Irrational bunching? Tax regimes, brackets, and taxpayer behaviors

Zanoni, Wladimi ; Ca illo-Maldonado, Paul; Pan ano, Juan; Chuquima ca, Nicolás
Wo king Pape
I a ional bunching? Tax egimes, b acke s, and axpaye
beha io s
IDB Wo king Pape Se ies, No. IDB-WP-1600
P o ided in Coope a ion wi h:
In e -Ame ican De elopmen Bank (IDB), Washing on, DC
Sugges ed Ci a ion: Zanoni, Wladimi ; Ca illo-Maldonado, Paul; Pan ano, Juan; Chuquima ca, Nicolás
(2024) : I a ional bunching? Tax egimes, b acke s, and axpaye beha io s, IDB Wo king Pape
Se ies, No. IDB-WP-1600, In e -Ame ican De elopmen Bank (IDB), Washing on, DC,
h ps://doi.o g/10.18235/0013005
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/299493
S anda d-Nu zungsbedingungen:
Die Dokumen e au EconS o dü en zu eigenen wissenscha lichen
Zwecken und zum P i a geb auch gespeiche und kopie we den.
Sie dü en die Dokumen e nich ü ö en liche ode komme zielle
Zwecke e iel äl igen, ö en lich auss ellen, ö en lich zugänglich
machen, e eiben ode ande wei ig nu zen.
So e n die Ve asse die Dokumen e un e Open-Con en -Lizenzen
(insbesonde e CC-Lizenzen) zu Ve ügung ges ell haben soll en,
gel en abweichend on diesen Nu zungsbedingungen die in de do
genann en Lizenz gewäh en Nu zungs ech e.
Te ms o use:
Documen s in EconS o may be sa ed and copied o you pe sonal
and schola ly pu poses.
You a e no o copy documen s o public o comme cial pu poses, o
exhibi he documen s publicly, o make hem publicly a ailable on he
in e ne , o o dis ibu e o o he wise use he documen s in public.
I he documen s ha e been made a ailable unde an Open Con en
Licence (especially C ea i e Commons Licences), you may exe cise
u he usage igh s as speci ied in he indica ed licence.
h ps://c ea i ecommons.o g/licenses/by/3.0/igo/
I a ional Bunching? Tax Regimes, B acke s,
and
Taxpaye Beha io s
Wladimi Zanoni
Paul Ca illo-Maldonado
Juan Pan ano
Nicolás Chuquima ca
WORKING PAPER No IDB-WP-1600
In e -Ame ican De elopmen Bank
Coun y Depa men Andean G oup
May 2024
I a ional Bunching? Tax Regimes, B acke s,
and
Taxpaye Beha io s
Wladimi Zanoni
Paul Ca illo-Maldonado
Juan Pan ano
Nicolás Chuquima ca
In e -Ame ican De elopmen Bank
Coun y Depa men Andean G oup
May 2024
Ca aloging-in-Publica ion da a p o ided by he
In e -Ame ican De elopmen Bank
Felipe He e a Lib a y
I a ional bunching? ax egimes, b acke s, and axpaye beha io s / Wladimi
Zanoni, Paul Ca illo-Maldonado, Juan Pan ano, Nicolás Chuquima ca.
p. cm. — (IDB Wo king Pape Se ies ; 1600)
Includes bibliog aphical e e ences.
1. Income ax-Ecuado . 2. Sel -employed-Taxa ion-
Ecuado . 3. P og essi e
axa ion-Ecuado . I. Zanoni López, Wladimi , 1975-. II. Ca illo-
Maldonado, Paul.
III. Pan ano, Juan. IV. Chuquima ca, Nicolás. V. In e -
Ame ican De elopmen
Bank. Coun y Depa men Andean G oup. VI. Se ies.
IDB-WP-1600
h p://www.iadb.o g
Copy igh © 2024 In e -Ame ican De elopmen Bank ("IDB"). This wo k is subjec o a C ea i e
Commons license CC BY 3.0 IGO (h ps://c ea i ecommons.o g/licenses/by/3.0/igo/legalcode
). The
e ms and condi ions indica ed in he URL link mus be me and he espec i e ecogni ion mus be
g an ed o he IDB.
Fu he o sec ion 8 o he abo e license, any media ion ela ing o dispu es a ising unde such license
shall be conduc ed in acco dance wi h he WIPO Media ion Rules. Any dispu e ela ed o he use o he
wo ks o he IDB ha canno be se led amicably
shall be submi ed o a bi a ion pu suan o he
Uni ed Na ions Commission on In e na ional T ade Law (UNCITRAL) ules. The use o he IDB's name o
any pu pose o he han o a ibu ion, and he use o IDB's logo shall be subjec o a sepa a e w i en
license ag eemen be ween he IDB and he use and is no au ho ized as pa o his license.
No e ha he URL link includes e ms and condi ions ha a e an in eg al pa o his license.
The opinions exp essed in his wo k a e hose o he au ho s and do no necessa ily e lec he iews o
he In e -Ame ican De elopmen Bank, i s Boa d o Di ec o s, o he coun ies hey ep esen .
I a ional Bunching? Tax Regimes, B acke s, and
Taxpaye Beha io s *
Wladimi Zanoni†Paul Ca illo-Maldonado‡Juan Pan ano§
Nicolás Chuquima ca†
This e sion: Ap il 12, 2024
Abs ac
In his s udy, we examine he beha io o sel -employed axpaye s who “bunch" a
an income le el jus below a c i ical h eshold, which igge s a ansi ion om a sim-
ple ax egime o a mo e complex one. Unde he simple egime, indi iduals comple e
hei ax o ms independen ly, while he complex egime manda es he use o a pub-
lic accoun an o main aining accoun ing eco ds. U ilizing da a om he Ecuado ian
ax au ho i y om 2011 o 2014, we ini ially obse ed and documen ed he bunching
beha io p omp ed by he shi be ween egimes. Subsequen ly, we assess he impac
o his egime ansi ion on he amoun o axes paid by hose sel -employed axpay-
e s who choose o ill axes in he complex egime. Ou me hodology employs bo h
pa ame ic and semi-pa ame ic “donu ” es ima o s o e alua e hese e ec s. We ind
ha he egime shi indeed p omp s axpaye s o bunch below he income h eshold,
op ing o emain wi hin he simple egime. In e es ingly, hose who ansi ion in o
he complex egime end o pay less in axes. This pa e n holds ac oss a ious bunch-
ing windows and is consis en ac oss se e al es ima o s used. Ou esul s sugges ha
accoun an s a e he key mechanism behind he e ec s, o hey help axpaye s be e
na iga e ax deduc ions and bene i s, leading indi iduals o pay ze o axes.
Keywo ds: Tax Regimes, P og essi e Taxes, Pe sonal Income Tax, Bunching.
JEL Codes: H24, H26, D12
*We hank Osca Valencia, Luis Alejos, Alejand o Ras ele i, Osmel Manzano, and Sebas ian Gallegos o
hei eedback. Ou hanks a e also due o he pa icipan s o he esea ch wo kshop in Ecuado , acili a ed by
Julien Renaud, as well as he a endees o he IADB’s FMM and CAN esea ch wo kshops. We a e especially
hank ul o Jo ge Pa edes and Emily Díaz o hei in aluable esea ch assis ance.
†In e -Ame ican De elopmen Bank (IADB)
‡Uni e sidad de Las Ame icas (UDLA)
§Uni e si y o A izona

1. In oduc ion
In his pape , we in es iga e he beha io o sel -employed axpaye s in Ecuado , speci ically
as hei business income inc eases. This s udy’s conce n di e s om he adi ional ocus
on beha io al changes due o inc eases in ma ginal ax a es ac oss exis ing ax b acke s as
s udied in he ich li e a u e exempli ied by he wo ks in (Saez, 2001, 2010; Che y, F iedman,
Olsen, & Pis a e i, 2011; Kle en, 2016). Ins ead, we explo e he impac on sel -employed
axpaye beha io esul ing om ansi ions ac oss di e en ax egimes (i.e. di e en se s
o ules and p ocedu es designed o ensu e compliance wi h ax laws and obliga ions) as he
income ob ained om hei economic ac i i ies ises. In Ecuado , as in many coun ies, as
he business income, sales, o equi y o sel -employed indi iduals inc ease, he complexi y
o ax egimes hey a e subjec o, in e ms o compliance and o e sigh , ends o escala e.
Whene e he e a e h esholds ac oss hese dimensions ha igge egime changes, he e is
an oppo uni y o axpaye s o “bunch” in he neighbo hoods o hose h esholds o a oid
obliga ions induced by egime shi s. Ul ima ely, hei s a egic beha io calib a ing hei e -
o and/o epo ing would in luence he amoun o ax hey pay. In his pape , ou esea ch
ques ion is wo old. On he one hand, we seek o es he hypo hesis ha sel -employed in-
di iduals in Ecuado bunch when hey a e subjec o a change in ax egime change ( om an
easy o a complex- o- ile ax egime) igge ed by a h eshold in hei business income. On
he o he hand, we wan o e alua e he causal impac o being subjec o a mo e complex
egime on axes paid by sel -employed indi iduals.
Answe ing ou esea ch ques ion is ele an o a ious easons. P ima ily, i con ibu es o
he discou se on he design o op imal ax sys ems s a ed by Mi lees (1971). By desc ibing
he complexi ies o he beha io o axpaye s in esponse o he egula o y and en o cemen
amewo ks hey encoun e , ou esea ch o e s insigh s o ax au ho i ies in hei pu sui o
maximizing e enue and c a ing e ec i e ax sys ems as done in (Bes , B ockmeye , Kle en,
Spinnewijn, & Waseem, 2015). While he a ionale o bunching a lowe ax a es is con-
en ionally o minimize ax liabili ies, ou s udy es s whe he bunching a income h esh-
olds ha igge egime changes (i.e. changes in egula ions and en o cemen amewo ks)
migh be mo i a ed by di e en ac o s, such as ci cum en ing he inc eased obliga ions
and sc u iny o mo e complex egimes. This insigh is c ucial o ax au ho i ies c a ing
p og essi e ax sys ems. They aim o balance iscal e iciency and ai ness by implemen ing
ax b acke s and egula ions ha ensu e adhe ence o ax laws. Ye he e ec i eness o ax
egimes achie ing hose dual objec i es is b oadly unde - esea ched, pa icula ly in e ms
o unde s anding how egimes (again, di e en guidelines o ax compliance) in e ac wi h
ax b acke s o de ine p og essi e ax sys ems. Ou con ibu ion expands he unde s anding
o bunching esponses o dis inc ax incen i es, no limi ed o hose igge ed by changes
in ma ginal axes ac oss b acke s income as i is pu sued by (Ha ju, Ma ikka, & Rauha-
nen, 2019; Liu, Lockwood, Almunia, & Tam, 2021; Akcigi , Philippe, Lequien, G a oueille,
& S an che a, 2022).
To add ess ou esea ch ques ion, we build a case s udy om Ecuado . Be ween 2011 and
2014, Ecuado ian sel -employed axpaye s had o ansi ion by law om a simple o a com-
plex ax egime as hei business income su passed a speci ic h eshold (USD 100,000 in his
case). This ansi ion in ol ed a shi om a s aigh o wa d ax decla a ion p ocess o a
mo e elabo a e one, whe e he axpaye was also equi ed by law o keep de ailed accoun -
ing books, and a egis e ed accoun an had o sign he ax o ms when iling. We analyze
axpaye ilings and hei bunching nea his h eshold using adminis a i e eco ds om
1
he Ecuado ian ax au ho i y, employing es ablished me hodologies i s in oduced by Saez
(2001, 2010), and la e e ined and implemen ed by Che y e al. (2011); Kle en and Mazha
(2012); Bosch, Dekke , and S ohmaie (2020); Alosa (2023).
Ou analysis unco e s signi ican bunching a he income h eshold ha igge s a egime
change, e ealing a su p ising end: indi iduals unde he simple egime, whe e bunch-
ing is obse ed, ac ually pay mo e axes on a e age han hose unde he complex egime.
This appa en ly coun e -in ui i e beha io (appa en ly i a ional bunching) ha we disco -
e ed in he obse a ional da a mo i a ed us o explo e he causal in luence o he egime
change ( om simple o complex) on axes paid. We conduc ed his analysis using mul i-
ple e sions o he so-called donu es ima o de eloped by Dowd (2020). Ou wo k is indi-
ec ly linked wi h ha o Smi h and Mille (2021); Alosa (2023) in examining he esponses
o sel -employed axpaye s o dis inc ins i u ional amewo ks o incen i es on ax and
deduc ions dimensions. We expand he unde s anding o axpaye beha io beyond he
well-documen ed esponses o ax b acke changes Alosa (2023); Saez (2010). We also in-
c ease he knowledge on sel -employed axpaye beha io as in Boe i, Giupponi, K uege ,
and Machin (2020).
Ou wo k i s in o he li e a u e o epo ed income esponses o axa ion as in Ha ju e al.
(2019); Liu e al. (2021); Akcigi e al. (2022). The i s esea ch e o s conduc ed by Ha ju
e al. (2019); Liu e al. (2021) a e cen e ed on VAT egis a ion incen i es. Ha ju e al. (2019)
encoun e ha o small businesses, compliance cos s due o he VAT o ms explain mos
o he bunching beha io obse ed in he da a. Wo k by Liu e al. (2021) inds a nega i e
ela ion be ween bunching beha io and p oduc ma k-up, wi h he lowe ma k-up p oduc
i ms (i.e., he lowe di e ence be ween e ail p ice and uni cos ) inc easingly engaging in
bunching beha io .
Di e en ly om Ha ju e al. (2019); Liu e al. (2021) who, as we indica ed, e alua e changes
in axes paid ac oss b acke s, ou pape in es iga es he egime e ec on axes paid when
he e is an o e all change in he ax epo sys em, which includes new dimensions in e-
po ing and he hi ing o a licensed p o essional. Ou wo k also di e en ia es om ha in
Alosa (2023), whe e he in es iga e changes in axpaye beha io s de i ed om he subs i u-
ion o a p og essi e ax scheme based on p o i s by a la ax based on income. Ins ead, we
conside a egime change ha lea es he p og essi e scheme un ouched bu opens changes
in he egula o y and en o cemen amewo k and in oduces a new agen ( he accoun an ).
The axpaye who ansi ions egimes ecei es ained ad ice on key epo ing ma e s such
as deduc ions, income sou ces, and equi y a angemen s. Finally, ou wo k joins he discus-
sion on whe he di e en ia ed ax egimes c ea e ine iciencies and/o inc ease complexi y
ha ul ima ely cause e enue losses o go e nmen s (Adam & Mille , 2021). Ou s udy ex-
amines beha io s in esponse o ax egimes common o many ax sys ems, making ou ind-
ings applicable beyond he speci ic coun y and egime change analyzed he e. The insigh s
om ou pape a e ele an o ax policy and economic de elopmen in a ious economic
con ex s, ega dless o income le els.
The pape is s uc u ed as ollows: Sec ion 2 ou lines ou case s udy, ocusing on he an-
si ion o sel -employed indi iduals in Ecuado om a simple o a complex ax egime as
hei income inc eases. Sec ion 3 desc ibes he da a. Sec ion 4 documen s he occu ence
o “bunching” jus below he income h eshold a which he ax egime changes. Sec ion 5
e alua es he impac o a egime change on axes paid. Sec ion 6 desc ibes ideas o a simple
model ha desc ibes he axpaye beha io . The pape concludes wi h Sec ion 7, which syn-
2
hesizes ou indings and discusses hei b oade implica ions o ax policy and economic
beha io .
2. Ecuado ’s Sel -Employed Taxpaye s: A Case S udy
This sec ion p esen s he Ecuado ian case s udy documen ing how sel -employed axpaye s
in 2011-2014 had o ile hei income axes1. A ha ime, as o oday, sel -employed axpay-
e s we e equi ed o decla e all sou ces o income, encompassing hose om hei business
ac i i ies, wages, capi al gains, and o he sou ces.2Addi ionally, hey needed o epo hei
cos s and ela ed expenses o he ax au ho i y. Once he di e ence be ween income and ex-
penses was calcula ed, indi iduals could apply o deduc ions and exemp ions as ou lined
in he ax law (known in Spanish as Ley de Regimen T ibu a io In e no), esul ing in hei
axable income. Based on his income le el, axpaye s would be placed wi hin a p og essi e
ax b acke sys em, de e mining he axes hey would ul ima ely need o pay.
In ha ax sys em and ime (i.e. in he se o egula ions and en o cemen ha go along wi h
he ax b acke s schedule as income inc eases), di e ences in he ax egime a ec ing sel -
employed indi iduals we e igge ed by independen h esholds ac oss h ee dimensions:
1) he le el o hei business income; 2) hei business expenses, and; 3) he amoun o he
capi al associa ed o he business ac i i y. The income h eshold was se a USD 100,000,
o equi y a USD 60,000, and o business ac i i y expenses a USD 80,000. Indi iduals e-
po ing income, expenses, and capi al below hose h esholds would ile hei axes using a
simple ax o m named o m 102A ( he simple egime). I hey exceed any o hose h esh-
olds, hey would all in o a ax egime in which hey we e “obliga ed o keep accoun ing
books” (OKAB– he complex egime). In he OKAB egime, axpaye s we e equi ed o p o-
ide de ailed in o ma ion on asse s, liabili ies, and equi y h ough Fo m 102. This o m was
no only mo e complex han Fo m 102A, bu i manda ed axpaye s o main ain o mal ac-
coun ing eco ds and ob ain a ce i ied accoun an ’s signa u e when iling income axes. A
key ea u e o he OKAB egime is ha once indi iduals we e classi ied as OKAB axpaye s,
hey emained in ha egime in subsequen yea s, e en i hei income, expenses, and cap-
i al el below he h esholds in he u u e3. The compa ison o axpaye beha io s wi hin
he simple and he OKAB egimes p o ides a case s udy o designing ax sys ems aimed
a maximizing e enue collec ion and ensu ing ai ness. I is especially ele an o unde -
s anding beha io s in hose sys ems ha add egula ions o complemen he in as uc u e
o p og essi e ax b acke s.
No e ha sel -employed axpaye s in he OKAB egime we e equi ed o hi e SRI-ce i ied ac-
coun an s, enabling hem o seek assis ance in na iga ing he exemp ions and deduc ions
pe mi ed unde his egime. This equi emen c ea es incen i es o po en ial collusion:
accoun an s employed by axpaye s inc ease hei alue by mo e e ec i ely exploi ing he
deduc ions and exemp ions ou lined in his complex egime. Taxpaye s can also manip-
ula e hei income and expenses when hey s ay in he simple egime, bu his does no
equi e he legal expe ise o an accoun an . The deg ee o which a sel -employed axpaye
1O he s udies use Ecuado ian adminis a i e eco ds o explain he ax iling beha io o i ms. See, o
ins ance, Ca illo, Pome anz, and Singhal (2017) and Deza, Ca illo-Maldonado, and Ruiz-A anz (2021).
2The Ecuado ian ax au ho i y classi ies sel -employed indi iduals ac oss a ious occupa ions, b oadly
classi ied as p o essionals and en ep eneu s.
3Despi e a ely eques ed o g an ed, axpaye s could only e u n o he simple egime unde a pe sonal
eques a he SRI o ices, condi ional on jus i ying hey el below ac oss all he h esholds
3
can selec a ax egime is achie ed ei he by adjus ing e o up on o a oid su passing an
income/expenses/capi al h eshold o by managing epo s a e he ac —is de e mined
by a u ili y maximiza ion exe cise. This p ocess conside s indi idual p e e ences and con-
s ain s, wi h he isk o ax aud de ec ion playing a signi ican ole as a de e mining ac o .
Despi e he ac ha he h ee a iables ha igge egime changes (income, expenses, and
capi al) had h esholds ha p omp ed he OKAB egime, he mos p ominen and easily
en o ced c i e ion was business income. Figu e (1) shows a posi i e ela ionship be ween
cos s/expenses and income o axpaye s who epo be ween USD 80,000 o 120,000 in
g oss business income. The e o e, accoun ing o business income indi ec ly accoun s o
expenses in a posi i e manne . By su eying he income h eshold, we indi ec ly con ol he
cos h eshold. An essen ial poin o men ion is ha indi iduals who s ay unde he simple
egime do no epo equi y, which makes i di icul o adequa e de ec ion o occu . Con-
sequen ly, we use he business income h eshold o he pe iod 2011-2014 (USD 100,000) as
he salien on ie be ween he simple and OKAB egimes.
Figu e 1: Business Expenses s Income
Table (1) p esen s he axpaye s’ quan i y in he simple and OKAB egimes be ween 2011
and 2014 o dis inc g oss business income dis ibu ion anges.4. In e es ingly, we obse e
ha he numbe o axpaye s in he OKAB egime nea ly quad upled o he obse a ions
un es ic ed by any ange.
A e conside ing he da a only o he USD 80,000-120,000 ange, we obse e ha only
abou 10% o indi iduals emain unde he OKAB egime. This pe cen age dec eases o less
han 5% when we accoun o axpaye s unde he USD 90,000-110,000 ange. We buil ou
analy ic da abase wi h all he obse a ions in he USD 80,000-120,000 g oss business income
ange. F om now on, all he elemen s will be buil using he analy ic da abase.
4All includes he axpaye s who epo g oss business income g ea e han USD 0. USD 80,000-120,000
accoun s o axpaye s who epo g oss business income be ween USD 80,000 and 120,000. Finally, USD
90,000-110,000 accoun s o axpaye s only in ha in e al.
4
Figu e 2: Densi y Dis ibu ions o Tax Beha io Va iables by Regime, 2011–2014
(a) Non-Labo Income, USD (b) Wages, USD
(c) G oss Business Income1, USD (d) To al Income2, USD
(e) Deduc ions, USD ( ) Pe sonal Income Tax3, USD
No e: All he his og ams excep o panels (c) and ( ) use a wid h pa ame e o USD 2,500 o he his og am plo and a bandwid h pa ame e
o USD 2,000 o he densi y plo (See S a a kdensi y documen a ion o u he de ails).
1. The ed dashed line in he plo ep esen s he h eshold whe e axpaye s change egimes ( om he Simple o he OKAB). The wid h
pa ame e o he his og am plo is se o USD 500, and he densi y bandwid h is se a USD 2,000 (See S a a densi y documen a ion o
u he de ails).
2. To al Income = Non-Labo Income + Wages + G oss Business Income
3. The wid h pa ame e o he his og am plo is se o USD 250, and he densi y bandwid h is se a USD 100 (See S a a densi y documen-
a ion o u he de ails).
11

om ou side some a bi a y lowe and uppe bounds, o p edic he numbe o indi iduals
expec ed in each o hese USD 500 bins wi hin he cu en bunching window. Compa ing
hese coun e ac ual densi y p edic ions o he ac ual da a allows us o calcula e he mass o
indi iduals bunching on each side o he h eshold. We sys ema ically es a ious combina-
ions o lowe and uppe bounds and hen selec he models ha minimize he di e ence in
he abo e- e e enced mass di e ence be ween he igh and le sides o he h eshold (i.e.,
he excess mass c i e ia). The o mal de i a ion o his p ocedu e can be ound in Appendix
(A).
Table (4) p esen s he 10 models wi h he lowes excess mass and hei espec i e lowe and
uppe bounds o he bunching window. Fu he mo e, his able epo s he deg ee and he
oo mean squa ed e o (RMSE) o he unde lying local polynomial used in he coun e ac-
ual p edic ion.
Figu es (3) and (4) display bo h he ac ual and coun e ac ual densi y es ima es o indi id-
uals wi hin he G oss Business Income dis ibu ion, cen e ed a ound USD 100,000 o e e y
model in he able (4). The shaded egions in he igu es ep esen he mass o indi iduals
on each side o his h eshold. Ou o he 10 models, he bunching window is es ima ed o
s a as low as USD 92,000 and end as high as USD 109,500. The he e ogenei y in he lowe
and uppe bounds o he bunching window e lec s ha poin selec ion o he bunching
bounds is an a a he han a science. Condi ional on his limi a ion, he mo e conse a i e
app oach is o es wi hin a bunching window se , as we do in he p esen s udy.
Table 4: Bunching Window Se , Excess
Mass C i e ia
Model Excess Poly. Lowe Uppe RMSE
Ranking Mass1Deg ee Bound Bound
1 0.198 2 94,500 107,000 39.909
2 4.267 3 95,500 106,000 33.819
3 5.667 3 94,500 107,500 28.093
4 6.612 1 94,500 103,000 73.859
5 8.090 2 94,000 107,500 36.980
6 9.773 3 92,000 109,500 25.755
7 10.615 3 93,500 109,000 25.193
8 11.831 3 92,500 109,500 25.456
9 15.837 2 93,500 108,000 34.697
10 18.178 3 93,000 109,500 25.107
No e: The se o lowe and uppe bounds alues
es ed is zL∈{92000,92500,...,96000}, and zU∈
{102000,102500,...,112000}.
1 Excess mass epo ed in his able is he absolu e di e ence
be ween he obse ed and p edic ed equencies in he bins.
The igu es sugges ha axpaye s epo ed incomes jus below his h eshold o a oid an-
si ioning o he mo e complex OKAB egime. This clus e ing o epo ed incomes nea he
USD 100,000 ma k ep esen s a delibe a e esponse o he ax sys em’s s uc u e. This be-
ha io was consis en h oughou he 2011–2014 pe iod as shown in panels (11a) o (11d) in
Figu e (11) in he Appendix. Each yea exhibi s a simila income clus e ing jus below he
100,000 USD h eshold, indica ing a sus ained beha io al esponse a he han a ansien
phenomenon. The obus ness o he bunching phenomenon has also been e i ied ac oss a
ange o his og am bin sizes, spanning om USD 100 o USD 1,000. Comp ehensi e de ails
can be ound in Figu e (12) in he Appendix (B).
12
Figu e 3: Bunching Window Dis ibu ion, Models 1–5
(a) Model Ranking=1 (b) Model Ranking=2
(c) Model Ranking=3 (d) Model Ranking=4
(e) Model Ranking=5
13
Figu e 4: Bunching Window Dis ibu ion, Models 6–10
(a) Model Ranking=6 (b) Model Ranking=7
(c) Model Ranking=8 (d) Model Ranking=9
(e) Model Ranking=10
14
4.1 Do accoun an s bunch?
One way o quali y he beha io al d i e s o bunching is by examining he ax- ela ed beha -
io o accoun an s because, ou o all axpaye s, hey o m he g oup wi h he bes knowledge
abou how o na iga e ac oss he wo egimes (Simple and OKAB) o maximize u ili y. Find-
ing bunching among he accoun an s will sugges ha bunche s a e, in ac , paying less ax
in he Simple egime han in he OKAB egime because hei bunching beha io is heo e i-
cally no d i en by unce ain y abou he OKAB egime.
The his og am displayed in Figu e (5) shows he dis ibu ion o G oss Business Income among
accoun an s o e he 2011–2014 pe iod ( he e a e 384 obse a ions co esponding o 292 ac-
coun an s). Repo ed G oss Business Income d ops no iceably: he e is a clus e ing o ax-
paye s a ound he USD 100,000 h eshold depic ed by he ed dashed line, which sugges s
ha he e is bunching beha io a ha h eshold among accoun an s. To u he disen angle
he bunching e idence o accoun an s, we p esen a compa ison o ax a ibu es in Table
(5). We obse e ha al hough being in he OKAB egime, accoun an s, on a e age, epo
g oss business income below he eligibili y h eshold. This may occu when accoun an s
ansi ion egimes in yea and epo below he h eshold in yea +1. When a axpaye
ansi ions o he OKAB egime, he is expec ed o keep epo ing unde he new egime e en
i he does epo below he h eshold o subsequen yea 6. The deduc ions pa e n, how-
e e , emains unchanged. Accoun an s unde he OKAB egime deduc USD 12,326 mo e
on a e age in OKAB egime (USD 92,757 s. 80,431), a igu e below he USD 17,072 di e -
ence o deduc ions o all axpaye s (deduc ions di e ence in Table (3)). This indica es ha
accoun an s, e en unde he Simple egime, a e able o ind mo e deduc ions (a ound USD
2,000) compa ed o he g oup as a whole. This is a sugges ion ha accoun an s maximize
he deduc ion space independen ly o he ax egime.
Accoun an s epo ing unde he OKAB egime do pay mo e axes when compa ing axpay-
e s wi h axes paid g ea e han USD 0 (i.e. hose who ac ually pay USD 1,429 in he OKAB
egime s. USD 1,406 in he Simple egime). Howe e , when conside ing all obse a ions,
including hose who pay USD 0, he pa e n encoun e ed in Table (3) emains. Accoun an s
unde he OKAB egime pay less. The la e may be due o a g ea e sha e o accoun an s
who decide o engage in epo ing misconduc (8.1%), ha is, epo ing o e USD 100,000
wi hou a egime change. This does no emain unco e ed by he ax au ho i y, which o ces
hese axpaye s o epo unde he OKAB egime o he nex yea . Ano he way in which
he axes paid di e ence may be explained is o obse e he sha e o accoun an s ha e-
po below USD 100,000 bu ill he F102 o m. Tha sha e is mo e han 50%, p oposing ha
accoun an s, once hey ansi ioned, decide o epo below he h eshold and s ill bene i
om he g ea e complexi y and deduc ion space o e ed by he OKAB egime.
5. Wha is he impac o he OKAB egime on ax paid?
As we p e iously discussed in ela ion o Table (3), ou indings indica e ha , on a e age,
axpaye s posi ioned o he le o he h eshold and subjec o he Simple ax egime paid
mo e in axes compa ed o hose si ua ed o he igh o he h eshold. This inding appea s
6A axpaye can go back o he simple egime by illing a eques o m and mee ing below he h esholds
condi ions in he p e ious yea . The bu eauc a ic p ocedu e can be done online on he ax au ho i y web page
(Se icio de Ren as In e nas, SRI, 2024); howe e , in he 2011-2014 pe iod, his p ocess needed o be done in
he SRI o ices and was a ely g an ed.
15
Figu e 5: G oss Business Income among Accoun an s (2011–2014)
Table 5: Tax-Rela ed A ibu es by Tax Regime (2011–2014) o Accoun an s
(1) (2) (3)
Va iables Simple OKAB Di . (2-1) Obs.
G oss Business Income, USD 90,756 99,390 8,634*** 384
(7,719) (11,012) (1,042)
Deduc ions, USD 80,431 92,757 12,326*** 384
(16,346) (16,758) (1,996)
% Wi h Income Tax>0 0.686 0.466 -0.220*** 384
(0.465) (0.502) (0.058)
Taxes Paid, USD 965 666 -299* 384
(1,489) (1,283) (175)
Taxes Paid Among Those Who Pay, USD 1,406 1,429 23 244
(1,616) (1,569) (275)
% Wi h G oss Business Income (USD) < 100k 0.919 0.523 -0.396*** 384
(0.273) (0.502) (0.041)
% Wi h G oss Business Income (USD) > 100k10.081 0.477 0.396*** 384
(0.273) (0.502) (0.041)
Obse a ions 296 88 384 .
No e: The % o axpaye s by yea we e 24.48% in 2011, 18.49% in 2012, 28.65% in 2013, and 28.39% in 2014.
1 8.1% o axpaye s unde he Simple egime epo ed G oss Business Income abo e USD 100,000, all o whom ansi-
ioned om he Simple o he OKAB egime in subsequen pe iods a e a no ice om he ax au ho i y.
16

coun e in ui i e, gi en ha he exis ing li e a u e sugges s axpaye s end o clus e a he
lowe end o ax b acke s s a egically o minimize hei ax liabili y (Saez, 2001, 2010; Che y
e al., 2011; Ha ju e al., 2019; Akcigi e al., 2022). Howe e , no e ha i in o ma ion abou
he ax liabili ies unde he complex egime is unce ain, he s a egic beha io o in en ion-
ally “bunch” in he Simple egime may no be solely mo i a ed by a desi e o educe one’s
ax bu den wi hin his egime. A oiding isk migh be a d i e o such beha io .
An expe imen would be he gold s anda d o de e mine whe he he OKAB egime is causally
linked o he di e ences in axes paid be ween hose unde he OKAB and he Simple egimes.
This expe imen would en ail iden i ying axpaye s wi h simila p ede e mined cha ac e -
is ics and incomes a he eligibili y h eshold, ollowed by andom assignmen o some o
hem o he OKAB egime and o he s o he Simple egime. Subsequen ly, hei ax- ela ed
ou comes would be ca e ully measu ed and compa ed, and di e ences in hose ou comes
could be a ibu ed o he impac o he complex egime. Such an expe imen al app oach
would mi iga e po en ial con ounding a iables, enabling esea che s o isola e he causal
e ec o he OKAB egime on he obse ed paid ax dispa i ies wi h he Simple one. While
conduc ing a ull-scale expe imen was no easible in his con ex , adminis a i e eco ds
o e an oppo uni y o u ilize quasi-expe imen al me hods o es ima ing he impac in
which we a e in e es ed. Nex , we desc ibe he quasi-expe imen al app oaches and esul s
ob ained in ou a emp o answe he esea ch ques ion in he i le o his sec ion.
5.1 Implemen ing a “donu ” es ima o
Ou i s app oach o explo e he impac o he OKAB ax egime on axes paid compu es
an OLS eg ession whe e he amoun o axes paid is he ou come a iable. The indepen-
den a iable, which coe icien we seek o es ima e, is an indica o , which akes a alue o
one when he axpaye s ill ou axes unde he OKAB egime, and ze o i axes a e illed
in he Simple egime (we only ha e obse a ions o axpaye s in ei he o hose egimes).
In he absence o s a egic bunching beha io , he coe icien es ima e associa ed wi h ha
indica o a iable would cap u e he OKAB egime’s e ec on axes paid. Howe e , as p e i-
ously seen in Figu e (2c), axpaye s bunch, and consequen ly, we can expec hei s a egic
beha io o bias he es ima e o he OKAB ea men e ec when using such an OLS es i-
ma o . To a enua e he conce n om he ac ha axpaye s’ bunching can bias he e ec
o in e es , we implemen a e sion o he so-called “donu ” es ima o (Dowd, 2020). Ou
donu es ima o uses he same OLS eg ession amewo k jus explained bu excludes da a
on sel -selec ed indi iduals wi hin a bandwid h hole a ound he business income eligibili y
h eshold ha simula es he empi ical bunching window.
No e ha indi iduals who a e posi ioned nea he igh (le ) side o he USD 100,000 h esh-
old and ha e chosen he OKAB (Simple) egime likely made his choice because hey—
especially hose loca ed nea he h eshold—pe cei ed highe bene i s in ha pa icula
egime compa ed o he al e na i e one. One c ucial assump ion unde lying he DE es-
ima o is ha as we mo e u he away om he h eshold, axpaye s’ s a egic beha io
diminishes. This is because indi iduals u he om he h eshold a e mo e likely o epo
hei ac ual business income. The e o e, we can use da a om non-bunche s loca ed o
he le ( igh ) o he h eshold and ou side o he bandwid h hole o gain insigh s in o he
po en ial ou comes o axpaye s s a egically posi ioned on ei he side o he h eshold.
The implemen a ion o his es ima o is s aigh o wa d. Conside equa ion (1, whe e he
axes paid yi o axpaye ia e modelled as a unc ion o he indica o o selec ion in he
17
OKAB egime Di. The coe icien δ1in his equa ion p o ides he egime e ec es ima ion.
The model includes a co a ia es ec o Xin ended o hold cons an p ede e mined ac o s
ha explain he e ogenei y in he egime selec ion and axes paid. In he empi ical speci ica-
ion, he co a ia es we included we e age, sex, yea s in he ax sys em, ma i al s a us, wo k a
manu ac u e, ade o p o essional, p o ince o esidence, educa ion, and yea ixed-e ec s.
The “donu ” a ibu e o he es ima o comes om he ac ha axpaye s a ound he USD
100,000 h eshold a e symme ically excluded so ha he da a used o compu e he coe -
icien es ima es comes om axpaye s ou side o he se (bandwid h hole) de ined by he
lowe and uppe bounds zlkand zuko ha window as indica ed in equa ions (2) and (3).
yi=β0+δ1Di+Xβ+²i,∀z∈[80,000;zlk)∩(zuk;120,000] (1)
zlk=100,000−1,000k,k=0,1,2,...,10 (2)
zuk=100,000+1,000k,k=0,1,2,...,10 (3)
Figu e (6) shows a ious coe icien es ima es o he e ec s o he OKAB egime on ou ou -
comes using ha donu es ima o . We es ima ed he OKAB egime e ec s, no only on axes
paid, bu on (b) The amoun o axes paid among hose who pay some ax; (c) The p obabil-
i y o paying some axes s. ze o axes, and; (d) The p obabili y ha he axable income o
he indi idual is he lowes b acke (i.e. he 0% ax b acke ). As men ioned, he es ima es
di e along he ho izon al axis in ha we inc emen ally excluded axpaye s om wi hin
bandwid hs ha g ew symme ically a ound he business income h eshold in in e als o
USD 1,000 dolla s (hence o h called he bandwid h holes). Fo compa ison pu poses, we
included es ima es o he OKAB egime employing he donu es ima o , including he Xco-
a ia es ( he c oss-and-line plo s) and wi hou hose co a ia es (6 plo ).
The se o g aphs depic ed in Figu e (6) collec i ely analyze he e ec s o he OKAB ax
egime u ilizing he donu es ima o app oach, which, as we jus explained, conside s he
exclusion o da a poin s wi hin a ious bandwid h holes a ound a $100,000 income h esh-
old o co ec o possible biases due o axpaye s’ bunching beha io .
G aph (a) indica es ha he axes paid by indi iduals seem o inc ease ma ginally as he
bandwid h hole widens; ne e heless, he o e all e ec o he OKAB egime on ax paymen s
emains ela i ely s able, wi h no s a is ically signi ican a ia ions in he coe icien es i-
ma es ac oss di e en bandwid hs. This s abili y sugges s a consis en impac o he egime
ac oss di e en income anges ou side he bunching holes. In con as , G aph (c) sugges s
a sligh bu no s a is ically signi ican g ow h in he p obabili y o paying ze o axes as he
bandwid h expands. This end, while no subs an ial enough o indica e s a is ical signi i-
cance, hin s a a possible inc ease in ax a oidance o e asion s a egies as he income ap-
p oaches he h eshold. When i comes o axes paid among hose who pay a leas some-
hing (G aph b), he impac o he OKAB egime seems negligible ac oss he bandwid hs
explo ed, wi h almos no s a is ically signi ican e ec s obse ed. Finally, G aph (d) shows a
highe p obabili y o indi iduals unde he OKAB egime o all in o he ze o- ax b acke , a
inding ha when pai ed wi h he da a in G aph (c), sugges s a po en ial s a egy employed
by accoun an s o axpaye s o le e age he OKAB egime o educing ax liabili ies, possibly
by legally adjus ing epo ed incomes o quali y o he lowes ax b acke .
18
The di e ences be ween es ima es wi h and wi hou co a ia es sugges he deg ee o bias in-
duced by he selec ion o obse ables in he bunching p ocess. Mo e de ail on he eg ession
ou pu associa ed wi h he coe icien es ima es om Figu e (6) is p esen ed in Appendix Ta-
ble (12).
Figu e 6: Donu es ima o - Fi s App oach
(a) Taxes Paid1, USD (b) Taxes Paid Among Those Who Pay2, USD
(c) P obabili y o Paying Some Taxes s. Ze o
Taxes (d) P obabili y o being in he ax b acke 00%
No es: The igu es show he es ima ed coe icien s o he OKAB egime in models wi h and wi hou co a ia es o he di e en bandwid h
holes a ound he h eshold. Fo axes paid, igu e (c) in he absence o exclusions (Bandwid h = 0), he e ec o he egime is app oxi-
ma ely USD -270 o he model ha includes co a ia es and USD 350 o he one ha does no include hem. Remembe ha his i s
app oxima ion assumes a symme ical bandwid h. Tha is, he USD 5,000 bandwid h excludes all obse a ions be ween USD 95,000 and
105,000. The whiske s o he coe icien s show he 95% con idence in e al.
1 Taxes Paid e e s o ax paid by sel -employed indi iduals, including hose who paid USD 0.
2 Taxes Paid Among Those Who Pay comp ises he dolla amoun o axes paid among sel -employed indi iduals who pay mo e han USD
5 in pe sonal income ax.
The donu es ima o applied in ou analysis inhe en ly assumes ha axpaye s exhibi sym-
me ical beha io a ound he income eligibili y h eshold o he OKAB egime. Howe e , as
e idenced in Table (4), he ac ual bunching beha io begins a app oxima ely USD 92,000
and ex ends o USD 109,500, which de ia es om a symme ical dis ibu ion acco ding o
he excess mass c i e ia. Fu he mo e, his me hod sys ema ically omi s da a wi hin he
’donu hole’ — he designa ed bandwid h a ound he h eshold — p esupposing ha he
obse able cha ac e is ics on ei he side o he bandwid h a e homogeneous. This assump-
ion could lead o biased esul s i ce ain a iables, such as age, yea s in he ax sys em,
o he indus y o sel -employed indi iduals, di e signi ican ly be ween hose in he Sim-
ple and OKAB egimes. Addi ionally, he es ima o ’s linea na u e may no accu a ely e lec
19
he complex, po en ially non-linea ela ionships in he da a, which could ei he mi iga e
o exagge a e he pe cei ed e ec s o he ax egime. In ligh o hese po en ial limi a ions,
we ha e explo ed al e na i e es ima ion echniques ha elax some o he ini ial model’s
assump ions in hopes o mi iga ing hese sho comings.
5.2 A semi-pa ame ic app oach o he “donu ” es ima o
In his sec ion, we in oduce a donu es ima o ha elaxes he symme ic bandwid h and
he linea da a-gene a ing p ocess assump ions o sepa a ely p edic coun e ac ual ou -
comes o he bunching beha io below and abo e he USD 100,000 h eshold. The o e all
idea is ha by answe ing how axpaye s would ha e beha ed in he absence o he OKAB
egime a each side o he h eshold, we can measu e hei di e ence and es ima e he
egime’s ea men e ec . A new se o es ima es was gene a ed wi h e e ence o all he
bunching windows iden i ied in Table (4), and he ou ou comes s udied7. A mo e de ailed
explana ion ollows.
We assume ha he da a-gene a ing p ocess d i ing he beha io s o axpaye s ou side o
he bunching window can be le e aged o es ima e he coun e ac ual axpaying beha io
exhibi ed by indi iduals wi hin he bunching window. Unde ha assump ion, he di e -
ences in coun e ac ual ou comes be ween axpaye s immedia ely o he igh and le o he
OKAB eligibili y h eshold can be conside ed an es ima e o he e ec o he OKAB egime.
The dispa i y in he p edic ed coun e ac ual ou comes o axpaye s on ei he side o he
h eshold p ecisely a he h eshold alue helps us iden i y he impac o he ax egime.
We calcula e he alue o he coun e ac ual di e ence in he ou comes o in e es in ou
s eps. Fi s , we es ima e each ou come a iable as a unc ion o co a ia es se (age, sex,
yea s in he ax sys em, ma i al s a us, wo k a manu ac u e, ade o p o essional, p o ince
o esidence, educa ion, and yea ixed-e ec s) o all axpaye s le ( igh ) o he bunch-
ing window (sepa a ely). Second, we i each model and ob ain he p edic ed esiduals– he
a iabili y o he ou come a iable no explained by hose co a ia es– le and igh o he
bunching window. Thi d, we es ima e a model o hose p edic ed esiduals le ( igh ) as
a unc ion o g oss business income o eco e he coe icien es ima e o business income
on he esidualized ou come. Finally, we make coun e ac ual poin es ima es a he USD
100,000 h eshold bo h o he le and igh and compu e he di e ence in he coun e ac ual
p edic ions ( igh -le ): his is ou es ima e o he egime’s impac on he ou come a iable.
Mo e o mally, conside he ou come a iable yi o axpaye i. Income is se up as zi, and X
is a ma ix o dimensions i×jdeno ing jco a ia es o i axpaye s. The i s s age es ima es
he ou come as a unc ion o co a ia es Xusing he in o ma ion o he le ( igh ) o he
bunching window; see equa ions (4) and (5). Le zL,zUdeno e he bunching window lowe
and uppe bounds.
yi,le =β0+Xβj+εi,le ,∀zi∈[80,000,zL) (4)
yi, igh =β0+Xβj+εi, igh ,∀zi∈(zU,120,000] (5)
7(a) The o al amoun o axes paid (including ze o axes); (b) The amoun o axes paid among hose who
pay some ax; (c) The p obabili y o paying some axes s. ze o axes; and (d)The p obabili y ha he axable
income o he indi idual is he lowes b acke (i.e. he 0% ax b acke )
20
Figu e 10: P obabili y o Being in he 0.00% Tax B acke
(a) Bunching Window 1 (b) Bunching Window 2
(c) Bunching Window 3 (d) Bunching Window 4
(e) Bunching Window 5 ( ) Bunching Window 6
(g) Bunching Window 7 (h) Bunching Window 8
(i) Bunching Window 9 (j) Bunching Window 10
No e: The sub- igu es include all he bunching windows es ima ed in Table 4. Gi en ha he ou come o being in a p ede ined ax b acke
is a bina y a iable, esidual es ima ion sca e poin s will end o clus e a ound ze o. The e o e, we decided o plo he obse ed da a as
a locally weigh ed eg ession using he lowess p ocedu e in S a a. The use o a sca e a ei he he le o igh o each bunching window
would esul in do accumula ion a ound ze o in each sub- igu e. The lowess p ocedu e co ec ly cap u es he mean p obabili y beha io
alongside he G oss Business Income dis ibu ion.
27

pe o mance in comple ing he ax o ms and ax au ho i y en o cemen capaci ies in hose
labo supply decisions? In a amewo k whe e, as in ou case, a ax au ho i y delinea es wo
ax egimes ha di e in hei complexi y and a e di ided by an income h eshold, labo sup-
ply choices o sel -employed axpaye s a e d i en by he incen i e o maximize hei u ili y
Kle en (2016); Alosa (2023); Bas ani and Selin (2014); Saez (2010). This u ili y is shaped by
he ne income e ained a e axes and he pe sonal cos o complying wi h ax obliga ions.
Taxpaye s wi h incomes below he h eshold can ile axes independen ly unde a ela i ely
s aigh o wa d sys em, whe eas hose abo e a e compelled o engage wi h a mo e complex
sys em ha manda es hi ing an accoun an .
In ou analysis, an uncon en ional elemen o unce ain y in he labo supply decision o
sel -employed indi iduals could a ise when hey delibe a e on join ly selec ing he OKAB
egime and in es ing in p o essional accoun ing se ices. This in es men is d i en by he
expec ed u ili y de i ed om he accoun an ’s e o s, which inhe en ly ca ies a deg ee o
unp edic abili y. Fo a gi en ee o he accoun an , axpaye s mus conjec u e on he po en-
ial e u ns om hese se ices, which will subsequen ly in o m hei choice o ax egime.
The highe he expec ed bene i , he mo e likely hey a e o op o a egime ha — hough
possibly mo e complex—p omises g ea e deduc ions o lowe ax liabili ies. Howe e , he
p ecise alua ion o hese se ices is ob usca ed by he in icacies and nuances o ax legisla-
ion, ende ing he decision-making p ocess o axpaye s especially subjec o unce ain y.
Ano he elemen o conside in he choice o labo supply o sel -employed indi iduals is
he s ic ness o ax en o cemen , o i can c i ically condi ion axpaye beha io ega ding
income decla a ion. Enhanced en o cemen mechanisms dissuade he epo ing o w ong
income igu es by inc easing he p obabili y o de ec ion and subsequen penaliza ion. This
isk o noncompliance in luences axpaye s’ decisions no only on hei income epo ing
bu also on he selec ion o hei ax egime. In ou se ing, he le el o en o cemen is no
only pa icula ly pi o al a income h esholds ha delinea e di e en ax b acke s bu also
nea he h eshold o he OKAB egime. Choices o accu a ely epo e o , labo supply,
and income a e unde sco ed by he po en ial o subs an ial impac s on ax liabili y esul ing
om he egime choices.
In essence, he a chi ec u e o he ax sys em—including i s b acke s and egimes—shapes
how he sel -employed epo hei income and alloca e hei labo . They adjus hei wo k
and epo ea nings s a egically o selec he mos a o able ax condi ions. A s aigh o -
wa d and p edic able ax sys em gene ally simpli ies hese decisions, leading o mo e con-
sis en epo ing o incomes a le els ha a e ax-e icien . Explo ing how unce ain y and
en o cemen in luence labo supply and ax egime choice wi hin he con ex o a labo sup-
ply model emains a complex ask. We del e in o his opic in a companion pape , which
complemen s he cu en analysis.
7. Conclusions
In his pape , we inqui ed abou how he beha io o sel -employed axpaye s in Ecuado
changes in esponse o ansi ions ac oss di e en ax egimes as hei business income in-
c eases, pa icula ly wi h espec o bunching a income h esholds o a oid inc eased ax
obliga ions, and wha is he causal impac o ansi ioning o a mo e complex ax egime on
he amoun o axes paid by hese indi iduals.
28
We explo ed he ela ionship be ween ax b acke s and ax egimes in he con ex o pe -
sonal income ax o sel -employed axpaye s in Ecuado . Ac oss he globe, ax au ho i ies
aim o make pe sonal income axes mo e p og essi e by implemen ing ax a es ixed wi hin
income b acke s ha inc ease p og essi ely along he income dis ibu ion. Taxpaye s o en
espond o hese ising ax a es by employing a s a egy known as “bunching.” Bunching
in ol es axpaye s clus e ing hei epo ed incomes a he uppe end o a lowe ax b acke
(Che y e al., 2011; Saez, 2001), a a ional esponse when hey possess comple e in o ma-
ion and aim o maximize hei wel a e unde gi en en o cemen and ansac ion cos con-
di ions.
He e we de elop he idea ha , in e ac ing wi h ax b acke s, he “ ax egime” is also an im-
po an ea u e o he ax sys em. Those ax egimes ep esen di e en le els o en o ce-
men , o e sigh , and egula o y obliga ions ha inc ease wi h income le els, e en hough
hey a e no de ined by le els o axable income. Tax egimes de ine he ules o engage-
men be ween axpaye s and go e nmen s and a e a c ucial componen o ax sys ems.
Ou esea ch ocused on i s documen ing whe he axpaye s engage in bunching beha io
a ound income h esholds ha igge ansi ions be ween ax egimes and hen examining
wha implica ions he change in egime had o ax collec ions in he p esence o ha kind
o s a egic beha io .
An essen ial aspec o ou esea ch cen e s on he di icul y axpaye s ace in assessing he
ne bene i o choosing a pa icula ax egime due o he complex na u e o ax o ms, de-
duc ions, and exemp ions. When ansac ion cos s associa ed wi h hese assessmen s a e
high, axpaye s may choose o bunch o add ess he unce ain y, e en i his esul s in highe
ax paymen s.
Ou s udy analyzes he beha io o a subse o sel -employed axpaye s in Ecuado be ween
2011 and 2014, a pe iod du ing which he ax code classi ied hese indi iduals in o di e en
ax egimes based on whe he hei income exceeded a USD 100,000 h eshold. Le e ag-
ing indi idual-le el longi udinal da a om he Ecuado ian ax au ho i y, we ound e idence
ha indi iduals o en decla ed incomes jus below his h eshold, indica i e o bunching
beha io .
Examining he di e ences in ax paid be ween hose who sel -selec ed in o he Simple egime
and hose in he OKAB (complex) egime a ound he USD 100,000 h eshold, we ind ha , on
a e age, indi iduals who op o he Simple egime pay highe axes han hose who choose
he complex one. This su p ising esul challenges he assump ion ha bunching always
leads o he paying o a lowe amoun o ax.
We in es iga e he mechanisms behind his phenomenon and disco e ha axpaye s who
bunch and op o he Simple ax egime may be encoun e ing highe unce ain y. Despi e
he complexi y o he o m hey mus ile, non-bunche s ha e mo e oppo uni ies o eques
exemp ions and deduc ions, hus paying a lowe amoun o ax. The a ailabili y o accoun-
an s eme ges as a key mechanism o educing ha unce ain y.
O e all, ou esea ch unco e s ha axpaye s may engage in bunching beha io o mi iga e
unce ain y, e en i i esul s in highe ax paymen s. This inding challenges con en ional
wisdom and emphasizes he impo ance o unde s anding he beha io al esponses o ax-
paye s o complex ax sys ems, ax egimes, and ansac ion cos s. I has b oade implica-
ions o ax policy design, en o cemen mechanisms, and he balance be ween e iciency
and equi y in ax sys ems. Ul ima ely, ou esea ch con ibu es o he ongoing discou se on
29
op imal ax heo y and o e s aluable insigh s o policymake s seeking o e ine ax sys ems
and enhance e enue collec ion.
30
Re e ences
Adam, S., & Mille , H. (2021). Taxing wo k and in es men ac oss legal o ms: Pa hways o
welldesigned axes (No. R184). IFS Repo .
Akcigi , U., Philippe, A., Lequien, M., G a oueille, M., & S an che a, S. (2022). Tax simplici y
o simplici y o e asion? e idence om sel -employmen axes in ance. Wo king
Pape .
Alosa, F. (2023). Es ima ing he elas ici y o u no e om bunching: P e e en ial ax egimes
o solo sel -employed in i aly. Wo king Pape (1).
Bas ani, S., & Selin, H. (2014). Bunching and non-bunching a kink poin s o he swedish ax
schedule. Jou nal o Public Economics,109, 36–49.
Bes , M. C., B ockmeye , A., Kle en, H. J., Spinnewijn, J., & Waseem, M. (2015). P oduc ion
e sus e enue e iciency wi h limi ed ax capaci y: heo y and e idence om pak-
is an. Jou nal o poli ical Economy,123(6), 1311–1355.
Boe i, T., Giupponi, G., K uege , A. B., & Machin, S. (2020). Solo sel -employmen and al e -
na i e wo k a angemen s: A c oss-coun y pe spec i e on he changing composi ion
o jobs. Jou nal o Economic Pe spec i es,34(1), 170–195.
Bosch, N., Dekke , V., & S ohmaie , K. (2020). A da a-d i en p ocedu e o de e mine he
bunching window: An applica ion o he ne he lands. In e na ional Tax and Public
Finance,27(1), 951–979.
Ca illo, P., Pome anz, D., & Singhal, M. (2017, Ap il). Dodging he axman: Fi m mis epo -
ing and limi s o ax en o cemen . Ame ican Economic Jou nal: Applied Economics,
9(2), 144-64. Re ie ed om h ps://www.aeaweb.o g/a icles?id=10.1257/
app.20140495 doi: 10.1257/app.20140495
Che y, R., F iedman, J., Olsen, T., & Pis a e i, L. (2011). Adjus men cos s, i m esponses,
and mic o s. mac o labo supply elas ici ies: E idence om danish ax eco ds. The
Qua e ly Jou nal o Economics,126(2), 749–804.
Deza, M. C., Ca illo-Maldonado, P., & Ruiz-A anz, M. (2021, dic.). Im-
posición e ec i a a las emp esas en ecuado . X-pedien es Económicos,4(9),
6–30. Re ie ed om h ps://ojs.supe cias.gob.ec/index.php/X-pedien es
_Economicos/a icle/ iew/47
Dowd, C. (2020). Donu s and dis an la e’s: De i a i e bounds o d ex apola ion. A ailable
a SSRN 3641913.
Ha ju, J., Ma ikka, T., & Rauhanen, T. (2019). Compliance cos s s. ax incen i es: Why do
en ep eneu s espond o size-based egula ions? Jou nal o Public Economics,173,
139–164.
Kle en, H. J. (2016). Bunching. Annual Re iew o Economics,8, 435–464.
Kle en, H. J., & Mazha , W. (2012). Using no ches o unco e op imiza ion ic ions and
s uc u al elas ici ies: Theo y and e idence om pakis an. The Qua e ly Jou nal o
Economics,128, 669–723.
Liu, L., Lockwood, B., Almunia, M., & Tam, E. H. (2021). Va no ches, olun a y egis a ion,
and bunching: Theo y and uk e idence. Re iew o Economics and S a is ics,103(1),
151–164.
Mi lees, J. A. (1971). An explo a ion in he heo y o op imum income axa ion. The Re iew
o Economic S udies,38(2), 175–208. doi: 10.2307/2296779
Saez, E. (2001). Using elas ici ies o de i e op imal income ax a es. The Re iew o Economic
S udies,68(1), 205–229. doi: 10.1111/1467-937X.00166
31
Saez, E. (2010). Do axpaye s bunch a kink poin s? Ame ican Economic Jou nal: Economic
Policy,2(3), 180–212.
Se icio de Ren as In e nas, SRI. (2024). Au o ización pa a lle a con abilidad. Re-
ie ed om h ps://www.gob.ec/s i/ ami es/au o izacion-deja -lle a
-con abilidad (Accessed: 2024-04-04)
Smi h, K., & Mille , H. (2021). Capi al axa ion and en ep eneu ship. Job Ma ke Pape .
32

Appendix
A. Bunching Window Se Es ima ion P ocedu e
Conside cjas he numbe o obse a ions (i.e., axpaye s) wi h an income le el (o bin) zj.
De ine zL∈{zL1,zL2,...,} and zU∈{zU1,zu2,...,} as he bunching window lowe and uppe
bounds.
1. Es ima e a local polynomial model o deg ee p∈{1,2,...,5} ollowing equa ion (9),
excluding he cu en bunching window in e al (zLn,zUn):
cj=
p
X
i=0
βizi
j+²i,∀zj∉(zLn,zUn) (9)
2. Fi he model o all income le els, including he ou -o -sample in e al (zLn,zUn), and
ob ain he coun e ac ual ˆ
cj equencies. See equa ion 10:
ˆ
cj=
p
X
i=0
ˆ
βjzp
j∀zj. (10)
3. Calcula e he excess mass below and abo e he h eshold (i.e., o he le and igh ) as
he di e ence in he obse ed numbe o obse a ions and he p edic ed alues.
ˆ
Mb,zLn =
z∗−1
X
j=zLn
(cj−ˆ
cj)
ˆ
Ma,zUn =
zUn
X
j=z∗
(cj−ˆ
cj)
Unde he cu en applica ion, ˆ
Mb,zLn is posi i e, because he bunching obse a ions
p oduce excess mass below z∗. On he o he hand, ˆ
Ma,zUn is nega i e because he
obse a ions ha commi in o bunching beha io lea e missing mass abo e z∗. Recall
ha o no ches, his is a common cha ac e is ic, bu i is uncommon o kinks.
4. Compu e he di e ence in he o al excess mass om he cu en bunching window
(zLn,zUn) as he di e ence be ween excess mass below and abo e. Recall ha ˆ
Ma,zUn
ca ies a nega i e sign since he e is a hole abo e he h eshold.
ˆ
MzLn,zUn =ˆ
Mb,zLn +ˆ
Ma,zUn (11)
5. Repea s eps 1–4 o all he combina ions o zLand zU.
The p ocedu e’s ou pu is a bunching window se in which all he bunching window combi-
na ions can be anked unde he excess mass c i e ia.
33
B. Figu es
Figu e 11: Densi y Dis ibu ion unde a USD 20,000 Bandwid h a ound he Regime Th esh-
old, 2011–2014
(a) 2011 (b) 2012
(c) 2013 (d) 2014
No e: The his og ams use a wid h pa ame e o USD 500 o he his og am plo and a bandwid h pa ame e o USD 2,000 o he densi y
plo (See S a a kdensi y documen a ion o u he de ails).
34
Figu e 12: Densi y Dis ibu ions unde a USD 20,000 Bandwid h a ound he Regime Th esh-
old by Bin Size, 2011–2014
(a) $100 bin size (b) $200 bin size
(c) $300 bin size (d) $400 bin size
(e) $500 bin size ( ) $600 bin size
(g) $700 bin size (h) $800 bin size
(i) $900 bin size (j) $1,000 bin size
No e: The yea s 2015 and 2016 a e no included.
35
C. Le and Righ eg essions o Ou come esiduals
Table 7: Impac o G oss Business Income on Tax Paid Uncondi ional Ne o co a ia es - Op imal Polynomial Models
Le and Righ o he Bunching Window
(1) (2) (3) (4) (5)
Bunching Window Bunching Window Bunching Window Bunching Window Bunching Window
94.5K-107.0K 95.5K-106.0K 94.5K-107.5K 94.5K-103.0K 94.0K-107.5K
Va iables Le Righ Righ Righ Le Righ Le Righ Le Righ
G oss Business Income USD 0.0105*** 0.0112 0.0107*** 0.00251 0.0105*** 0.0124* 0.0105*** 0.00614 0.00980*** 0.0124*
(0.00270) (0.00711) (0.00245) (0.00646) (0.00270) (0.00750) (0.00270) (0.00478) (0.00283) (0.00750)
Cons an -909.6*** -1270.1 -933.8*** -282.5 -909.6*** -1403.5* -909.6*** -681.3 -846.0*** -1403.5*
(233.6) (804.7) (213.5) (727.8) (233.6) (850.6) (233.6) (530.8) (244.4) (850.6)
RMSE 1,451.8538 1,154.3145 1,454.7634 1,175.6606 1,451.8538 1,153.1386 1,451.8538 1,182.0880 1,450.4657 1,153.1386
Obse a ions 16,781 2,551 17,710 2,030 16,781 2,551 16,781 2,551 16,336 1,790
(6) (7) (8) (9) (10)
Bunching Window Bunching Window Bunching Window Bunching Window Bunching Window
92.6K-109.5K 93.5K-109.0K 92.5K-109.5K 93.5K-108.0K 93.0K-109.5K
Va iables Le Righ Righ Righ Le Righ Le Righ Le Righ
G oss Business Income USD 0.0102*** 0.0153 0.00689** 0.0187** 0.00950*** 0.0153 0.00689** 0.0113 0.00834*** 0.0153
(0.00351) (0.0101) (0.00296) (0.00913) (0.00330) (0.0101) (0.00296) (0.00797) (0.00313) (0.0101)
Cons an -870.3*** -1758.5 -593.4** -2134.2** -814.4*** -1758.5 -593.4** -1287.8 -717.2*** -1758.5
(300.2) (1158.9) (255.4) (1043.8) (283.1) (1158.9) (255.4) (905.8) (268.9) (1158.9)
RMSE 1,461.7045 1,179.0522 1,443.9890 1,157.1717 1,456.3711 1,179.0522 1,443.9890 1,164.5854 1,450.3981 1,179.0522
Obse a ions 14,531 1,451 15,883 1,722 15,020 1,451 15,883 1,722 15,450 1,451
No es: Signi icance le els a e ∗(p< .10),∗∗(p< .05), ∗∗∗(p< .01).
36
Online Appendix
A. Balance ables a Bunching Windows Ma gins
A.1 Bunching Window 1, USD USD 94,500 – 107,000
Table 14: Socio-Demog aphic A ibu es by Tax Regime a USD 1,500 ma gins om
he Bunching Window
Bunching Window: USD 94,500 – 107,000
Ma gins: USD 93,000 – 94,500 and USD 107,000 – 108,500
(1) (2) (3)
Simple OKAB
lb:93.0K lb: 107.0K
Va iables up:94.5K up: 108.5K Di . (2-1) Obse a ions
Age 41.709 42.500 0.791 1,369
(9.690) (9.479) (0.868)
Yea s in he ax sys em19.018 10.870 1.852*** 1,369
(6.438) (6.669) (0.580)
Single 0.366 0.290 -0.077* 1,369
(0.482) (0.455) (0.043)
Ma ied 0.579 0.645 0.066 1,369
(0.494) (0.480) (0.044)
O he Ma i al S a us20.054 0.065 0.011 1,369
(0.227) (0.248) (0.021)
Female 0.361 0.391 0.031 1,369
(0.480) (0.490) (0.043)
Region: Sie a30.453 0.493 0.039 1,369
(0.498) (0.502) (0.045)
Region: Cos a40.496 0.471 -0.025 1,369
(0.500) (0.501) (0.045)
Region: Amazon and Galapagos50.051 0.036 -0.015 1,369
(0.220) (0.188) (0.020)
Sec o : Manu ac u e 0.004 0.000 -0.004 1,369
(0.064) (0.000) (0.005)
Sec o : T ade 0.041 0.022 -0.019 1,369
(0.197) (0.146) (0.017)
Sec o : P o essional 0.010 0.007 -0.003 1,369
(0.098) (0.085) (0.009)
Sec o : O he s 0.359 0.239 -0.120*** 1,369
(0.480) (0.428) (0.043)
G oss Business Income, USD 93,757 107,784 14,027*** 1,369
(431) (436) (39)
% Wi h Income Tax>0 0.766 0.471 -0.295*** 1,369
(0.424) (0.501) (0.039)
Taxes Paid6, USD 953 577 -376*** 1,369
(1,549) (1,327) (137)
Taxes Paid Among Those Who Pay7, USD 1,244 1,225 -19 1,008
(1,665) (1,721) (214)
% in Tax B acke 00% 0.229 0.529 0.300*** 1,369
(0.420) (0.501) (0.039)
Obse a ions 1,231 138 1,369 .
No e: Bunching Window 1 is de ined as he space be ween USD 94,500 and 107,000 in G oss Business Income. This able conside s all axpaye s who lie inside
he in e als [93,000;94,500) and (107,000;108,500] co esponding o a USD 1,500 ma gin below and abo e he bunching window lowe and uppe bounds. 1.
Taxpaye ’s yea s in he ax sys em, which is he di e ence be ween he ax o m yea and he axpaye ’s i s egis e ed yea . 2. O he Ma i al S a us include
di o ced, widowed, and ee union. 3. Sie a includes Ca chi, Imbabu a, Pichincha, Co opaxi, Tungu ahua, Bolí a , Chimbo azo, Caña , Azuay, and Loja
p o inces, all o which a e loca ed in he highland cen e o he coun y. 4. Cos a includes Esme aldas, Manabí, San o Domingo de los Tsáchilas, Los Ríos,
Guayas, San a Elena, and El O o p o inces. Cos a egion si es a e in he Paci ic coas side o he coun y, wes o he highlands. 5. Amazon and Galápagos
include Sucumbíos, Napo, O ellana, Pas aza, Mo ona San iago, Zamo a Chinchipe, and Galápagos p o inces, all o which, excep o Galápagos, an a chipelago
in he Paci ic Ocean, a e loca ed eas o he highlands. 6. Taxes Paid e e s o ax paid by sel -employed indi iduals, including hose who paid USD 0. 7. Taxes
Paid Among Those Who Pay comp ise he dolla amoun o axes paid among sel -employed indi iduals who pay mo e han USD 5 in pe sonal income ax.
43

A.2 Bunching Window 2, USD 95,500 – 106,000
Table 15: Socio-Demog aphic A ibu es by Tax Regime a USD 1,500 ma gins om
he Bunching Window
Bunching Window: USD 95,500 – 106,000
Ma gins: USD 94,000 – 95,500 and USD 106,000 – 107,500
(1) (2) (3)
Simple OKAB
lb: 94.0K lb: 106.0K
Va iables up: 95.5k up: 107.5K Di . (2-1) Obse a ions
Age 41.508 40.884 -0.624 1,414
(9.967) (8.475) (0.909)
Yea s in he ax sys em18.984 9.899 0.916 1,414
(6.628) (5.494) (0.603)
Single 0.343 0.349 0.006 1,414
(0.475) (0.478) (0.044)
Ma ied 0.596 0.566 -0.030 1,414
(0.491) (0.498) (0.045)
O he Ma i al S a us20.061 0.085 0.025 1,414
(0.239) (0.280) (0.022)
Female 0.328 0.271 -0.057 1,414
(0.470) (0.446) (0.043)
Region: Sie a30.455 0.535 0.080* 1,414
(0.498) (0.501) (0.046)
Region: Cos a40.485 0.419 -0.066 1,414
(0.500) (0.495) (0.046)
Region: Amazon and Galapagos50.060 0.047 -0.013 1,414
(0.237) (0.211) (0.022)
Sec o : Manu ac u e 0.002 0.000 -0.002 1,414
(0.039) (0.000) (0.003)
Sec o : T ade 0.042 0.031 -0.011 1,414
(0.201) (0.174) (0.018)
Sec o : P o essional 0.008 0.016 0.008 1,414
(0.088) (0.124) (0.008)
Sec o : O he s 0.346 0.287 -0.059 1,414
(0.476) (0.454) (0.044)
G oss Business Income, USD 94,751 106,783 12,033*** 1,414
(427) (452) (40)
% Wi h Income Tax>0 0.739 0.473 -0.266*** 1,414
(0.439) (0.501) (0.041)
Taxes Paid6, USD 950 780 -170 1,414
(1,585) (1,541) (146)
Taxes Paid Among Those Who Pay7, USD 1,285 1,650 365 1,011
(1,723) (1,900) (229)
% in Tax B acke 00% 0.257 0.527 0.270*** 1,414
(0.437) (0.501) (0.041)
Obse a ions 1,285 129 1,414 .
No e: Bunching Window 2 is de ined as he space be ween USD 95,500 and 106,000 in G oss Business Income. This able
conside s all axpaye s who lie inside he in e als [94,000;95,500) and (106,000;107,500] co esponding o a USD 1,500
ma gin below and abo e he bunching window lowe and uppe bounds. 1. Taxpaye ’s yea s in he ax sys em, which is
he di e ence be ween he ax o m yea and he axpaye ’s i s egis e ed yea . 2. O he Ma i al S a us include di o ced,
widowed, and ee union. 3. Sie a includes Ca chi, Imbabu a, Pichincha, Co opaxi, Tungu ahua, Bolí a , Chimbo azo,
Caña , Azuay, and Loja p o inces, all o which a e loca ed in he highland cen e o he coun y. 4. Cos a includes Esme -
aldas, Manabí, San o Domingo de los Tsáchilas, Los Ríos, Guayas, San a Elena, and El O o p o inces. Cos a egion si es
a e in he Paci ic coas side o he coun y, wes o he highlands. 5. Amazon and Galápagos include Sucumbíos, Napo,
O ellana, Pas aza, Mo ona San iago, Zamo a Chinchipe, and Galápagos p o inces, all o which, excep o Galápagos,
an a chipelago in he Paci ic Ocean, a e loca ed eas o he highlands. 6. Taxes Paid e e s o ax paid by sel -employed
indi iduals, including hose who paid USD 0. 7. Taxes Paid Among Those Who Pay comp ise he dolla amoun o axes
paid among sel -employed indi iduals who pay mo e han USD 5 in pe sonal income ax.
44
A.3 Bunching Window 3: USD 94,500 – 107,500
Table 16: Socio-Demog aphic A ibu es by Tax Regime a USD 1,500 ma gins om
he Bunching Window
Bunching Window: USD 94,500 – 107,500
Ma gins: USD 93,000 – 94,500 and USD 107,500 – 109,000
(1) (2) (3)
Simple OKAB
lb: 93.0K lb: 107.5K
Va iables up: 94.5K up: 109.0K Di . (2-1) Obse a ions
Age 41.709 42.088 0.378 1,368
(9.690) (9.330) (0.870)
Yea s in he ax sys em19.018 10.496 1.478** 1,368
(6.438) (6.483) (0.580)
Single 0.366 0.263 -0.104** 1,368
(0.482) (0.442) (0.043)
Ma ied 0.579 0.657 0.078* 1,368
(0.494) (0.476) (0.044)
O he Ma i al S a us20.054 0.080 0.026 1,368
(0.227) (0.273) (0.021)
Female 0.361 0.409 0.048 1,368
(0.480) (0.493) (0.043)
Region: Sie a30.453 0.467 0.014 1,368
(0.498) (0.501) (0.045)
Region: Cos a40.496 0.467 -0.028 1,368
(0.500) (0.501) (0.045)
Region: Amazon and Galapagos50.051 0.066 0.015 1,368
(0.220) (0.249) (0.020)
Sec o : Manu ac u e 0.004 0.000 -0.004 1,368
(0.064) (0.000) (0.005)
Sec o : T ade 0.041 0.015 -0.026 1,368
(0.197) (0.120) (0.017)
Sec o : P o essional 0.010 0.000 -0.010 1,368
(0.098) (0.000) (0.008)
Sec o : O he s 0.359 0.204 -0.155*** 1,368
(0.480) (0.405) (0.043)
G oss Business Income, USD 93,757 108,273 14,516*** 1,368
(431) (431) (39)
% Wi h Income Tax>0 0.766 0.467 -0.299*** 1,368
(0.424) (0.501) (0.039)
Taxes Paid6, USD 953 504 -450*** 1,368
(1,549) (1,138) (136)
Taxes Paid Among Those Who Pay7, USD 1,244 1,078 -166 1,007
(1,665) (1,471) (214)
% in Tax B acke 00% 0.229 0.533 0.304*** 1,368
(0.420) (0.501) (0.039)
Obse a ions 1,231 137 1,368 .
No e: Bunching Window 3 is de ined as he space be ween USD 94,500 and 107,500 in G oss Business Income. This
able conside s all axpaye s who lie inside he in e als [93,000;94,500) and (107,500;109,000] co esponding o a USD
1,500 ma gin below and abo e he bunching window lowe and uppe bounds. 1. Taxpaye ’s yea s in he ax sys em,
which is he di e ence be ween he ax o m yea and he axpaye ’s i s egis e ed yea . 2. O he Ma i al S a us include
di o ced, widowed, and ee union. 3. Sie a includes Ca chi, Imbabu a, Pichincha, Co opaxi, Tungu ahua, Bolí a ,
Chimbo azo, Caña , Azuay, and Loja p o inces, all o which a e loca ed in he highland cen e o he coun y. 4. Cos a
includes Esme aldas, Manabí, San o Domingo de los Tsáchilas, Los Ríos, Guayas, San a Elena, and El O o p o inces.
Cos a egion si es a e in he Paci ic coas side o he coun y, wes o he highlands. 5. Amazon and Galápagos include
Sucumbíos, Napo, O ellana, Pas aza, Mo ona San iago, Zamo a Chinchipe, and Galápagos p o inces, all o which, excep
o Galápagos, an a chipelago in he Paci ic Ocean, a e loca ed eas o he highlands. 6. Taxes Paid e e s o ax paid by
sel -employed indi iduals, including hose who paid USD 0. 7. Taxes Paid Among Those Who Pay comp ise he dolla
amoun o axes paid among sel -employed indi iduals who pay mo e han USD 5 in pe sonal income ax.
45
A.4 Bunching Window 4: USD 94,500 – 103,000
Table 17: Socio-Demog aphic A ibu es by Tax Regime a USD 1,500 ma gins om
he Bunching Window
Bunching Window: USD 94,500 – 103,000
Ma gins: USD 93,000 – 94,500 and USD 103,000 – 104,500
(1) (2) (3)
Simple OKAB
lb: 93.0K lb: 103.0K
Va iables up: 94.5K up: 104.5K Di . (2-1) Obse a ions
Age 41.709 42.215 0.506 1,361
(9.690) (9.382) (0.891)
Yea s in he ax sys em19.018 11.138 2.121*** 1,361
(6.438) (7.014) (0.599)
Single 0.366 0.315 -0.051 1,361
(0.482) (0.466) (0.044)
Ma ied 0.579 0.608 0.028 1,361
(0.494) (0.490) (0.046)
O he Ma i al S a us20.054 0.077 0.022 1,361
(0.227) (0.268) (0.021)
Female 0.361 0.308 -0.053 1,361
(0.480) (0.463) (0.044)
Region: Sie a30.453 0.515 0.062 1,361
(0.498) (0.502) (0.046)
Region: Cos a40.496 0.438 -0.057 1,361
(0.500) (0.498) (0.046)
Region: Amazon and Galapagos50.051 0.046 -0.005 1,361
(0.220) (0.211) (0.020)
Sec o : Manu ac u e 0.004 0.000 -0.004 1,361
(0.064) (0.000) (0.006)
Sec o : T ade 0.041 0.023 -0.018 1,361
(0.197) (0.151) (0.018)
Sec o : P o essional 0.010 0.000 -0.010 1,361
(0.098) (0.000) (0.009)
Sec o : O he s 0.359 0.269 -0.090** 1,361
(0.480) (0.445) (0.044)
G oss Business Income, USD 93,757 103,715 9,958*** 1,361
(431) (460) (40)
% Wi h Income Tax>0 0.766 0.438 -0.328*** 1,361
(0.424) (0.498) (0.040)
Taxes Paid6, USD 953 419 -534*** 1,361
(1,549) (913) (138)
Taxes Paid Among Those Who Pay7, USD 1,244 956 -288 1,000
(1,665) (1,182) (224)
% in Tax B acke 00% 0.229 0.562 0.332*** 1,361
(0.420) (0.498) (0.040)
Obse a ions 1,231 130 1,361 .
No e: Bunching Window 4 is de ined as he space be ween USD 94,500 and 103,500 in G oss Business Income. This
able conside s all axpaye s who lie inside he in e als [93,000;94,500) and (103,000;104,500] co esponding o a USD
1,500 ma gin below and abo e he bunching window lowe and uppe bounds. 1. Taxpaye ’s yea s in he ax sys em,
which is he di e ence be ween he ax o m yea and he axpaye ’s i s egis e ed yea . 2. O he Ma i al S a us include
di o ced, widowed, and ee union. 3. Sie a includes Ca chi, Imbabu a, Pichincha, Co opaxi, Tungu ahua, Bolí a ,
Chimbo azo, Caña , Azuay, and Loja p o inces, all o which a e loca ed in he highland cen e o he coun y. 4. Cos a
includes Esme aldas, Manabí, San o Domingo de los Tsáchilas, Los Ríos, Guayas, San a Elena, and El O o p o inces.
Cos a egion si es a e in he Paci ic coas side o he coun y, wes o he highlands. 5. Amazon and Galápagos include
Sucumbíos, Napo, O ellana, Pas aza, Mo ona San iago, Zamo a Chinchipe, and Galápagos p o inces, all o which, excep
o Galápagos, an a chipelago in he Paci ic Ocean, a e loca ed eas o he highlands. 6. Taxes Paid e e s o ax paid by
sel -employed indi iduals, including hose who paid USD 0. 7. Taxes Paid Among Those Who Pay comp ise he dolla
amoun o axes paid among sel -employed indi iduals who pay mo e han USD 5 in pe sonal income ax.
46
A.5 Bunching Window 5: USD 94,000 – 107,500
Table 18: Socio-Demog aphic A ibu es by Tax Regime a USD 1,500 ma gins om
he Bunching Window
Bunching Window: USD 94,000 – 107,500
Ma gins: USD 92,500 – 94,000 and USD 107,500 – 109,000
(1) (2) (3)
Simple OKAB
lb: 92.5K lb: 107.5K
Va iables up: 94.0K up: 109.0K Di . (2-1) Obse a ions
Age 42.143 42.088 -0.055 1,351
(9.682) (9.330) (0.869)
Yea s in he ax sys em19.075 10.496 1.421** 1,351
(6.343) (6.483) (0.573)
Single 0.369 0.263 -0.106** 1,351
(0.483) (0.442) (0.043)
Ma ied 0.578 0.657 0.079* 1,351
(0.494) (0.476) (0.044)
O he Ma i al S a us20.053 0.080 0.028 1,351
(0.224) (0.273) (0.021)
Female 0.355 0.409 0.054 1,351
(0.479) (0.493) (0.043)
Region: Sie a30.428 0.467 0.039 1,351
(0.495) (0.501) (0.045)
Region: Cos a40.520 0.467 -0.053 1,351
(0.500) (0.501) (0.045)
Region: Amazon and Galapagos50.052 0.066 0.014 1,351
(0.222) (0.249) (0.020)
Sec o : Manu ac u e 0.004 0.000 -0.004 1,351
(0.064) (0.000) (0.005)
Sec o : T ade 0.044 0.015 -0.029 1,351
(0.204) (0.120) (0.018)
Sec o : P o essional 0.014 0.000 -0.014 1,351
(0.118) (0.000) (0.010)
Sec o : O he s 0.386 0.204 -0.181*** 1,351
(0.487) (0.405) (0.043)
G oss Business Income, USD 93,258 108,273 15,015*** 1,351
(441) (431) (40)
% Wi h Income Tax>0 0.753 0.467 -0.286*** 1,351
(0.432) (0.501) (0.040)
Taxes Paid6, USD 913 504 -409*** 1,351
(1,468) (1,138) (130)
Taxes Paid Among Those Who Pay7, USD 1,212 1,078 -134 978
(1,581) (1,471) (204)
% in Tax B acke 00% 0.244 0.533 0.289*** 1,351
(0.430) (0.501) (0.039)
Obse a ions 1,214 137 1,351 .
No e: Bunching Window 5 is de ined as he space be ween USD 94,000 and 107,500 in G oss Business Income. This
able conside s all axpaye s who lie inside he in e als [92,500;94,000) and (107,500;109,000] co esponding o a USD
1,500 ma gin below and abo e he bunching window lowe and uppe bounds. 1. Taxpaye ’s yea s in he ax sys em,
which is he di e ence be ween he ax o m yea and he axpaye ’s i s egis e ed yea . 2. O he Ma i al S a us include
di o ced, widowed, and ee union. 3. Sie a includes Ca chi, Imbabu a, Pichincha, Co opaxi, Tungu ahua, Bolí a ,
Chimbo azo, Caña , Azuay, and Loja p o inces, all o which a e loca ed in he highland cen e o he coun y. 4. Cos a
includes Esme aldas, Manabí, San o Domingo de los Tsáchilas, Los Ríos, Guayas, San a Elena, and El O o p o inces.
Cos a egion si es a e in he Paci ic coas side o he coun y, wes o he highlands. 5. Amazon and Galápagos include
Sucumbíos, Napo, O ellana, Pas aza, Mo ona San iago, Zamo a Chinchipe, and Galápagos p o inces, all o which, excep
o Galápagos, an a chipelago in he Paci ic Ocean, a e loca ed eas o he highlands. 6. Taxes Paid e e s o ax paid by
sel -employed indi iduals, including hose who paid USD 0. 7. Taxes Paid Among Those Who Pay comp ise he dolla
amoun o axes paid among sel -employed indi iduals who pay mo e han USD 5 in pe sonal income ax.
47
A.6 Bunching Window 6: USD 92,000 – 109,500
Table 19: Socio-Demog aphic A ibu es by Tax Regime a USD 1,500 ma gins om
he Bunching Window
Bunching Window: USD 92,000 – 109,500
Ma gins: USD 90,500 – 92,000 and USD 109,500 – 111,000
(1) (2) (3)
Simple OKAB
lb: 90.5K lb: 109.5
Va iables up: 92.0K up: 111.0K Di . (2-1) Obse a ions
Age 42.203 42.986 0.783 1,466
(10.115) (9.640) (0.895)
Yea s in he ax sys em18.997 10.336 1.339** 1,466
(6.828) (6.310) (0.603)
Single 0.351 0.343 -0.009 1,466
(0.478) (0.476) (0.042)
Ma ied 0.590 0.607 0.017 1,466
(0.492) (0.490) (0.044)
O he Ma i al S a us20.058 0.050 -0.008 1,466
(0.234) (0.219) (0.021)
Female 0.357 0.257 -0.100** 1,466
(0.479) (0.439) (0.042)
Region: Sie a30.463 0.443 -0.020 1,466
(0.499) (0.499) (0.044)
Region: Cos a40.492 0.521 0.030 1,466
(0.500) (0.501) (0.044)
Region: Amazon and Galapagos50.045 0.036 -0.010 1,466
(0.208) (0.186) (0.018)
Sec o : Manu ac u e 0.003 0.000 -0.003 1,466
(0.055) (0.000) (0.005)
Sec o : T ade 0.024 0.014 -0.010 1,466
(0.154) (0.119) (0.013)
Sec o : P o essional 0.014 0.007 -0.007 1,466
(0.119) (0.085) (0.010)
Sec o : O he s 0.379 0.264 -0.115*** 1,466
(0.485) (0.443) (0.043)
G oss Business Income, USD 91,242 110,253 19,010*** 1,466
(423) (440) (38)
% Wi h Income Tax>0 0.705 0.457 -0.248*** 1,466
(0.456) (0.500) (0.041)
Taxes Paid6, USD 845 515 -330** 1,466
(1,482) (1,092) (129)
Taxes Paid Among Those Who Pay7, USD 1,198 1,126 -72 999
(1,641) (1,390) (210)
% in Tax B acke 00% 0.289 0.543 0.254*** 1,466
(0.453) (0.500) (0.041)
Obse a ions 1,326 140 1,466 .
No e: Bunching Window 6 is de ined as he space be ween USD 92,000 and 109,500 in G oss Business Income. This
able conside s all axpaye s who lie inside he in e als [90,500;92,000) and (109,500;111,000] co esponding o a USD
1,500 ma gin below and abo e he bunching window lowe and uppe bounds. 1. Taxpaye ’s yea s in he ax sys em,
which is he di e ence be ween he ax o m yea and he axpaye ’s i s egis e ed yea . 2. O he Ma i al S a us include
di o ced, widowed, and ee union. 3. Sie a includes Ca chi, Imbabu a, Pichincha, Co opaxi, Tungu ahua, Bolí a ,
Chimbo azo, Caña , Azuay, and Loja p o inces, all o which a e loca ed in he highland cen e o he coun y. 4. Cos a
includes Esme aldas, Manabí, San o Domingo de los Tsáchilas, Los Ríos, Guayas, San a Elena, and El O o p o inces.
Cos a egion si es a e in he Paci ic coas side o he coun y, wes o he highlands. 5. Amazon and Galápagos include
Sucumbíos, Napo, O ellana, Pas aza, Mo ona San iago, Zamo a Chinchipe, and Galápagos p o inces, all o which, excep
o Galápagos, an a chipelago in he Paci ic Ocean, a e loca ed eas o he highlands. 6. Taxes Paid e e s o ax paid by
sel -employed indi iduals, including hose who paid USD 0. 7. Taxes Paid Among Those Who Pay comp ise he dolla
amoun o axes paid among sel -employed indi iduals who pay mo e han USD 5 in pe sonal income ax.
48

A.7 Bunching Window 7: USD 93,500 – 109,000
Table 20: Socio-Demog aphic A ibu es by Tax Regime a USD 1,500 ma gins om
he Bunching Window
Bunching Window: USD 93,500 – 109,000
Ma gins: USD 92,000 – 93,500 and USD 109,000 – 110,500
(1) (2) (3)
Simple OKAB
lb: 92.0K lb: 109.0K
Va iables up: 93.5K up: 110.5K Di . (2-1) Obse a ions
Age 42.396 43.029 0.632 1,387
(9.796) (9.864) (0.874)
Yea s in he ax sys em19.250 10.357 1.107* 1,387
(6.522) (6.310) (0.579)
Single 0.349 0.336 -0.013 1,387
(0.477) (0.474) (0.042)
Ma ied 0.597 0.629 0.031 1,387
(0.491) (0.485) (0.044)
O he Ma i al S a us20.054 0.036 -0.018 1,387
(0.226) (0.186) (0.020)
Female 0.340 0.271 -0.069 1,387
(0.474) (0.446) (0.042)
Region: Sie a30.438 0.457 0.019 1,387
(0.496) (0.500) (0.044)
Region: Cos a40.505 0.507 0.002 1,387
(0.500) (0.502) (0.045)
Region: Amazon and Galapagos50.057 0.036 -0.021 1,387
(0.232) (0.186) (0.020)
Sec o : Manu ac u e 0.002 0.000 -0.002 1,387
(0.049) (0.000) (0.004)
Sec o : T ade 0.039 0.021 -0.018 1,387
(0.194) (0.145) (0.017)
Sec o : P o essional 0.013 0.000 -0.013 1,387
(0.113) (0.000) (0.010)
Sec o : O he s 0.380 0.257 -0.123*** 1,387
(0.486) (0.439) (0.043)
G oss Business Income, USD 92,722 109,693 16,971*** 1,387
(438) (434) (39)
% Wi h Income Tax>0 0.738 0.436 -0.302*** 1,387
(0.440) (0.498) (0.040)
Taxes Paid6, USD 813 404 -409*** 1,387
(1,308) (990) (114)
Taxes Paid Among Those Who Pay7, USD 1,102 926 -175 981
(1,414) (1,333) (186)
% in Tax B acke 00% 0.258 0.564 0.306*** 1,387
(0.438) (0.498) (0.040)
Obse a ions 1,247 140 1,387 .
No e: Bunching Window 7 is de ined as he space be ween USD 93,500 and 109,000 in G oss Business Income. This
able conside s all axpaye s who lie inside he in e als [92,000;93,500) and (109,000;110,500] co esponding o a USD
1,500 ma gin below and abo e he bunching window lowe and uppe bounds. 1. Taxpaye ’s yea s in he ax sys em,
which is he di e ence be ween he ax o m yea and he axpaye ’s i s egis e ed yea . 2. O he Ma i al S a us include
di o ced, widowed, and ee union. 3. Sie a includes Ca chi, Imbabu a, Pichincha, Co opaxi, Tungu ahua, Bolí a ,
Chimbo azo, Caña , Azuay, and Loja p o inces, all o which a e loca ed in he highland cen e o he coun y. 4. Cos a
includes Esme aldas, Manabí, San o Domingo de los Tsáchilas, Los Ríos, Guayas, San a Elena, and El O o p o inces.
Cos a egion si es a e in he Paci ic coas side o he coun y, wes o he highlands. 5. Amazon and Galápagos include
Sucumbíos, Napo, O ellana, Pas aza, Mo ona San iago, Zamo a Chinchipe, and Galápagos p o inces, all o which, excep
o Galápagos, an a chipelago in he Paci ic Ocean, a e loca ed eas o he highlands. 6. Taxes Paid e e s o ax paid by
sel -employed indi iduals, including hose who paid USD 0. 7. Taxes Paid Among Those Who Pay comp ise he dolla
amoun o axes paid among sel -employed indi iduals who pay mo e han USD 5 in pe sonal income ax.
49
A.8 Bunching Window 8: USD 92,500 – 109,500
Table 21: Socio-Demog aphic A ibu es by Tax Regime a USD 1,500 ma gins om
he Bunching Window
Bunching Window: USD 92,500 – 109,500
Ma gins: USD 91,000 – 92,500 and USD 109,500 – 111,000
(1) (2) (3)
Simple OKAB
lb: 91.0K lb: 109.5K
Va iables up: 92.5K up: 111.0K Di . (2-1) Obse a ions
Age 42.450 42.986 0.536 1,496
(10.063) (9.640) (0.890)
Yea s in he ax sys em19.186 10.336 1.150* 1,496
(6.722) (6.310) (0.593)
Single 0.327 0.343 0.015 1,496
(0.469) (0.476) (0.042)
Ma ied 0.612 0.607 -0.005 1,496
(0.487) (0.490) (0.043)
O he Ma i al S a us20.060 0.050 -0.010 1,496
(0.238) (0.219) (0.021)
Female 0.342 0.257 -0.085** 1,496
(0.475) (0.439) (0.042)
Region: Sie a30.464 0.443 -0.021 1,496
(0.499) (0.499) (0.044)
Region: Cos a40.483 0.521 0.038 1,496
(0.500) (0.501) (0.044)
Region: Amazon and Galapagos50.053 0.036 -0.017 1,496
(0.224) (0.186) (0.020)
Sec o : Manu ac u e 0.002 0.000 -0.002 1,496
(0.047) (0.000) (0.004)
Sec o : T ade 0.024 0.014 -0.010 1,496
(0.154) (0.119) (0.013)
Sec o : P o essional 0.011 0.007 -0.004 1,496
(0.105) (0.085) (0.009)
Sec o : O he s 0.374 0.264 -0.110** 1,496
(0.484) (0.443) (0.043)
G oss Business Income, USD 91,735 110,253 18,518*** 1,496
(441) (440) (39)
% Wi h Income Tax>0 0.729 0.457 -0.272*** 1,496
(0.444) (0.500) (0.040)
Taxes Paid6, USD 872 515 -358*** 1,496
(1,488) (1,092) (129)
Taxes Paid Among Those Who Pay7, USD 1,196 1,126 -70 1,053
(1,627) (1,390) (208)
% in Tax B acke 00% 0.265 0.543 0.277*** 1,496
(0.442) (0.500) (0.040)
Obse a ions 1,356 140 1,496 .
No e: Bunching Window 8 is de ined as he space be ween USD 92,500 and 109,500 in G oss Business Income. This
able conside s all axpaye s who lie inside he in e als [91,000;92,500) and (109,500;111,000] co esponding o a USD
1,500 ma gin below and abo e he bunching window lowe and uppe bounds. 1. Taxpaye ’s yea s in he ax sys em,
which is he di e ence be ween he ax o m yea and he axpaye ’s i s egis e ed yea . 2. O he Ma i al S a us include
di o ced, widowed, and ee union. 3. Sie a includes Ca chi, Imbabu a, Pichincha, Co opaxi, Tungu ahua, Bolí a ,
Chimbo azo, Caña , Azuay, and Loja p o inces, all o which a e loca ed in he highland cen e o he coun y. 4. Cos a
includes Esme aldas, Manabí, San o Domingo de los Tsáchilas, Los Ríos, Guayas, San a Elena, and El O o p o inces.
Cos a egion si es a e in he Paci ic coas side o he coun y, wes o he highlands. 5. Amazon and Galápagos include
Sucumbíos, Napo, O ellana, Pas aza, Mo ona San iago, Zamo a Chinchipe, and Galápagos p o inces, all o which, excep
o Galápagos, an a chipelago in he Paci ic Ocean, a e loca ed eas o he highlands. 6. Taxes Paid e e s o ax paid by
sel -employed indi iduals, including hose who paid USD 0. 7. Taxes Paid Among Those Who Pay comp ise he dolla
amoun o axes paid among sel -employed indi iduals who pay mo e han USD 5 in pe sonal income ax.
50
A.9 Bunching Window 9: USD 93,500 – 108,000
Table 22: Socio-Demog aphic A ibu es by Tax Regime a USD 1,500 ma gins om
he Bunching Window
Bunching Window: USD 93,500 – 108,000
Ma gins: USD 92,000 – 93,500 and USD 108,000 – 109,500
(1) (2) (3)
Simple OKAB
lb: 92.0K lb: 108.0K
Va iables up: 93.5K up: 109.5K Di . (2-1) Obse a ions
Age 42.396 42.073 -0.323 1,397
(9.796) (10.054) (0.849)
Yea s in he ax sys em19.250 10.093 0.843 1,397
(6.522) (6.243) (0.561)
Single 0.349 0.300 -0.049 1,397
(0.477) (0.460) (0.041)
Ma ied 0.597 0.640 0.043 1,397
(0.491) (0.482) (0.042)
O he Ma i al S a us20.054 0.060 0.006 1,397
(0.226) (0.238) (0.020)
Female 0.340 0.367 0.027 1,397
(0.474) (0.484) (0.041)
Region: Sie a30.438 0.493 0.055 1,397
(0.496) (0.502) (0.043)
Region: Cos a40.505 0.447 -0.059 1,397
(0.500) (0.499) (0.043)
Region: Amazon and Galapagos50.057 0.060 0.003 1,397
(0.232) (0.238) (0.020)
Sec o : Manu ac u e 0.002 0.000 -0.002 1,397
(0.049) (0.000) (0.004)
Sec o : T ade 0.039 0.020 -0.019 1,397
(0.194) (0.140) (0.016)
Sec o : P o essional 0.013 0.000 -0.013 1,397
(0.113) (0.000) (0.009)
Sec o : O he s 0.380 0.207 -0.173*** 1,397
(0.486) (0.406) (0.041)
G oss Business Income, USD 92,722 108,758 16,036*** 1,397
(438) (437) (38)
% Wi h Income Tax>0 0.738 0.413 -0.324*** 1,397
(0.440) (0.494) (0.039)
Taxes Paid6, USD 813 418 -395*** 1,397
(1,308) (1,089) (111)
Taxes Paid Among Those Who Pay7, USD 1,102 1,012 -90 982
(1,414) (1,511) (186)
% in Tax B acke 00% 0.258 0.587 0.328*** 1,397
(0.438) (0.494) (0.038)
Obse a ions 1,247 150 1,397 .
No e: Bunching Window 9 is de ined as he space be ween USD 93,500 and 108,000 in G oss Business Income. This
able conside s all axpaye s who lie inside he in e als [92,000;93,500) and (108,000;109,500] co esponding o a USD
1,500 ma gin below and abo e he bunching window lowe and uppe bounds. 1. Taxpaye ’s yea s in he ax sys em,
which is he di e ence be ween he ax o m yea and he axpaye ’s i s egis e ed yea . 2. O he Ma i al S a us include
di o ced, widowed, and ee union. 3. Sie a includes Ca chi, Imbabu a, Pichincha, Co opaxi, Tungu ahua, Bolí a ,
Chimbo azo, Caña , Azuay, and Loja p o inces, all o which a e loca ed in he highland cen e o he coun y. 4. Cos a
includes Esme aldas, Manabí, San o Domingo de los Tsáchilas, Los Ríos, Guayas, San a Elena, and El O o p o inces.
Cos a egion si es a e in he Paci ic coas side o he coun y, wes o he highlands. 5. Amazon and Galápagos include
Sucumbíos, Napo, O ellana, Pas aza, Mo ona San iago, Zamo a Chinchipe, and Galápagos p o inces, all o which, excep
o Galápagos, an a chipelago in he Paci ic Ocean, a e loca ed eas o he highlands. 6. Taxes Paid e e s o ax paid by
sel -employed indi iduals, including hose who paid USD 0. 7. Taxes Paid Among Those Who Pay comp ise he dolla
amoun o axes paid among sel -employed indi iduals who pay mo e han USD 5 in pe sonal income ax.
51
A.10 Bunching Window 10: USD 93,000 – 109,500
Table 23: Socio-Demog aphic A ibu es by Tax Regime a USD 1,500 ma gins om
he Bunching Window
Bunching Window: USD 93,000 – 109,500
Ma gins: USD 91,500 – 93,000 and USD 109,500 – 111,000
(1) (2) (3)
Simple OKAB
lb: 91.5K lb: 109.5K
Va iables up: 93.0K up: 111.0K Di . (2-1) Obse a ions
Age 42.317 42.986 0.669 1,411
(9.972) (9.640) (0.885)
Yea s in he ax sys em19.112 10.336 1.224** 1,411
(6.659) (6.310) (0.590)
Single 0.343 0.343 -0.000 1,411
(0.475) (0.476) (0.042)
Ma ied 0.601 0.607 0.006 1,411
(0.490) (0.490) (0.044)
O he Ma i al S a us20.056 0.050 -0.006 1,411
(0.230) (0.219) (0.020)
Female 0.334 0.257 -0.077* 1,411
(0.472) (0.439) (0.042)
Region: Sie a30.445 0.443 -0.002 1,411
(0.497) (0.499) (0.044)
Region: Cos a40.504 0.521 0.018 1,411
(0.500) (0.501) (0.045)
Region: Amazon and Galapagos50.051 0.036 -0.015 1,411
(0.220) (0.186) (0.019)
Sec o : Manu ac u e 0.003 0.000 -0.003 1,411
(0.056) (0.000) (0.005)
Sec o : T ade 0.027 0.014 -0.012 1,411
(0.161) (0.119) (0.014)
Sec o : P o essional 0.015 0.007 -0.008 1,411
(0.121) (0.085) (0.011)
Sec o : O he s 0.369 0.264 -0.105** 1,411
(0.483) (0.443) (0.043)
G oss Business Income, USD 92,234 110,253 18,019*** 1,411
(425) (440) (38)
% Wi h Income Tax>0 0.724 0.457 -0.267*** 1,411
(0.447) (0.500) (0.040)
Taxes Paid6, USD 860 515 -346*** 1,411
(1,455) (1,092) (127)
Taxes Paid Among Those Who Pay7, USD 1,189 1,126 -63 984
(1,592) (1,390) (204)
% in Tax B acke 00% 0.271 0.543 0.271*** 1,411
(0.445) (0.500) (0.040)
Obse a ions 1,271 140 1,411 .
No e: Bunching Window 10 is de ined as he space be ween USD 93,000 and 109,500 in G oss Business Income. This
able conside s all axpaye s who lie inside he in e als [91,500;93,000) and (109,500;111,000] co esponding o a USD
1,500 ma gin below and abo e he bunching window lowe and uppe bounds. 1. Taxpaye ’s yea s in he ax sys em,
which is he di e ence be ween he ax o m yea and he axpaye ’s i s egis e ed yea . 2. O he Ma i al S a us include
di o ced, widowed, and ee union. 3. Sie a includes Ca chi, Imbabu a, Pichincha, Co opaxi, Tungu ahua, Bolí a ,
Chimbo azo, Caña , Azuay, and Loja p o inces, all o which a e loca ed in he highland cen e o he coun y. 4. Cos a
includes Esme aldas, Manabí, San o Domingo de los Tsáchilas, Los Ríos, Guayas, San a Elena, and El O o p o inces.
Cos a egion si es a e in he Paci ic coas side o he coun y, wes o he highlands. 5. Amazon and Galápagos include
Sucumbíos, Napo, O ellana, Pas aza, Mo ona San iago, Zamo a Chinchipe, and Galápagos p o inces, all o which, excep
o Galápagos, an a chipelago in he Paci ic Ocean, a e loca ed eas o he highlands. 6. Taxes Paid e e s o ax paid by
sel -employed indi iduals, including hose who paid USD 0. 7. Taxes Paid Among Those Who Pay comp ise he dolla
amoun o axes paid among sel -employed indi iduals who pay mo e han USD 5 in pe sonal income ax.
52