This part of the book is transitional in nature, for we conclude some of the strands in the arguments presented in previous parts, and we broaden the scope of our inquiry into other directions. One issue that stands out above all others in this context is the validation of the theory. The theoretical analysis of Part I identified clearly the main causes or determinants of shift-work at the level of the firm, but the discussion of estimation and the review of the empirical literature in Part II revealed the difficulty of undertaking appropriate empirical tests that will corroborate or deny the existence of these causal relationships. The individual-country results presented in this chapter bring together these two aspects of the issue by providing evidence on the validity of the theory after correcting for the econometric problems that cloud the interpretation of previous results. In Chapter 8 we move on to other dimensions of the topic. There we present evidence on the extent to which the theory previously developed explains differences in shift-work among the countries in our UNIDO data, and we summarize the available evidence on the extent and the main characteristics of shift-work in various countries. This summary provides some motivation for the subsequent discussion in Part IV, which focuses on the effects of shift-work at the economy-wide level as well as at the firm level.

As mentioned in the Introduction, much of the recent interest in shift-work among those concerned with developing countries derives from the hope that increased use of multiple shifts will contribute to solution of the employment problem. In Chapter 0 we explained why the phenomenon of substitution qualifies the favorable effects of shift-work on employment. To recapitulate, if the capital stock of the factory is taken as given, or if substitution possibilities are deemed to be absent, then increasing the number of shifts from one to two will double total employment per unit of capital stock. However, when the capital stock is regarded as endogenous and substitution possibilities are present, the choice of the double-shift system increases total employment per unit of capital stock by less than 100%, perhaps much less. One of the purposes of the present chapter is to present a theoretical treatment of this question.

The analysis of policy changes in a decentralized economy depends critically on whether or not the capital stock of factories is regarded as fixed. If it is, then a “big-push” policy package of increasing aggregate demand becomes quite appealing, because it seems that firms can increase output quite substantially by adopting multiple shifts. But if the capital stock of each factory can change, then the increase in aggregate demand may lead to an investment boom and to increased inflation instead of to the desired increase in multiple-shift operation.

As we mentioned in the previous chapter, the individual-country results do not shed much light on one issue, namely, the existence of an inverse relationship between the level of the shift differentials and shift-work (Proposition 3). Our within-country analysis provides no information on this relationship because of the absence of direct information on these differentials. Nevertheless, the analysis of intercountry differences in shift-work will allow us to overcome to some degree this absence of information and to provide indirect evidence on the existence of such a relationship. Moreover, this analysis will also broaden the scope of our inquiry. From the confines of the causes of shift-work in the context of the theory of the firm, we move into the realm of the implications of shift-work for the firm and the economy.

A useful starting point in our present endeavor is a brief discussion of the existing literature on capital utilization and the level of economic development (Section 8.1). This discussion sets the stage for the econometric analysis of differences in shift-work among the countries in our UNIDO data (Section 8.2); moreover, it also provides the setting for a review of the available evidence on the extent and main characteristics of shift-work in various countries (Section 8.3). Our concluding remarks (Section 8.4) highlight several issues raised by the results in this chapter. These issues will be taken up in Part IV.

In this chapter we shall delve more deeply into the decision to work shifts; this will be done with the aid of specific assumptions about the productive process. The most important of these assumptions is the use of a constant elasticity of substitution (CES) production function (Arrow et al. 1961) throughout the chapter to describe the extent of ex-ante substitution possibilities between capital and labor services. Specific functional forms are also relied on to investigate more thoroughly other aspects of the productive process that are potentially important in the decision to work shifts. Finally, we also find specific functional forms to be indispensable in our subsequent empirical analysis. Thus this chapter provides the functional forms of the cost ratio that will be implemented empirically.

In the next section the CES functional form will be employed to describe ex-ante substitution possibilities in the cost-minimization model of Chapter 1. The CES assumption has already been used in the shift-work literature by Winston (1974a) and Betancourt and Clague (1975), among others. We shall follow the approach used in the latter reference because it facilitates separation of the influence of capital intensity from that of ex-ante substitution possibilities as causal factors in the shift-work decision. This separation also highlights the role of the elasticity of substitution in modifying the effect of factor prices on shift-work.

Bridging the gap between the theory and the data is frequently a difficult task in economic research. Successful accomplishment of this task is the aim of the estimation process. This process, which is the focus of the second part of this book, will be critically influenced by the nature and implications of the theory, the characteristics of the data, and the purpose of the estimation.

The principal purpose of the estimation in our context is, of course, the testing of the theory of capital utilization presented in the previous part of this book. We lay the foundation for such testing in this chapter through an analysis of the implications of certain characteristics of the theory (and to a lesser extent of the data) for the estimation process. Without going into details at this point, it is worth noting that the analysis in this chapter yields insights of interest and significance for the empirical analysis of other qualitative variables as well as shiftwork. By contrast, in the next chapter the discussion focuses exclusively on capital utilization or shift-work; there we use the theoretical results from Part I and the results from this chapter to provide a novel interpretation of previous attempts at testing the theory. In so doing we furnish a review of the existing econometric literature on capital utilization. In Chapter 6 we describe in detail the procedures employed to implement the models empirically.

In the introductory chapter we indicated the reasons for focusing on the decision to work shifts at the time of the investment decision, that is, before a plant is built. The shift-work decision then involves the simultaneous choices of the level of capital stock and the rate of utilization of that capital stock. Throughout this book this decision will be formulated in terms of choosing an optimal system of operation (i.e., a single-shift, double-shift, or triple-shift system) and an optimal level of capital stock. Thus a choice is made with respect to a discrete variable and a continuous one simultaneously. There are two reasons underlying this approach. First, we believe it corresponds closely to the actual decision-making process. Second, it facilitates the analysis considerably without eliminating any theoretical insights obtainable by formulating this process in terms of two continuous variables. On the contrary, it brings out some important aspects of the decision that can be easily overlooked in a continuous formulation.

It will be assumed in this chapter that the level of output to be produced during the day is stable and remains the same regardless of which system of operation is used. This assumption converts the decision problem into a cost-minimization problem for all three systems of operation. Whatever objectives the firm may have, the considerations involved in comparing the costs of all systems at every level of output will always play a major role in the choice between systems.

Real growth and fiscal drag

Given the statutory structure of the personal income tax of a country, the average tax rate (i.e., the ratio of total income tax revenue to gross income) will increase whenever the country's income at current prices rises. It will not matter whether the increase in income is due to price changes, to real growth, or to both. The percentage increase in the average tax rate over a given period of time will depend on the rate of growth of current income and on the elasticity of the tax. The absolute increase, however, will depend also on the tax rate itself. The higher the tax rate at a given moment, the greater the absolute increase in the rate brought about by a given percentage change in income (given the elasticity).

The automatic increase in revenue brought about by increases in current income is often referred to as “fiscal drag” or “fiscal dividend.” The fiscal drag can be labeled an inflationary (or monetary) fiscal drag or a real fiscal drag, depending on whether it is brought about by income changes that reflect inflation or real growth. Indexation for price changes, as discussed in Chapters 2 and 3, will remove the inflationary fiscal drag but not the real fiscal drag. Therefore, if real growth is fast, the elasticity of the tax is high, and the average tax rate is substantial, the absolute increase in the tax rate brought about by real fiscal drag can be significant.

In Chapter 11, which dealt with interest incomes, it was found that the available evidence indicated that economic agents, at least over the period covered by the statistical analysis, suffered from fiscal illusions. This meant that the market rates of interest did not adjust enough for the effects of inflation and taxation to leave lenders with unchanged real incomes and borrowers with unchanged real costs. Borrowers were definitely the gainers from inflation. This result was not considered surprising, because it was associated with a type of distortion – that of the taxable base – that has not been obvious, until very recently, even to economists.

For wages, however, the basic tax-related distortion is strictly the result of the progressivity of the income tax. In Chapters 2 and 3 of this study we saw that the distortions caused by progressivity (1) are far simpler to understand than those in the bases, (2) have been known for a long time, and (3) have been dealt with by many countries – through indexation or tax-cut policies. In view of this, it would be strange if wage earners – especially members of national unions with research staffs – suffered widely from progressivity-related fiscal illusions; or, if they did not, if they made no attempt at protecting themselves from the effect of higher marginal tax rates by bargaining on the basis of net-of-tax wages.

Inflation affects individuals in myriad ways: as consumers, wage earners, savers, asset holders, borrowers, lenders, taxpayers, and so forth. This book will not deal with all of these effects but it will, instead, concentrate on the relationship between inflation and personal income taxes. That taxes and prices are somehow related has, of course, been known for a long time. In 1945, Colin Clark caused quite a stir with his controversial thesis that a high level of taxation sets in motion forces of an economic, political, and psychological nature that inevitably lead a country toward inflation. Clark's thesis sounded plausible in 1945, a year of high (for that time) taxes and inflation. But then, with the end of the war, inflation became a problem of the past and interest in Clark's thesis eventually subsided. That thesis, and much of the public-finance textbooks' discussion of tax shifting, dealt with the effects of taxes on prices. Today, however, inflation is no longer seen as a temporary aberration or as a disease of a few irresponsible countries, but, rather, it has become a more or less permanent condition in a universal context. As a consequence, it was inevitable that there should be a growing interest in the reverse of the abovementioned relationship. The focus of interest now is less on the effect of taxes on prices than on the effect of changes in price levels on tax systems.

The literature dealing with the impact of inflation on taxation has been biased by the recent experiences of the industrialized countries. For these countries, inflation has generally been associated with increases in the real value of income tax revenues, so that many authors have been led to believe that the main inflation-induced problems are the prevention of this supposedly unwanted, or at least unlegislated, increase in revenue and the neutralization of the inevitable effects on the redistribution of the tax burden among income groups. However, the increase in real revenue is likely to occur mainly when (1) the lags in the collection of income taxes are short, and (2) these taxes are elastic. However, while these characteristics seem to prevail in many industrialized countries, they are not common to all countries.

When one deals with countries with income taxes collected with somewhat longer lags and with elasticities not much greater, or even less, than unity, the consequences of inflation can be very different, especially when the rate of inflation becomes high. Unfortunately, these alternative situations are not products of an economist's imagination but, on the contrary, have either existed or continue to exist in many developing countries and perhaps even in some industrialized countries. For these countries, the problem has not been an increase, but rather an inflation-induced fall, in real income tax revenue. In many cases, this fall has itself become a contributing factor in the inflationary process when the affected governments have financed the fiscal deficits through the printing of new money.

In this chapter, I outline the practical experiences of several countries in trying to maintain a constant real progression of the rates through the indexation of exemptions and brackets. The countries are classified into four groups on the basis of the type of adjustment made. For the first group, the connection between the adjustment made and the rate of inflation is direct and complete, apart from discrepancies introduced by some of the special problems mentioned in Chapter 2. For the second group, the connection is still direct, but it is no longer complete. For the third group, the connection between inflation and the adjustment made is somewhat more tenuous, because substantial discretionary elements enter into the adjustment mechanisms. Finally, for the fourth group, although one can still talk about inflationary adjustments, and inflationary changes remain the most important factor, other elements enter the picture, so that one is hesitant to continue referring to these adjustments as indexation for inflation.

Countries with full, annual, automatic inflationary adjustments

The laws which describe the adjustment mechanisms are not always easy to interpret, but it appears that Canada, Australia, and Argentina are the only countries that qualify for this category. Up to 1974, Uruguay also belonged to this group. In these countries, the exemptions and deductions in fixed amounts, as well as the brackets, are fully, annually, and automatically increased to reflect changes in the consumer price index.

In this chapter, we assume that the only inflation-induced distortion in the personal income tax system is caused by the progressivity. In particular, we assume that (1) either there is no capital income problem, or (2) that capital incomes have already been adjusted. We also ignore the issue of collection lags. In this way we can focus on one basic aspect of the problem: the fact that inflation pushes taxpayers into brackets levied with higher tax rates even when their real income has not changed.

Effects of inflation on personal income tax liability

In most countries, personal incomes above certain exempted levels are taxed with progressive statutory rates. These rates apply to income brackets specified in money terms in the tax legislation. Personal exemptions, as well as some of the deductions or tax credits, are also fixed in current values. Such a legal structure for the personal income tax guarantees that, in the absence of evasion, higher taxable incomes will be subjected to higher average tax rates. This feature has generally been considered desirable from both an equity and a stabilization point of view: because it promotes the objective of income redistribution and conforms to the principle of ability to pay; and because the progressivity of the income tax generally increases its built-in-flexibility and this increase is supposed to moderate cyclical fluctuations.

In addition to incomes from wages and salaries, capital gains, and interest, some individuals receive dividends, rents, and profits from unincorporated business enterprises and partnerships. These other incomes are also distorted by inflation and require corrections that extend beyond that provided by bracket indexation. Although much of the discussion in this chapter is of some relevance to corporate profits and, consequently, to dividends, as our subject is personal income taxation, the problems as well as the solutions typical to corporations are not dealt with in detail.

By and large, business accounting relies on historical data. If prices never changed, historical data would provide accurate information for the preparation of balance sheets and income statements of business enterprises. However, in the real world, some prices are changing at all times and these changes may be either around a relatively stable average or, as it has been more often the case in the period since the end of World War II, around an average that has itself been moving in an upward direction. When prices are changing, and these changes are not taken into account, the balance sheets and the income statements of enterprises will be distorted, and this will affect their tax liabilities, as their measure of taxable profits will be distorted.

The balance sheet will be distorted because the value of assets (land, buildings, machines, etc.) will be kept at the original, and largely outdated, acquisition prices.