增值稅外文翻譯--淺析增值稅的形成原因及其影響（節(jié)選）

1、<p>　　3200單詞，15800英文字符，4150漢字</p><p>　　出處： Michael Keen, Ben Lockwood. The value added tax: Its causes and consequences: Journal of Development Economics 2010 (92) 138–151</p><p>　　Added ta

2、x: Its causes and consequences</p><p><b>　　? </b></p><p>　　Michael Keen,  Ben Lockwood</p><p><b>　　Abstract</b></p><p>　　This paper explore

3、s the causes and consequences of the remarkable rise of the value added tax (VAT), asking what has shaped its adoption and, in particular, whether it has proved an especially effective form of taxation. It is ?rst shown

4、that a tax innovation, such as the introduction of a VAT, reduces the marginal cost of public funds if and only if it also leads an optimizing government to increase the tax ratio. This leads to the estimation, on a pane

5、l of 143 countries for 25 years, of a syst</p><p>　　Keywords: Value added tax; Tax reform</p><p>　　1. Introduction</p><p>　　Fifty years ago the value-added tax (VAT) was rarely

6、 heard of outside of France and some specialist texts. Now it is found in over 130 countries, where it commonly raises 20% or more of all tax revenue. Widely adopted in sub-Saharan Africa and elsewhere, it has been the c

7、enterpiece of tax reform in many developing countries. By any standards, the rise of the VAT has been the most signi?cant development in tax policy and administration of recent decades. And it is not over. Several Caribb

8、ean coun</p><p>　　On the ?rst of these, a key claim made by proponents of the VAT, especially for developing countries, has been that it would enhance efforts to mobilize much-needed tax revenue, not only d

9、irectly but through wider improvements in tax administration and compliance. The same argument is sometimes turned around, however, by those distrustful of the uses to which government would put additional revenue. This

10、political economy concern has been especially in?uential in the United States, and was a ke</p><p>　　Underlying both views–of the VAT as a bene?cial expansion of the armory of tax instruments available to go

11、vernments, or as all too easy a source of revenue–is a common presumption of fact: that adoption of a VAT makes it easier to raise revenue, and in that sense improves the ef?ciency of the overall tax system. This presump

12、tion can be challenged on theoretical grounds, as we touch on shortly. The concern of this paper, however, is to test it empirically: Is there any evidence that the VAT has, </p><p>　　The second broad questi

13、on–what has led countries to adopt (or not) a VAT?–is closely related to the ?rst, since the likely revenue impact has presumably been a major consideration in their decisions. Moreover, controlling for this adoption dec

14、ision is important if one isto answer the empirical ‘money machine’ question just posed. Beyond this, however, the question is of interest and importance in itself. What have been the roles, for example, of regional emul

15、ation, and of the international ?nanc</p><p>　　domestic tax design transmit themselves across countries. A potentially informative aspect of the VAT in this context is that the direct cross-border spillov

16、er effects of adoption are relatively limited (compared, for instance, to those from corporate tax reform), so that competition for tax base itself is unlikely to play a signi?cant role.</p><p>　　This pape

17、r explores these and other empirical issues related to the performance and adoption of the VAT, using an unbalanced panel of 143 countries over 26 years. There being no obvious way to test directly for ef?ciency gains f

18、rom the VAT, we develop an indirect approach. We ?rst show that, in principle (and under weak conditions), access to a more ef?cient tax instrument would be expected to lead an optimizing (but not necessarily benevolent)

19、 government to increase the revenue ratio (the rati</p><p>　　The literature on the VAT is surprisingly sparse. On theoretical aspects, while there is of course a large literature on the optimal design of tax

20、es on ?nal consumption, few contributions concern themselves with the crediting and refund mechanisms (and their potential imperfections) that make the critical difference between a VAT and, for example, a retail sales t

21、ax—and which are therefore the distinctive source of any ef?ciency gain or loss associated with the VAT. The literature establishing c</p><p>　　raised by, the full tax system. Nellor (1987), proceeding essen

22、tially by case study, provided an early assessment of the revenue impact of a few VAT adoptions. Closest to the present analysis, however, is Ebrill and others (2001), who address similar questions but do so using only a

23、 single cross-section of data, and – as also do Desai and Hines (2005) and the other papers cited above–taking the presence or absence of a VAT as exogenous. These restrictions mean that these previous results are sub<

24、;/p><p>　　The plan of the paper is as follows. Section 2 develops a simple framework for thinking about the adoption and revenue impact of the VAT, using this to derive a direct test for the presence of ef?cien

25、cy gains associated with its implementation. The estimation strategy and dataset are developed and described in Section 3. Results are reported and discussed in Section 4, and the country-specificity of the effects of t

26、he VAT to which they point is explored in Section 5. Section 6 summarizes and c</p><p>　　2.Modeling the impact and adoption of the VAT</p><p>　　This section discusses the nature and likely impa

27、ct of the VAT, and develops a methodology for modeling its adoption and evaluating its impact on the overall effectiveness of national tax systems.</p><p>　　2.1.Background</p><p>　　The essence o

28、f the VAT is that it is levied on all transactions but, being credited against tax due on traders' sales, ultimately comes to bear–if the crediting chain is unbroken, and tax properly charged on imports and remitted

29、on exports–only on ?nal consumption. The principal claim made by advocates of the VAT (and indeed by some of its opponents, as noted above) is that, if well-designed, this structure makes it a particularly ef?cient tax.

30、There are several potential sources of ef?ciency ga</p><p>　　preserving the gains in production ef?ciency from moving producer prices closer to world</p><p>　　prices. At a more general level–and

31、 especially in developing countries–adoption of the VAT is often seen as the central element in a program of modernizing tax administration, developing the use of methods of self-assessment whose generalization is expec

32、ted ulti- mately to ease administration and compliance in relation to other taxes too.</p><p>　　But it could be, on the other hand, that the VAT has actually reduced ef?ciency, for it clearly has potential w

33、eaknesses. When the VAT chain is broken–as, for many reasons, it commonly is–production inef?ciencies may be created, which could in principle offset the bene?ts of greater assurance of revenue collection. The VAT has al

34、so proved vulnerable to high pro?le criminal attack: ‘carousel fraud,’ for example, which exploits arrangements for the taxation of intra- community trade within the Eur</p><p>　　cation–stressed and pursued

35、in Keen (2008)–that most of the VAT collected in developing</p><p>　　countries is in fact levied on imports, and so functions precisely as a tariff for purchasers in the informal sector). Thus the question

36、 of whether the VAT has enhanced the ef?ciency with which governments ?nance themselves is (or ideally would be) ultimately an empirical one.</p><p>　　The issue then is how any such ef?ciency gains (or losse

37、s) might be detected empirically.7</p><p>　　2.2.The revenue impact of a tax innovation</p><p>　　To address this, consider a stylized economy consisting of a single representative consumer with

38、preferences U (C, G) de?ned over private consumption C and public expenditure G. Assume that U is strictly increasing in both arguments, quasi-concave, and–a minimal simpli?cation–that C is normal in demand. Public expen

39、diture is ?nanced by some form of taxation that generates revenue R, so that the consumer's welfare is simply U (Y ? R, R), where Y denotes gross income; which, we suppose, is a func</p><p>　　of taxation

40、 and the nature of the tax instruments available, the latter parameterized by a scalar</p><p>　　V. (In the empirics below, V will be a dummy variable taking the value unity if a VAT is present and zero if it

41、 is not). Capturing the in ef?ciency loss from taxation, pre-tax income decreases with R, with ? YR N 0 (denoting derivatives by subscripts) being the marginal deadweight loss from taxation. Marginal dead- weight loss, w

42、e shall suppose, increases with the amount of revenue raised, so that YRR b 0.8 A tax innovation, corresponding to an increase in V, thus raises the ef?ciency of the tax</p><p>　　To derive testable implicati

43、ons of such an ef?ciency gain, suppose ?rst that the government is wholly benevolent. For any V (taken as given for the moment), it then chooses R to maximize U [Y (R,V) ? R, R]. The necessary condition for this gives th

44、e modi?ed Samuelson rule:</p><p>　　This is essentially the standard form of tax effort equation estimated in the literature, but augmented by an additional term (and interaction) relating to the ef?ciency of

45、 the available tax instruments. Eq. (3) is the basis of the ?rst element of the estimation strategy developed further below, focused on the ef?ciency impact of the VAT once it is in place.</p><p>　　The focus

46、 of empirical interest will be the impact on r of an increase in V, corresponding to the adoption of a VAT. This speaks directly to the ‘ money machine’ argument, since it identi ? es the ultimate revenue impact, al

47、l else equal, of introducing a VAT. More</p><p>　　fundamentally, however, the following shows that it is informative on the nature and magnitude of any ef?ciency gains from the VAT:</p>&

48、lt;p>　　PROPOSITION. rV (V, Y) has the same sign as YRV. Proof. See Appendix A.</p><p>　　Thus the tax ratio will be positively associated with the presence of a VAT if and only deployment of a VAT reduces

49、the marginal deadweight loss of the tax system. The intuition is simple. As it becomes less costly to raise revenue at the margin, so more will optimally be raised; and so, conditional on observed income Y, the tax ratio

50、 will rise.</p><p>　　The discussion so far has presumed benevolent policy-making. This is clearly not the view of the world, however, of those who–like some members of the President's Advisory Panel–resi

51、st the introduction of a VAT on the grounds that it may lead to an undesirable increase in the size of government. But the approach pursued here does not require that one take any particular view of government and policy

52、 making. At a purely descriptive level, as noted earlier, the empirical results speak directly to </p><p>　　To see this, suppose, for instance, that policy is made to maximize some function Ψ(L,U)defined ove

53、r an item of public expenditure L that bene?ts only the policy-maker and on private utility. Thus policymakers attach some but not sole importance to the welfare of their citizens, if not through sheer benevolence then f

54、rom self-interest in terms of reelection or the avoidance of other threats to their position. The policy problem in this case can be written showing marginal willingness to pay for </p><p>　　To derive the ef

55、fect of an exogenous increase in the ef?ciency of the tax system on the amount of tax revenue that is optimally raised, and a suitable estimating equation to investigate the effect empirically, denote the revenue ratio b

56、y r ≡ R / Y and make the arguments of necessary condition (1) explicit to rewrite it as:</p><p>　　This implicitly de?nes the tax ratio as a function r(V,Y) of the level of gross income and the state of the t

57、ax system, V, taking a ?rst order approximation to which gives:</p><p>　　A different approach is taken by Bovenberg (1987), who seeks to quantify the potential ef?ciency gains from the VAT in a computable

58、general equilibrium model.</p><p>　　If the ‘some form of taxation’ in mind in this paragraph is thought of as involving more than one tax instrument, then it needs to be assumed that all are optimally and so

59、 has the same general form as that examined above. In particular, Proposition 1 continues to apply, so that–given the irrelevance of the preference structure to that result–the derivative of the tax ratio with

60、respect to V , conditional on observed income, continues to have the same</p><p>　　sign as YRV. The empirical results below thus bear, it should be stressed, on the ef?ciency of</p><p>　　the VA

61、T, not on its welfare impact.</p><p>　　In using these results to guide the empirical investigation, it is important to note that the presence or absence of a VAT is to be related to total revenue, not simply

62、 to revenue from commodity taxation. When it is adopted in tandem with trade liberalization, for instance, one would expect a VAT to raise more revenue than the predecessor sales tax simply because of the switch of reven

63、ue away from trade taxes and towards sales taxes. Similarly, the concern of the panelists cited above was not wi</p><p>　　淺析增值稅的形成原因及其影響</p><p><b>　　摘要</b></p><p>　　本篇文章

64、通過分析是什么促使了采用增值稅以及強(qiáng)調(diào)指出含稅價值是否被證實(shí)為一種特別有效的稅收政策。在開始階段事實(shí)表現(xiàn)為，如果在稅制創(chuàng)新能夠同時導(dǎo)致一個最優(yōu)化的政府提高稅率，那么一次稅制創(chuàng)新，如增值稅的引進(jìn)，就能夠減少公共資金的邊際成本。來自143個國家的專家組成的評估小組歷時25年對此做出了一個系統(tǒng)的評價，該評價描述了采用增值稅后的可能性以及增值稅對收入的影響。結(jié)果指出非常贊成采用增值稅，并且指出其對于收益率具有十分深遠(yuǎn)但是非常復(fù)雜的影響。該篇評估

65、報告非常短暫地指出，大多數(shù)采用了增值稅的國家因此增加了一種更加有效的稅收手段，盡管這點(diǎn)在撒哈拉以南的非洲表現(xiàn)得沒有那么明顯。</p><p>　　關(guān)鍵詞:增值稅,稅收改革</p><p><b>　　1簡介</b></p><p>　　在五十年前，在法國和一些專門的文獻(xiàn)之外很少提到過增值稅（VAT）?，F(xiàn)在增值稅已經(jīng)普及到130多個國家，那里所有

66、的稅收收入普遍提高了20%以上。在撒哈拉以南的非洲和其他地方的廣泛采用，它已成為許多發(fā)展中國家的稅制改革的核心。不論從采用何種標(biāo)準(zhǔn)評判，增值稅的發(fā)展是近幾十年的稅收政策和管理發(fā)展進(jìn)程中的重要標(biāo)志。它還沒有結(jié)束。一些加勒比國家正在采用增值稅的進(jìn)程中，例如，在利比亞、敘利亞和阿拉伯聯(lián)合酋長國計劃在未來幾年內(nèi)采取增值稅。這已被廣泛認(rèn)可。然而，無論是在理論或?qū)嵶C上，增值稅發(fā)展的原因和后果沒有引起重視。</p><p>　

67、　首先，增值稅的支持者提出的一個關(guān)鍵要求，特別是對于發(fā)展中國家來說，一直認(rèn)為它會提高努力籌集急需的稅收收入，不僅直接，但通過更廣泛的改進(jìn)稅收征管和遵守。然而通過政府將額外收入置于那些不受信任的用途時，同樣的論點(diǎn)有時會截然不同。這種政治經(jīng)濟(jì)問題一直在美國尤其有影響力，為什么最近總統(tǒng)小組未能達(dá)成協(xié)議是否采納增值稅的一個關(guān)鍵原因（只有經(jīng)濟(jì)合作及發(fā)展組織國家沒有增值稅）是可取的：“一些專家小組成員……關(guān)注的是引入增值稅會導(dǎo)致更高的稅收總量，隨著

68、時間的推移和促進(jìn)發(fā)展的一個更大的聯(lián)邦政府，換句話說，即增值稅將是一個‘賺錢機(jī)器’?！?lt;/p><p>　　同時相關(guān)意見——增值稅作為實(shí)益膨脹，提供給政府的稅收工具，或其中一個</p><p>　　收入來源——是一個常見的事實(shí)推定：采用增值稅使得它更容易提高收入，并在這個意義上提高整體稅制的效率?？梢源送贫ㄔ诶碚撋系奶魬?zhàn)，因?yàn)槲覀兘佑|時間不長。然而，本文的關(guān)注，是實(shí)踐上的測試它：是否有任何證

69、據(jù)顯示，增值稅事實(shí)上被證明是一個特別有效的征稅形式？進(jìn)一步探索，我們不禁要問：增值稅稅收制度的有效性變化的影響帶有國家特性，例如，被或多或少的標(biāo)記與欠發(fā)達(dá)經(jīng)濟(jì)體？所有這一切都沒有表明，但應(yīng)強(qiáng)調(diào)，增值稅的收入籌措能力的唯一標(biāo)準(zhǔn)就是通過它應(yīng)該判斷，或?qū)嶋H上增值稅并不是必然可取的，即使顯示出它與收入的增加聯(lián)系在一起。這里關(guān)注的是簡單的事實(shí)問題，即是否有任何的跡象，和如果在什么情況下</p><p>　　，采用的增值稅確

70、實(shí)導(dǎo)致了一個更有效的稅收制度。</p><p>　　第二個廣泛的問題——是什么導(dǎo)致了國家（不）采取增值稅？——是最密切相關(guān),</p><p>　　因?yàn)槭杖肟赡苡绊懰麄兊臎Q定，想必收入是一個重要的考慮因素。此外，如果一個人要回答剛提出的實(shí)證上的“賺錢機(jī)器”的問題，控制通過此決定是非常重要的。然而，除此之外，問題在于其自身的利益和重要性。例如，在區(qū)域競賽和國際金融組織中、在增值稅的推廣過程中扮

71、演著什么樣的角色？這不僅對增值稅經(jīng)驗(yàn)的歷史理解，而且對衡量其進(jìn)一步傳播的前景，對國內(nèi)稅收設(shè)計創(chuàng)新如何傳送到各個國家自己解決更廣泛的問題來說十分重要。在此背景下，就增值稅的一個潛在的信息方面來說</p><p>　　，采用直接跨境溢出效應(yīng)相對有限例如企業(yè)稅制改革，使稅基本身的競爭不可能發(fā)揮顯著的作用。</p><p>　　本文將探討與增值稅的采用和表現(xiàn)的相關(guān)問題和其他實(shí)證性的有關(guān)問題。由于沒

72、有明顯的方式來證實(shí)增值稅提高了效率，所以我們開發(fā)了一個間接的方法。首先證明，在原則上（并在更微弱的條件下），更加有效的稅收工具預(yù)計將導(dǎo)致一個最優(yōu)化的（但并不一定是仁政的）政府，使得收益率增加（收益率，也就是收入與國內(nèi)生產(chǎn)總值的比例）。遵循著一理論的結(jié)果，我們再實(shí)證性檢測增值稅的存在對收益率是否存在任何影響。這歸結(jié)為添加一個虛擬的存在的增值稅——重要的是各種交互影響的條件——從長久以來建立的“稅收努力”的作品中回歸熟悉，這些作品試圖解釋在

73、稅率上存在的跨國的差異。然后我們問道，這些影響是否是正面積極的。當(dāng)然，所有者一切也可以被認(rèn)為是作為一個簡單的直接的測試增值稅是否已被證明是一個“賺錢機(jī)器”。要控制內(nèi)生性偏差，并且要調(diào)查增值稅采納的決定性因素，增值稅假設(shè)量本身也明確模擬及卷取估計方程。</p><p>　　關(guān)于增值稅的文獻(xiàn)非常少。在理論方面來說，雖然有關(guān)于最終的最佳消費(fèi)稅設(shè)</p><p>　　計的大量文獻(xiàn)，但是很少有貢獻(xiàn)的

74、關(guān)注他們與抵稅、退稅機(jī)制（和他們的潛在的缺陷），使之成為增值稅與，例如，零售銷售稅之間的關(guān)鍵差異——因此任何效率增益或損耗的來源與增值稅存在聯(lián)系。例如，建立在同一商品稅是最優(yōu)的條件下的文獻(xiàn)同樣適用于任何其他形式的消費(fèi)稅和增值稅。關(guān)于增值稅的實(shí)證研究也很少。幾篇論文試圖通過遵守增值稅的常用的估計收益最大化的速率來對稅收收入建造模型</p><p>　　。Desai和Hines（2005）考慮增值稅對國家貿(mào)易的影響，

75、發(fā)現(xiàn)——特別是比其他收入高的國家——開放性和出口績效與增值稅和收入的規(guī)模都存在負(fù)相關(guān)。然而，在之前關(guān)于對增值稅收線建模師沒有工作的。只有兩個我們了解的研究探究了增值稅對效率的影響與（密切相關(guān)的，正如我們將看到的）由完整的稅收制度提出的收入的關(guān)系。Nellor（1987）進(jìn)行一個本質(zhì)上是由案例研究組成的研究，提供了一個由于增值稅的采用對收入產(chǎn)生的影響的早期評估。然而，最接近現(xiàn)狀的分析是Ebrill等人（2001）提出類似的問題，但僅僅只用

76、了一個單一的數(shù)據(jù)的橫截面，以及Desai和Hines（2005）也是這么做的，并且其他論文評論上面提到的分析——認(rèn)為增值稅的采用與否是外源性的。這些限制以為這這些先前的結(jié)果有潛在的嚴(yán)重偏見。</p><p>　　2對增值稅采用及影響建模</p><p>　　本節(jié)通過建模和評估其對國家稅收系統(tǒng)整體效能的影響的方法來討論增值稅的性質(zhì)和可能造成的影響。</p><p>&

77、lt;b>　　2.1背景介紹</b></p><p>　　增值稅的本質(zhì)是對所有交易征收，但由于交易者的銷售的稅額可以抵扣，所以是最終消費(fèi)者——如果抵扣鏈?zhǔn)峭暾模⑶以谶M(jìn)口處收取和出口處退回適當(dāng)?shù)亩愵~——來承擔(dān)的。如果是精心設(shè)計的，通過對增值稅的支持者主債權(quán)（實(shí)際上是由它的一些競爭對手，如上面提到的），這種結(jié)構(gòu)會使得它成為一種特別有效的稅收手段。存在一些潛在的與增值稅采用相關(guān)的效率資源。在實(shí)踐中

78、，增值稅一般會取代營業(yè)稅（通常是所有交易中都要承擔(dān)的）或者單級銷售稅（在零售或其他水平征收</p><p>　　）。但是營業(yè)稅是由中間業(yè)務(wù)承擔(dān)，從而導(dǎo)致生產(chǎn)效率低下。銷售稅經(jīng)常這么做（因?yàn)楹茈y從最終銷售中分辨出來的），此外，銷售稅是很容易逃稅和避稅的（因?yàn)槿绻驗(yàn)槟承┰?，稅收在最后的銷售點(diǎn)無法正確的征收，那收入是會完全失去的），</p><p>　　這樣往往最終只能在狹窄的稅基上征收高額

79、稅率。在其他情況下，增值稅已作為一攬子貿(mào)易自由化的一部分被采用，旨在彌補(bǔ)降低關(guān)稅同時保持移動生產(chǎn)者價格接近世界水平的價格及生產(chǎn)效率獲得的收入損失。在更廣泛的層面上——尤其是在發(fā)展中國家——采用增值稅通常被看做是一個現(xiàn)代化的稅收管理程序的核心要素，能夠開發(fā)利用自我評估的繁華預(yù)計最終以便于管理和遵守有關(guān)的其他稅收的方法。另一方面，也許增值稅在實(shí)際上獎勵了效率，因?yàn)樗@然存在潛在的弱點(diǎn)。當(dāng)增值稅鏈條斷裂時——由于許多原因，就如通常情況那樣——

80、可能造成生產(chǎn)效率低下，這在原則上可以抵消稅收征管效益更大的保證。事實(shí)證明，增值稅容易導(dǎo)致高調(diào)的犯罪攻擊：例如“旋轉(zhuǎn)木馬欺詐”，即利用共同體內(nèi)部貿(mào)易在歐盟的稅收安排，已達(dá)到1.5%-2.5%左右的凈收入，在英國甚至更多（Keen和Smith，2006）。退稅制度的不完善，和</p><p>　?。ɑ颍┻^度的法定豁免，可能意味著增值稅在實(shí)踐運(yùn)作時，在很大程度上依賴出口和中間產(chǎn)品征稅，所以傾向于減少出口和國民產(chǎn)出。非正

81、式對增值稅也有十分重要的影響。Piggott和Whalley（2001）的研究表明，擴(kuò)大消費(fèi)稅的基礎(chǔ)可以減少由于低效非正式生產(chǎn)的規(guī)模擴(kuò)大而產(chǎn)生的社會福利，相似的，在Emran和Stiglitz（2005）的研究中認(rèn)為增值稅也許會遜色于非正規(guī)部門中存在的關(guān)稅（有實(shí)質(zhì)性的資格——在Keen（2008）的研究中強(qiáng)調(diào)并且熱衷追求的——大多數(shù)發(fā)展中國家采集的增值稅實(shí)際上是對進(jìn)口征稅，因此，準(zhǔn)確來說，它的功能是作為在非正式領(lǐng)域?qū)I家征收關(guān)稅的手段）

82、。</p><p>　　因此，增值稅是否提高了工作效率與（在理想情況下）政府資助自己的問題最終是一個實(shí)證性問題。然后任何這樣的效率收益（或虧損）的問題可能都要通過實(shí)踐檢測。</p><p>　　2.2稅收創(chuàng)新對收入的影響</p><p>　　為了解決這個問題，可以考慮通過定義私人消費(fèi)C和公共支出G形成一個有代表性的經(jīng)濟(jì)組成的消費(fèi)者偏好函數(shù)U(C,G)。在兩種觀點(diǎn)中設(shè)

83、U是嚴(yán)格遞增的，凹形曲線，最小的簡化C是正常的需求。公共開支的資金通過一些形式的稅收，產(chǎn)生收益為R，使消費(fèi)者的福利只是U（Y-R，R），Y表示總收入，其中我們假設(shè)，一個功能Y(R,V)有雙方的稅收水平和提供的稅務(wù)文書的性質(zhì)，后者參數(shù)由標(biāo)量V表示。（在以下的經(jīng)驗(yàn)主義中，如果增值稅存在，V將是一個取相同價值量的啞變量，如果增值稅不存在，V是零）。捕捉稅收的低效虧損，稅前收入隨R和—YR>0下降（下標(biāo)表示衍生工具）編輯稅收凈損失。我們假

84、設(shè)，邊際凈損失隨收入的增加而增加，使得YRR<0。當(dāng)且僅當(dāng)它意味著減少的邊際凈損失，使YRV>0創(chuàng)新的稅收，相應(yīng)增加在V，從而提高效率的稅收制度。</p><p>　　為了得到這樣的效率增益測試的影響，首先假設(shè)政府是完全德政的。對于任何</p><p>　　V（暫時采用），然后選擇R使U[Y（R，V）-R，R]最大化。這個必要條件給出了修改后的薩繆爾森規(guī)則：</p>

85、<p>　　這表示支付社會福利的邊際意愿等于邊際稅收的凈損失。為了獲得稅收總額優(yōu)化提出外源性增加稅收制度的效率，合適的估計方程實(shí)</p><p>　　證調(diào)查效果的影響，表示收入比R≡R/Y參數(shù)的必要條件（1）明確把它改寫為：</p><p>　　這隱含將稅率定義為一個與國家稅收系統(tǒng)的總收入水平V相關(guān)的函數(shù)r（V，Y），這個函數(shù)的一階近似，給出的稅收比率：</p>

86、<p>　　在文獻(xiàn)中，這本質(zhì)上是在標(biāo)準(zhǔn)形式的稅收努力方程估計，但增添了一個附加項（相互影響）與可用的稅收工具的效率有關(guān)。公式（3）是根據(jù)下面進(jìn)一步發(fā)展的，一旦它在估計策略的首要因素，必是專注于增值稅的效率影響。實(shí)踐感興趣的重點(diǎn)將是由于V的增加引起r的變動，相應(yīng)的是增值稅的采納。由于引入增值稅，所有的都平等，且它確定了最終收入的影響，所以直接說增值稅是“賺錢機(jī)器”。</p><p>　　命題：rV（V，