e millions o U.S. dolla s (USD). As
Fig. 2 indica es, he MKT a iable is po en ially
non-s a iona y and ending, while such da a
ea u es a e no ob ious o ne income ha
exhibi s occasional ola ili y ou bu s s.
Tab. 1 p esen s he summa y s a is ics
o MKT and NI. On a e age, NI is lowe han
MKT and his con i ms AMZN’s commi men o
la ge ma ke ing budge . S anda d de ia ions o
bo h a iables a e la ge han he means, hus,
indica ing highly ola ile and o e dispe sed
a iables. Wi h espec o skewness, NI is
oughly no mally dis ibu ed and MKT is
posi i ely skewed. This is in line wi h he
obse ed da a ea u es in Fig. 2. Finally,
bo h se ies display excess ku osis which
emphasizes la ge ail isk.
NI and MKT a e conside ed in i s -di e ence
e ms in o de o a oid heo e ical p oblems
o in e ence wi h non-s a iona y a iables. To
suppo his claim, s anda d Dickey-Fulle and
Phillips-Pe on uni oo s es s a e pe o med
on bo h a iables (Tab. 2). MKT is ound o be
in eg a ed o o de one and NI is bo de line
s a iona y a he 5% signi i cance le el. When
Fig. 2: Ne income and ma ke ing cos o Amazon.com [AMZN]
Sou ce: own
EM_3_2017.indd 212EM_3_2017.indd 212 7.9.2017 10:34:407.9.2017 10:34:40
213
3, XX, 2017
Ma ke ing and T ade
Va iable Obs Mean S d. De . Min Max Median Skewness Ku osis
NI 70 25.13 181.94 -545.14 416 40.50 -0.64 3.84
MKT 70 209.32 272.63 7.77 1,133 79.97 1.78 5.25
Sou ce: own
Va iable PP p- alue ADF p- alue Lags
NI 0.035 0.051 13
MKT 0.997 0.998 5
Change in NI 0.000 0.000 1
Change in MKT 0.000 0.006 9
Sou ce: own
Tab. 1: Summa y s a is ics
Tab. 2: Non-s a iona i y es s
Fig. 3: Causali y es s o ΔMKT and ΔNI o Amazon.com
Sou ce: own
EM_3_2017.indd 213EM_3_2017.indd 213 7.9.2017 10:34:407.9.2017 10:34:40
214 2017, XX, 3
Ma ke ing a obchod
he a iables we e ans o med o hei qua e ly
changes, he null hypo hesis o non-s a iona i y
is ejec ed a he 99% signi i cance le el.
5. Resul s
This sec ion epo s he esul s o causali y
es s in he equency domain o he bi a ia e
sys em ha con ains change in MKT (MKT)
and change in NI (NI). Bo h Dickey Fulle and
Phillips-Pe on es s ejec he null hypo hesis
o a uni oo in he MKT and NI ime se ies
a he 1% signi i cance le el. Acco ding o he
AIC c i e ion, a VAR(8) model was selec ed o
bo h sys ems.
Fig. 3 p esen s he causali y measu e
be ween MKT and NI o all equencies
(ω
(0,π)) along wi h he 5% c i ical alue
(5.99) ha is ep esen ed wi h a ho izon al
dashed line. The op panel indica es ha he null
hypo hesis o no causali y om NI o MKT
is ejec ed when <0.20 which co esponds
o e y low equencies wi h a wa eleng h o
oughly 7-8 yea s (31 qua e s = 2/ω). Hence,
i akes a long pe iod o ime be o e any changes
in NI impac changes in MKT. These i ndings
show o a ce ain ex en an independence o
ma ke ing decisions ela i e o he p o i abili y
o he company.
The bo om panel o Fig. 3 e eals e idence
o causali y om MKT o NI in he medium
un o ω
[1,1.7] (be ween 3.69 and 6.28
qua e s, i.e., he lag leng h is be ween one
and one and a hal yea s) and long un
ω
[0.2,0.5] (i.e., 3-7 yea s). These i ndings
demons a e ha changes in MKT a e able o
p edic changes in NI a bo h medium and long
ho izons. In o he wo ds, ma ke ing expense
is an impo an de e minan o p o i abili y wi h
a dual impac , one ha esul s in he sho o
medium un esponse o NI (1-1.5 yea -lag)
and he o he ha d i es p o i abili y a business
cycle equencies (3-7 yea s).
In he con ex o ou “spi al o success”
depic ed in Fig. 1, we show ha AMZN has no
only been success ul in acing high isk when
he o ganiza ional changes we e in ense (in
he sho o medium un), bu also when he
o ganiza ional changes we e no in ense (in
he long un). Speci i cally, AMZN was able o
en e new cycles o en ep eneu ial ac i i ies
and ma ke ing, and c ea e a dis inc i e ma ke
ad an age ha esul ed in hei p o i abili y.
Mo e b oadly, as he company has been
cons an ly inno a ing and di e si ying o e he
sample yea s, AMZN has alida ed ou p emise
ha a success ul company needs o en e mo e
equen de elopmen al cycles wi h a g ea e
le el o di e si i ca ion.
Conclusions
A la ge body o esea ch sugges s ha
en ep eneu ially-o ien ed businesses need
ma ke ing and ha en ep eneu ial ac i i ies
a e necessa y o ma ke ing-o ien ed
businesses. The in e ac ion be ween he wo
disciplines is hus a po en ially ich esea ch
a ea. The goal o his pape is o lay ou
mul idisciplina y ounda ions o he o mal
heo e ical and p ac ical ea men o his
impo an esea ch ques ion. On he heo e ical
side, we s ess ha he in e ac ion be ween
ma ke ing and en ep eneu ship leads o
a pa adoxical si ua ion: when o ganiza ional
changes a e in ense and also when hey a e
no , an o ganiza ion aces high isk. Such
de elopmen s necessi a e a new cycle o
en ep eneu ial (i.e., inno a ion) ac i i ies ha
will esul in e en mo e isk, bu hey will cause
a swi ch o a new de elopmen al cycle, un il he
inno a i eness o en ep eneu ial o ganiza ion
and/o i s ma ke ing c ea e a cycle o dis inc i e
ma ke ad an age. Wi h his a icle, we
emphasize he complex ela ionship be ween
ma ke ing, en ep eneu ship and i nance ha
is equency-dependen and ollows ce ain
pe iodici y. We de elop a heo e ical model and
es i s alidi y on ma ke da a.
On he empi ical side, we s udy he
in e ac ion o ma ke ing, p o i abili y and
en ep eneu ial ac i i ies in he equency
domain o he in o ma ion ob ained om he
i nancial s a emen s o Amazon.com, Inc. We
documen ha ma ke ing e o s a e able o
p edic changes in ne income a bo h medium
and long ho izons. In pa icula , Amazon.
com, Inc. was able o en e new cycles o
en ep eneu ial ac i i ies and ma ke ing c ea e
a dis inc i e ma ke ad an age ha esul ed in
hei p o i abili y.
Ne e heless, in he mos ecen yea s
(2014 and 2015), he p o i abili y o Amazon
has become mo e a iable. The unp eceden ed
nega i e ne income was eco ded in
2014 (-241 USD$mil), bu he company’s
inno a i e business model p e ailed in 2015
and gene a ed a s ong posi i e ne income
o 328 USD$mil. Meanwhile, he ma ke
showed a s ong suppo o he company
EM_3_2017.indd 214EM_3_2017.indd 214 7.9.2017 10:34:417.9.2017 10:34:41
215
3, XX, 2017
Ma ke ing and T ade
and, consequen ly, he s ock p ice ollowed an
upwa d end o e he yea s, un il he end o
2015. Ou model sugges s ha he obse ed
success o igina ed in he equency o new
de elopmen al cycles. Indeed, a a ie y o
Amazon’s inno a ions and new se ices such
as Fi e TV, Dash, T a el, Echo, Alexa and
P imeAi (d one deli e y) eme ged in 2014 and
2015. This ecen e idence and ou empi ical
i ndings suppo ou heo e ical conjec u e ha
a success ul company needs o en e mo e
equen de elopmen al cycles wi h a g ea e
le el o di e si i ca ion. Ou u u e esea ch
e o s will ocus on concep ualizing a dynamic
model o a panel o companies ha ha e been
success ully di e si ying hei business model.
The au ho s would like o hank he Edi o
and he wo anonymous e e ees o hei
help ul commen s and sugges ions.
Re e ences
Al a ez, S. A., & Ba ney, J. B. (2005).
How Do En ep eneu s O ganize Fi ms
Unde Condi ions o Unce ain y? Jou nal
o Managemen , 31(5), 776-793. doi:
10.1177/0149206305279486
Amabile, T. M., Con i, R., Coon, H.,
Lazenby, J., & He on, M. (1996). Assessing
he Wo k En i onmen o C ea i i y. Academy
o Managemen Jou nal, 39(5), 1154-1184.
doi:10.2307/256995.
A uahene-Gima, K., & Ko, A. (2001). An
Empi ical In es iga ion o he E ec o Ma ke
O ien a ion and En ep eneu ship O ien a ion
Alignmen on P oduc Inno a ion. O ganiza ion
science, 12(1), 54-74. doi:10.1287/
o sc.12.1.54.10121.
Beche e , R. C., & Mau e , J. G. (1998). The
Mode a ing E ec o En i onmen al Va iables on
he En ep eneu ial and Ma ke ing O ien a ion
o En ep eneu -led Fi ms. En ep eneu ship
Theo y and P ac ice, 22(1), 47-58.
Be lin, I. (1958). Two Concep s o Libe y. In
I. Be lin (Ed.), Fou Essays on Libe y (1969).
Ox o d: Ox o d Uni e si y P ess.
Bhuian, S. N., Menguc, B., & Bell, S. J. (2005).
Jus En ep eneu ial Enough: The Mode a ing
E ec o En ep eneu ship on he Rela ionship
Be ween Ma ke O ien a ion and Pe o mance.
Jou nal o Business Resea ch, 58(1), 9-17.
doi:10.1016/S0148-2963(03)00074-2.
B ei ung, J., & Candelon, B. (2006). Tes ing o
sho - and long- un causali y: A equency-domain
app oach. Jou nal o Econome ics, 132(2),
363-378. doi:10.1016/j.jeconom.2005.02.004.
Buseni z, L. W., Plumme , L. A., Klo z,
A. C., Shahzad, A., & Rhoads, K. (2014).
En ep eneu ship Resea ch (1985–2009)
and he Eme gence o Oppo uni ies.
En ep eneu ship Theo y and P ac ice, 38(5),
981-1000. doi:10.1111/e ap.12120.
C ame , J. S., Ha og, J., Jonke , N., &
Van P aag, C. M. (2002). Low Risk A e sion
Encou ages he Choice o En ep eneu ship:
An Empi ical Tes o a T uism. Jou nal o
Economic Beha io & O ganiza ion, 48(1), 29-
36. doi:10.1016/S0167-2681(01)00222-0.
Danneels, E., & Kleinschmid , E. J. (2001).
P oduc Inno a i eness om he Fi m’s
Pe spec i e: I s Dimensions and Thei Rela ion
wi h P ojec Selec ion and Pe o mance. Jou nal
o P oduc Inno a ion Managemen , 18(6),
357-373. doi:10.1016/S0737-6782(01)00109-6.
Da is, D., Mo is, M., & Allen, J. (1991).
Pe cei ed En i onmen al Tu bulence and
I s E ec on Selec ed En ep eneu ship,
Ma ke ing, and O ganiza ional Cha ac e is ics
in Indus ial Fi ms. Jou nal o he Academy
o Ma ke ing Science, 19(1), 43-51.
doi:10.1007/BF02723423.
Geweke, J. (1982). Measu emen o
linea dependence and eedback be ween
mul iple ime se ies. Jou nal o he Ame ican
S a is ical Associa ion, 77(378), 304-324.
doi:10.2307/2287238.
Gilmo e, A. (2011). En ep eneu ial
and SME ma ke ing. Jou nal o Resea ch
in Ma ke ing and En ep eneu ship, 13(2),
137-145. doi:10.1108/14715201111176426.
G adoje ic, N., & Len o, C. (2015).
Mul iscale analysis o o eign exchange
o de l ows and echnical ading p o i abili y.
Economic Modelling, 47, 156-165.
doi:10.1016/j.econmod.2015.02.028.
G adoje ic, N., & Doba dzic, E. (2013).
Causali y be ween egional s ock ma ke s:
A equency domain app oach. Panoeconomicus,
60(5), 633-647. doi:10.2298/PAN1305633G.
He on, L., Sapienza, H. J., & Smi h-Cook,
D. (1992). En ep eneu ship Theo y om
an In e disciplina y Pe spec i e: Volume II.
En ep eneu ship Theo y and P ac ice, 16(3), 5-11.
Hills, G. E., & Hul man, C. M. (2011).
Academic Roo s: The Pas and P esen o
En ep eneu ial Ma ke ing. Jou nal o Small
Business & En ep eneu ship, 24(1), 1-10.
doi:10.1080/08276331.2011.10593521.
EM_3_2017.indd 215EM_3_2017.indd 215 7.9.2017 10:34:417.9.2017 10:34:41