Economics

VAT is not an instrument to combat informality


By Ro­drigo Ibar­rola

Af­ter months of un­cer­tainty, the Ex­ec­u­tive Power fi­nally ve­toed com­pletely the Law that would have al­lowed ser­vice providers to deduct Value Added Tax (VAT) on pur­chases of food prod­ucts, cloth­ing and other items that would not usu­ally be de­ductible un­der the cur­rent law, un­less they are re­lated to the ser­vice pro­vided. On the other hand, it is­sued a de­cree al­low­ing a par­tial de­duc­tion of up to 30%. The bill re­turns to Con­gress, where both cham­bers must de­cide whether to ac­cept the veto or rat­ify its ini­tial sanc­tion. Deputies ac­cepted the veto, only the rat­i­fi­ca­tion of the Sen­ate re­mains.

Judg­ing by the ap­proach of the press, as well as by the ac­tors ad­dressed, it is pos­si­ble to in­fer that most of the me­dia have adopted a fa­vor­able per­spec­tive to­wards the law. So much so that it was very un­usual to find op­pos­ing voices in the daily in­ter­views. This asym­met­ri­cal cov­er­age of the po­si­tions gives sub­stance to a pro­posal that, in prin­ci­ple, seems rea­son­able, but is not so when an­a­lyzed in de­tail.

The ex­plana­tory mem­o­ran­dum of the law men­tions that the non-in­clu­sion of su­per­mar­ket re­ceipts among the de­ductible ex­penses “con­sti­tutes a blunt blow to the work­ing class”, a brief men­tion is made of the pro­mo­tion of smug­gling and there is no men­tion at all of in­for­mal­ity.

This point is in­ac­cu­rate: work­ers to­tal some 3.6 mil­lion peo­ple, while the po­ten­tial ben­e­fi­cia­ries of this law are only some 273 thou­sand (of which only 65 thou­sand are taxed, ac­cord­ing to the Trea­sury). It is worth ask­ing whether the re­main­ing ones are not con­sid­ered work­ers, since the par­lia­men­tary ini­tia­tive would de­prive this group, which rep­re­sents 93% of the work­ers, of the ben­e­fit, thus threat­en­ing tax eq­uity. This un­de­ni­able fact -al­ready pointed out by the Min­istry of Fi­nance- has caused pri­vate pro­mot­ers of this law (such as Al­berto Sborovski, of the Paraguayan Cham­ber of Su­per­mar­kets, CA­PASU, and the an­a­lyst Amil­car Fer­reira) to change their ar­gu­ments to­wards the fight against in­for­mal­ity.

Table 1. Gen­eral VAT tax­pay­ers for per­sonal ser­vices, 2021 – In mil­lions of guaraníes

Source: Sub­sec­re­taría de Es­tado de Trib­utación.

Re­gard­ing Fer­reira, on No­vem­ber 1, in the pro­gram En Con­texto, hosted by Luis Bareiro, he men­tioned the Uruguayan case as a suc­cess­ful ex­am­ple, with­out ex­press­ing what he means by “suc­cess­ful”. There­fore, it is per­ti­nent to make some com­ments on the men­tioned case.

increasing the deductibility of VAT as an instrument to promote formality has no empirical basis. It will generate tax inequity by granting benefits to higher income sectors (the informal sector earns approximately half the income of the formal sector and still pays taxes), it will have no impact on informality, and the multiplier effect will not compensate for the decrease in tax collection.

First, the Uruguayan pol­icy con­sisted of mod­i­fy­ing the nom­i­nal rate, not the de­ductibil­ity of the set­tle­ment; sec­ond, the ob­jec­tive was to fos­ter fi­nan­cial in­clu­sion, not to re­duce in­for­mal­ity. In­deed, in 2014, Uruguay re­duced the nom­i­nal VAT rate from 22% to 20%. Then, in 2017, it low­ered it again to 18%. Ob­vi­ously, such a mea­sure had no ef­fect on in­for­mal­ity, which can be eas­ily no­ticed graph­i­cally.

Fig­ure 1. La­bor in­for­mal­ity in Uruguay, 2006-2020, in per­cent­ages

Source: Own, with data from Ob­ser­va­to­rio Ter­ri­to­rio Uruguay.

This is log­i­cal, since, ba­si­cally, the Uruguayan gov­ern­men­t’s aim was to en­cour­age the use of elec­tronic means of pay­ment, through an in­cen­tive in the VAT rate, i.e., only ap­plic­a­ble to those trans­ac­tions, and thus pro­mote fi­nan­cial in­clu­sion. In fact, the reg­u­la­tion it­self is called “Fi­nan­cial In­clu­sion Law“. It is true, it was a suc­cess, the per­cent­age of peo­ple over 15 years old with a bank ac­count went from 45% in 2014 to 64% in 2017, to stand at 74% in 2021.

Fur­ther­more, when we talk about in­for­mal­ity we can re­fer -in most cases- to the pro­por­tion of work­ers who do not con­tribute to so­cial se­cu­rity (in which case it is not clear how the VAT rate could af­fect it) or to the in­for­mal econ­omy, which is the value of eco­nomic ac­tiv­ity that is not reg­is­tered, and this can be car­ried out by for­mal ac­tors, such as, for ex­am­ple, smug­gling or drug traf­fick­ing us­ing le­gal fa­cades.

About the re­duc­tion of la­bor in­for­mal­ity in Uruguay, this was the re­sult of a se­ries of poli­cies adopted since 2005 that in­cluded the im­ple­men­ta­tion of col­lec­tive wage bar­gain­ing su­per­vised by the State, a broad (non-sec­toral) tax re­form, in­cen­tives for pro­duc­tive in­vest­ment through the ex­emp­tion of cor­po­rate in­come tax and the in­cor­po­ra­tion of the con­cept of job qual­ity. In ad­di­tion, with sus­tained eco­nomic growth, fis­cal space was cre­ated for the im­ple­men­ta­tion of so­cial se­cu­rity and em­ploy­ment plans, coun­ter­cycli­cal fis­cal pol­icy, so­cial pro­tec­tion with in­come trans­fers to the most dis­ad­van­taged sec­tors, among oth­ers. In short, the re­duc­tion of in­for­mal­ity was ob­tained thanks to a high and sus­tained eco­nomic growth, aligned with a se­ries of poli­cies pro­mot­ing em­ploy­ment gen­er­a­tion and ad­e­quate pro­tec­tion. And not through VAT.

On the other hand, in view of the pos­si­ble col­lec­tion loss of about 100 mil­lion dol­lars, es­ti­mated by the Un­der­sec­re­tary of State for Tax­a­tion (SET), in case this mea­sure is im­ple­mented, both Fer­reira and Sborovski ar­gued that the sav­ings gen­er­ated to con­sumers would gen­er­ate a “boom” in con­sump­tion that would com­pen­sate for the re­signed amount. This idea is rooted in eco­nomic the­ory and is known as the “in­come ef­fect”, which ba­si­cally stip­u­lates that al­low­ing the VAT de­duc­tion puts money in con­sumers’ pock­ets, and they, in turn, make ad­di­tional pur­chases, which stim­u­lates spend­ing and eco­nomic ac­tiv­ity.

Un­der this hy­poth­e­sis, and as­sum­ing that all pur­chases to be de­ducted are taxed at a rate of 10%, a boost would be needed such that the over­all in­crease in spend­ing -within the for­mal econ­omy- would be about US$1 bil­lion (2.5% of GDP), i.e., a mul­ti­plier ef­fect of 10 times what is de­ducted. How­ever, sev­eral of the goods af­fected al­ready have the re­duced rate of 5%, so the mul­ti­plier ef­fect should cer­tainly be greater than 10. Wait­ing for a 10-fold in­crease in spend­ing, when the em­pir­i­cal ev­i­dence places it at around 2.7, is to­tally un­sub­stan­ti­ated.

In­for­mal­ity is the cause of a lack of eco­nomic and in­sti­tu­tional de­vel­op­ment. It ex­ists be­cause it of­fers flex­i­bil­ity for em­ploy­ment in economies con­strained by low la­bor and busi­ness pro­duc­tiv­ity, and by a State that is in­ef­fi­cient in the pro­vi­sion of ser­vices and in­ef­fec­tive in its reg­u­la­tory and tax bur­den. Given these con­di­tions, if there were no in­for­mal­ity, there would be greater un­em­ploy­ment, poverty, con­flict, and crime.

The con­clu­sion we can draw from this is that in­creas­ing the de­ductibil­ity of VAT as an in­stru­ment to pro­mote for­mal­ity has no em­pir­i­cal ba­sis. It will gen­er­ate tax in­equity by grant­ing ben­e­fits to higher in­come sec­tors (the in­for­mal sec­tor earns ap­prox­i­mately half the in­come of the for­mal sec­tor and still pays taxes), it will have no im­pact on in­for­mal­ity, and the mul­ti­plier ef­fect will not com­pen­sate for the de­crease in tax col­lec­tion. Al­though in­equities al­ready ex­ist, the right thing to do would be to ad­dress them com­pre­hen­sively and not in a sec­toral man­ner with ini­tia­tives that fur­ther un­der­mine the al­ready low tax bur­den and, above all -but not least- to use rea­son­able ar­gu­ments in the dis­cus­sion.

Cover im­age: Agen­cia IP

68 views

Write a comment...

Your email address will not be published. Required fields are marked *