The City of Philadelphia

And

The Living Wage

May 8, 1998

Prepared by: Andrew J. Buck
Professor of Economics
Temple University
Philadelphia, PA 19122

A plentiful subsistence increases the bodily strength of the laborer and the comfortable hope of bettering his condition and ending his days perhaps in ease and plenty animates him to exert that strength to the utmost. Adam Smith, Wealth of Nati ons, 1776.

Low wages are by no means identical with cheap labour. From a purely quantitative point of view the efficiency of labour decreases with a wage which is physiologically insufficient … the present-day average Silesian mows, when he exerts himself to the full, little more than two-thirds as much land as the better paid and nourished Pomeranian or Mecklenburger, and the Pole, the further East he comes from, accomplishes progressively less than the German. Low wages fail even from a purely business point of view wherever it is a question of producing goods which require any sort of skilled labour, or the use of expensive machinery which is easily damaged, or in general wherever any greater amount of sharp attention or of initiative is required. Here low wag es do not pay, and their effect is the opposite of what was intended. Max Weber, The Protestant Ethic and the Spirit of Capitalism (Scribner, New York, 1925, p. 61)

… high paid labour is generally efficient and therefore not dear labour; a fact which though it is more full of hope for the future of the human race than any other that is known us, will be found to exercise a very complicating influence on the theory of distribution. Alfred Marshall, Principles of Economics (Macmillan: London 1920)

There is now enough experience around the country with living wage ordinances to summarize in tabular form (see Table 2) their essential features. All ordinances have three broad areas they address, the terms of employment (wages and benefits), the coverage of firms and workers, and compliance. There have been many proposals for the determination of the living wage rate. Although it is attractive to index the living wage so that it need not be revisited frequently, this is not common. Most commonly the living wage is stated in fixed dollar terms with periodic increases. The initial wage rate is often related to the current poverty rate in the municipality. Some ordinances are very narrow in the terms of employment and do not address anything other than wage rate. Some provide for either a health insurance benefit or an additional hourly wage in its absence. In those areas that have adopted a living wage ordinance firm coverage is most often limited to those doing business with the city government o r that have received some financial assistance. Employee coverage is most often restricted to those working on the municipal project for which there is a contract or for which a subsidy has been granted. Compliance is usually monitored via employer filing s and violation results in a fine.

Philadelphia's ordinance is not atypical in most respects. The Philadelphia proposal sets the hourly rate at the family-of-four poverty guideline established by Health and Human Services. The Philadelphia ordinance is unusual in that future changes are tied to the greater of either the poverty level for a family of four or the CPI. Philadelphia's proposed ordinance does not offer employer's the discretion to choose between health insurance and paying a higher wage. The Philadelphia ordinance also seems to extend the health insurance benefit to part-time workers. Furthermore, the 'adequate health care insurance' language of the ordinance is rather nebulous. The Philadelphia ordinance only implicitly deals with paid vacation. The ordinance prohibits firm s from funding compliance with the minimum wage by reducing benefits, including days off.

Employee coverage under the Philadelphia ordinance is typical of such laws elsewhere. Coverage of firms is somewhat more encompassing than is common. For example, there is no explicit minimum on the dollar value of the covered contract. This could be p roblematic for bids on small contracts or compliance with the ordinance by small vendors. How will a firm comply with the law when it uses only part of the time of a full time employee to fulfill the contract? Will the firm pay a different wage depending on the task being done by the worker? A larger firm might be more able to convert the worker to working on the municipal project full time since the employees wage would only be a small fraction of the firm's total wage bill. Philadelphia also has a provi sion for employment practices in a particular industry with specific reference to the characteristics of new hires. This is atypical. Waivers and exemptions seem to be in line with those granted in other localities.

Compliance is to be monitored in Philadelphia by the Department of Procurement (or Finance in special cases). The kind of information to be reported is consistent with other living wage ordinances. The reporting frequency is every six months. This is f ar less burdensome than the biweekly reporting period in the Los Angeles ordinance.

Living wage ordinances are nothing more than minimum wage laws confined to narrow geographic localities, classes of firms, and categories of workers. This being the case, our knowledge of the impacts of a living wage ordinance can be couched in ter ms of the minimum wage debate. This paper opened with comments on the cost of labor from three economists of an earlier era. Max Weber would probably find himself most closely associated with today's living wage movement. The other two writers are regarded as the fathers of modern economic theory. It is the recent interpretation of the supply and demand paradigms of Adam Smith and Alfred Marshall which form the basis for the resistance to the imposition of and increases in the minimum wage.

What we will learn from the models discussed in this section is that sometimes we can place too much reliance on the principle of Occam's razor, the idea that the simplest explanation is the best. In the first part of this section the minimum wage anta gonists' model is presented. In the conventional argument against a minimum wage an increase in the minimum causes employment to fall. The next sub-section shows that some corrections to the model will produce an increase in employment in response to an i ncrease in the minimum wage.

In the final part of this section some alternative models with a more macro orientation are considered. The conventional antagonists' model does not consider effects of higher earned incomes on the rest of the economy. Nor does that model consider the inducements for innovation resulting from higher labor costs. When these considerations are made, the adverse employment effects of a minimum wage are not so certain.

The simplest, most powerful paradigm in the economist's toolbox is that of supply and demand (Byrns and Stone). It is so simple and straightforward that it is taught as part of every college and high school principles of economics course. There are several premises on which the paradigm is based. First, it is assumed to be nearly universally true that as the price of, say apples, rises the consumer will buy fewer apples, all other things equal. If there are a few consumers in the market for which t his is not true, then in aggregation over all consumers those few are swamped and we observe that there is an inverse relationship between the price of an apple and the quantity of apples consumers are willing to purchase. The stylized consumers of this m odel are well informed about the prices of substitutes for apples and the prices of commodities which they might consume in conjunction with apples, say flour for pie crusts.

Second, the owner of the apple orchard will bring more apples to market only if the price of an apple rises to cover the cost of planting an additional tree, the extra cost of hiring more labor to pick the apples on the additional tree, and the added c ost of transporting the apples to market. The orchard owner, like the consumer is also thought to be well informed. S/He knows the value of the alternative uses to which s/he can put the land, equipment and labor employed in the orchard.

Third, the naïve paradigm assumes that there are many consumers and many sellers of apples. Since there are so many orchards offering identical apples to the market, no one of them is able to exert influence over the price at which apples are trad ed. Similarly, there are so many consumers that the purchases of any one of them is incidental to the number of bushels of apples being bought and sold.
In the context of this simple paradigm there will be a single price at which apples are bought and sold. Every farmer who wants to sell an apple at that market clearing price is able to do so. Every consumer who wants an apple at the market clearing price is able to buy one. If there is a blight which makes it more expensive to grow a given quantity of apples, then the market clearing price will rise. If it is suddenly discovered that apples contain a carcinogen, then at a given price consumers will buy f ewer apples and the market price clearing price must fall.

As a first cut at explaining observed economic phenomenon the supply and demand paradigm is as powerful as they come. Consider the decline in the price of gasoline over the last twenty-five years. At one time OPEC was a powerful cartel able to restrict the supply of gasoline to the market. Seeing the opportunity for gain, non-OPEC members brought more gas to the market, increasing the quantity available for sale at any given price. At the same time, we have started driving more fuel efficient cars. At any given price we needed less gas to drive to work. These two forces served to drive down the market clearing price of gas.

Let us return to apples. For the sake of argument we'll say that the current price of one of the identical apples offered for sale in the market is $1.00. At the end of the market day there are no apples left unsold and all who wanted to buy an apple w ere able to do so. Suppose that the United States Department of Agriculture has decided that apples should be bought and sold for no less than $1.25, instead of the $1.00 which prevails. The new minimum price is imposed by fiat. No one is permitted to off er an apple for less than $1.25 and no one may purchase an apple for less than $1.25. If consumers' tastes have not changed then they will not be willing to purchase as many apples at $1.25 as they did at $1.00. Similarly, orchard owners will bring more a pples to the market at the higher price. At the close of trading there will be some apples left unsold.

The supply and demand paradigm that seems so effective in explaining oil prices and excess supply in the market for apples is the basis for the assertion that a legally mandated minimum wage, regardless of the name, will result in some workers losing t heir jobs. It is the wisdom of this simple paradigm that firms will hire additional workers up to the point at which an additional worker's wage is just covered by the gain in sales revenue accruing from hiring an additional worker. Since the added revenu e from hiring one more worker cannot increase without limit, there must be some point beyond which the additions to revenue decline with the additions to the size of the firm's workforce. Hence, there is an inverse relationship between the wage rate and t he number of workers the firm will employ. Furthermore, no firm employs more than a small fraction of the workforce, so it has no wage setting power.

In order to induce more workers to enter the workforce or to forego more of their leisure time they must be offered a higher wage. Suppose that all those willing to work at $5.15 per hour are able to do so, and firms are able to employ all of the worke rs they need at that hourly wage. Also, one worker is interchangeable for another and there are large numbers of them. Since there are no Eric Lindros' in the market for unskilled workers, no single worker has an impact on the going wage rate.

If none of the assumptions as just outlined are violated, then there will not be any unemployment and each worker will be paid according to her value to the firm. It is observed that, by some standard, the market clearing wage is 'too low' so one day a higher authority mandates a minimum wage of, say, $7.60 per hour. In the paradigm as presented thus far it is incontrovertible that some workers must lose their jobs. The last worker hired before the imposition of the minimum was just worth the $5.15 the firm had to pay. Since the addition to revenue from hiring the last worker is just $5.15, an increase in the minimum allowable wage will cause the firm to lay off workers up to the point where the value of the incremental worker is just equal to the wage that the firm must pay. On the other side of the market, people who would not work for $5.15 may now be attracted by the new $7.60 minimum and will seek jobs.

In the paradigm the question is not whether workers will lose their jobs, but how many. The question of how many will depend on how sensitive the firm is to a change in the wage rate. If in response to a 10% increase in the wage rate, the firm reduces its employment level by a greater percent, say 15%, then we say that the demand for labor is elastic. For these numbers, not only will employment go down, but the total income of employed workers will fall as well. If the firm's response to the 10% wage i ncrease is to decrease employment by a smaller percent, say only 5%, then the demand for labor is said to be inelastic. When the demand for labor is inelastic, although the number of those employed will fall, the aggregate earnings of those who have jobs will increase.

But all is not well with this paradigm. To begin with, workers are not homogeneous, as in the assumption of identical apples. Even among the unskilled there are differences in educational attainment and commitment to the world of work. Unlike the apple , the worker must strike a bargain in the market place in order to satisfy basic needs for survival. If a household is below the poverty level, then the worker does not enter freely into the employment contract, conferring bargaining power on the firm (Pr asch). If the orchard owner is not able to earn an adequate rate of profit, she can exit the industry and apply himself elsewhere. The worker doesn't have this luxury. Workers will also have different knowledge of employment opportunities and differing ge ographic mobility, so are not as all-knowing as the orchard owner. Finally, if a higher price is received for the apple, is its nature as an apple changed in any way?

Businesses also do not satisfy the assumptions implicit in the naïve paradigm. While there may be large numbers of firms hiring workers in, say, the Philadelphia labor market, they do not all have the same needs for low wage workers or put them to the same uses. Implicit in the apple example was the assumption that the orchard owner could substitute continuously between land and tractors, or between trees and labor. One of the things we know is that it is not always possible to substitute between the inputs to production. If transporting students to school is the goal, then employing two buses and no driver is not the same as employing one bus and one driver.

Another feature of the model of the markets for apples and labor presented here is that they do not consider related phenomenon. Suppose that all fruits are found to be carcinogenic, but also contain nutrients essential for life. Can the naïve sup ply and demand model confidently predict the impact of the carcinogen announcement on the price of apples?

One must now ask whether or not the model's assumptions are essential to its predictions of job loss. The answer is an emphatic yes, and alternative fix-ups to the paradigm, which produce employment increases, are considered in the next sub-section.

Suppose we weaken just the assumption that employers have no wage setting power in the labor market (Henderson and Quandt). For example, the Four Seasons Hotel may recognize that if it wants to increase the size of its housekeeping staff by adding additional workers it must bid away workers employed elsewhere or induce new workers to enter the workforce by offering a higher wage. But in so doing, it must increase the wage of those it already employs on its housekeeping staff or risk losing them to other employers. As a result, the wage paid to the additional worker is less than the cost to the firm of making that hire. A numerical example is illustrative.

At an hourly wage of $2 only two people come to work at the Four Seasons. If the wage rate is $3 then three people come to work, and if the wage rate is $4 then four people will come to work. Initially the hotel employs two workers for one hour and inc urs a wage bill of $4. The hotel subsequently decides that it needs three people for one hour each. The necessary wage to bring forth three workers is $3, resulting in a wage bill of $9. The increase in the wage bill resulting from increasing employment b y one worker is not $3, the wage rate, but $5, an increase of $1 for each of the first workers plus the wage of the new third worker. A little later the hotel decides that it needs four workers, not just three, for one hour each. The hotel must offer $4 t o each of the four workers, for a total wage bill of $16. The increase in the wage bill from increasing employment by one worker is $7, not the $4 wage rate. The total cost of an additional hire is increasing faster than the increase in the wage rate nece ssary to make the hire.

How does the hotel decide when to stop increasing employment? As in the naïve model, the firm hires additional workers to the point where the value to the firm of the last worker hired is just equal to the change in the wage bill resulting from th at hire. Suppose that in our example the value to the Four Seasons of the second worker is $6, the value of a third worker is $5, and the value of a fourth worker is $4. The hotel will hire three workers since the incremental cost of the third hire is $5, which is the same as the value of that worker to the hotel. It will not hire a fourth worker since the incremental cost is $16, but the value of the fourth hire to the firm is only $4.

Suppose that now a higher authority intervenes and imposes a minimum wage of $4, no matter how much labor the firm hires. If the Four Seasons hires two people their wage bill will be $8, for three people the bill is $12, for four people the wage bill i s $16. The incremental cost of one more hire is always $4. In this circumstance the hotel will choose to hire four people because the incremental cost of that fourth hire is now equal to the value of that hire to the hotel! Using the same machinery and re asoning as the conventional model we have demonstrated that a minimum wage can result in an increase in employment.

Return for the moment to the naïve model. In this model unemployment will cause wages to fall until there is no more unemployment. If a worker shirks he can be fired. But since there are many firms and workers, the shirker immediately finds a replacement job. But in reality firms know that some workers are better than others. At the time they hire and train the worker they cannot tell precisely who will be the more productive laborer. Firing loafers may serve only to build a bad reputation for the firm, requiring it to pay higher wages to attract workers. Alternatively, the firm could invest in costly monitoring of worker performance. In any event, because dismissal is not costless, high turnover remains a problem.

A better mechanism to attract higher quality workers to the firm, to elicit greater productivity from the workforce, and to avoid the cost of monitoring is for some firms to pay wages above the prevailing market rate (Shapiro and Stiglitz). At the high er rate of pay it becomes more costly to the worker to be discovered shirking and be dismissed from the firm. A wage distribution also serves to sort workers into lower and higher levels of productivity. If all firms adopt the higher pay strategy, workers will still have an incentive to be more productive. Although this seems paradoxical, it is a result of the presence of unemployed workers prepared to work for the premium wage earned by the person caught shirking. In the end the premium wage will reduce shirking, increase productivity, and reduce turnover of employees.

If paying a premium over the wage at which the market would otherwise clear is beneficial for the firm, why is a minimum wage necessary? The crux of the argument was first articulated by Webb (1912) and has been advanced by Stiglitz(1987), Rebitzer and Taylor (1995), and Carter(1995). The need for a minimum wage is motivated by the fact that the cost to the firm per effective unit of labor will differ across firms as well as within the firm. These differences are accounted for by differing rates and co sts to the firm of turnover and/or differing costs of monitoring worker effort, for example. Suppose that as the wage rate rises worker productivity rises at an increasing rate as a consequence of greater worker effort and job commitment, but then the rat e of increase begins to slow. As the wage continues to grow the turnover of employees falls and the firm invests in training so that productivity again grows more quickly than the wage rate before slowing once again. In these circumstances it can happen t hat the cost per effective unit of labor is minimized for, say, two different wages. If the firm can sell all of its output at the going price then it will always pay the lower wage. On the other hand, if the mandated minimum wage falls above the lower of the two minima but below the greater of the two, then it will be in the interest of the firm to pay a higher wage in order to achieve the second of the two minima.

In the naïve model there is no explicit discussion of the way in which labor is combined with the other factors of production (equipment, energy, land, etc.). The predictive success of the model does, however, depend on some implicit assumptio ns about production. The most critical implicit assumption is that the different inputs can be continuously combined in different quantities to produce a given level output. This flies in the face of every production process in Philadelphia, from oil refi ning to serving hamburgers at a fast-food restaurant.

With the passage of time technology may make it possible to use other inputs in place of labor, but this is possible to only a limited degree in the near term. Invariably labor must be combined with other inputs in relatively fixed proportions: The tra nsportation company needs a driver for its bus, the hotel needs a custodial staff to clean its rooms, the restaurant needs a waiter to serve the customers. This is such an immutable fact of business life that it is used in most economic studies of the imp act of new or expanded enterprises.

Indeed, one can obtain Input-Output tables from the U.S. Department of Commerce which show the proportional combinations of inputs necessary to produce any one of the outputs of more than 400 industries. One can use these tables to measure the employme nt impact of an inner city university on its environs or income and tax impacts of a new ballpark on the sponsoring municipality.

If fixed factor proportions are the production rule, then it is an easy matter to show that the employment effect of a wage increase will be nonexistent. The demand for labor is not driven by the value of the output produced, a consequence of the na&iu ml;ve model, but by the quantity the firm intends to produce (Garegnani, Prasch, and Card, Katz and Krueger).

It has long been recognized that every dollar a consumer spends on goods or services accrues to someone as income. The recipient of that income in turn spends part and saves part. That second round of spending accrues to a third party as income and part of it is spent again. This multiplier effect turns an increase in spending into a powerful engine for increasing aggregate demand for goods and services. It is the basis for the formulation of fiscal policy in Washington.

The responsiveness of labor demand to a change in the wage is inelastic. The higher minimum wage transfers income from profits, perquisites and overhead to the low wage sector. Any increase in wages will be translated into higher aggregate income for c overed workers. Part of the increase in income will be spent, resulting in a many fold increase in aggregate community demand for goods and services. The increase in demand for goods and services will increase overall employment and income of the communit y via the multiplier (Mazur, Prasch).

A possible counter argument is that an increase in the minimum wage will result, in the aggregate, in an increase in the cost of doing business. So although the wage increase stimulates aggregate demand, it is choked off by an increase in the cost of g oods sold. However, minimum wage employment is quite small as a percentage of total employment. The fraction of the cost of doing business attributable to low wage employment is lower still. In fact, it has been estimated that the 40% increase in the mini mum wage in Los Angeles raised the cost of goods sold for covered firms by only 1% - 3% (Pollin and Luce). Hence it is more likely that the change in aggregate demand will swamp the small change in aggregate supply.

One may turn the cold calculus of the naïve model back on its adherents. Suppose that one is running a firm with much waste and sloth. The fact that the inflation adjusted minimum wage is not much higher than its historic low is quite convenie nt. The low real wage allows one to use the low wage worker to subsidize the firm's otherwise inefficient operation. Confronted with a mandated higher wage for its lowest paid workers, the firm's owners and managers must find ways to make the firm more ef ficient. There is a Darwinian principle at work. An increase in the minimum wage will put inefficient firms out of their misery: Good riddance to bad rubbish (Mazur). An increase in the minimum wage could well be the engine for increased efficiency in the market place.

By making unskilled labor more expensive a minimum wage may also serve as an inducement for the firm to innovate. Confronted with an increase in the mandatory minimum wage firms will be propelled to find productivity enhancing innovations. More innovat ive ones will replace firms that fail to find new technologies. As a consequence, labor productivity will increase. Technology leading to increased labor productivity can easily offset the increase in a minimum wage. Indeed, the Luddites of the 19^th century to the contrary, labor enhancing technological advance has never reduced employment. At the end of the 20^th century does it matter that technological innovation is spurred by the effort to save labor costs rather than serendipity?

The debate about the actual employment effects of a minimum wage dates back to the end of the 19^th century and is not confined to the U.S. Some of the most compelling evidence on the effects of a minimum wage comes from Great Britain. Pr ior to 1993 a system of wage boards set minimum wages by industry and job class. While the wage boards were in existence there was no evidence that they pushed the wages of covered workers above those of uncovered workers and there were no apparent disemp loyment effects (Bell and Wright). After the last of the wage boards were abolished in 1993 there were wage falls in newly created jobs, but no overall employment gains (Machin and Manning). Similar, straightforward stories can also be found in the U.S. F irst, disguised as an entry level training wage, but really as a concession to opponents of the minimum wage, there has been a sub-minimum wage. But the sub-minimum wage is seldom used. This belies the conventional view of the employment effects of a mini mum wage (Card, Katz and Krueger). Second, some individual states have had their own versions of the Davis-Bacon Act. In recent years some states have repealed their 'little Davis-Bacon Acts'. In the repeal states there have been some employment increase, but not large enough to offset, in an aggregate sense, the loss in earnings from wage rollbacks.

In pursuit of a definitive measure of the employment effects of a minimum wage economists have studied aggregate employment figures across time, across states, and across time and states. More recently a number of case studies has been undertaken. Thes e aggregate and case studies are reviewed briefly in the next two sections. The overall conclusion is that if there is a disemployment effect of increasing the minimum wage, it is quite modest.

Like the U.S., Australia has a minimum wage. Like the U.S. the real value of that minimum wage has been declining. In Australia corporate profits as a share of national income and worker productivity have both been increasing for the last fifteen y ears. At the same time the real wage of the lowest paid workers has been stagnating, with no comparable changes in the level of employment (Harcourt, 1997). This conclusion is based on economy wide figures and may or may not be informative about the exper ience of low wage workers. Studies of the U.S. are much more carefully targeted at the experiences of those working in the low wage sector.

Among opponents of the minimum wage back of the envelope calculations of the employment effect seem quite popular. Typical of this is a short essay in the American Banker's Association Journal(1996). The author plotted the ratio of the minimum wage and the teen unemployment rate and observes that the two show a high positive correlation. The reader is then asked to conclude that an increase in the minimum wage is bad for teenage employment. Another essay in this category is by Deere, Murphy and Welch ( 1995) in which they produce some very sloppy correlations and reach the conclusion that an increase in the minimum wage is bad for employment. They also offer the specious reasoning that the minimum wage is undesirable because employer and worker have alr eady reached an agreement on what constitutes and hour of work at a lower wage. If a higher wage at a greater level of effort had been desirable then employer and worker would have bargained tothat solution. Even though these papers are so transparently f lawed it is still worth holding them up to closer scrutiny. In a study using data from the National Longitudinal Surveys of Youth Schiller(1994) reports that state wage floors appear to have no impact on youth employment or entry wages. Paradoxically cove rage exemptions can increase both employment and wages.

In only one of the publications reviewed for this paper was it reported that the minimum wage would reduce employment and earnings for the target group. Wessels (1993) goes even further to assert that allowing tips to satisfy the minimum wage requireme nt would create 360,000 jobs and increase total income for workers by 8%. Although interesting his calculations rely on rather unreasonable assumptions about patrons tipping behavior, the market capacity for new tipped establishments, and employment pract ices of restaurants.

In their recent book Card and Krueger (1995) offer a comprehensive review of the aggregate analyses of the minimum wage. The overarching conclusion is that when they are observed, the employment effects of a change in the minimum wage are small and fal l mostly on teenagers. Even their most ardent critics do not produce results that are qualitatively different from these conclusions.

In her study of employment in fast food restaurants in Atlanta Young(1991) reports on the effect of the increase in the minimum wage from $3.35 per hour. Basically there was no impact. This is not surprising since at the time the starting wa ge was $4.01 and the average crew wage was $4.42 per hour.

Card's (1992) study of the 1988 change in the California minimum wage from $3.35 to $4.25 per hour analyzes an entire market rather than just a particular industry. His methodology was to compare the difference in changes. That is he compared th e change in earnings and employment in California with earnings and employment changes in states which kept their minimum wage constant over the same period. While all covered workers experienced increases in earnings, the effect was especially important for teenagers. At the same time there was no apparent job loss. Using data from County Business Patterns (CBP), Kim and Taylor(1995) arrive at a different conclusion for retail trade in California. Their estimator for job loss was statistically biased tow ard making the demand for labor elastic. In their book Card and Krueger (1995) rework the CBP data and conclude that any job loss in retailing was not statistically significant.

In a study of fast food restaurants in Texas Katz and Krueger(1992) compared higher- and lower-wage restaurants to evaluate the effect of the change in the federal minimum wage in April 1991. Those restaurants which had to raise wages to meet th e new federal standard grew faster than those which already met the standard. Beyond that, those same restaurants did not take advantage of the availability of the sub-minimum training wage for their teenage workers.

In April 1992 New Jersey increased its minimum wage from $4.25 to $5.05 while Pennsylvania remained unchanged. The results of the Card and Krueger(1994) New Jersey - Pennsylvania study are now legend. It was their conclusion that New Jersey rest aurants which had to raise wages to meet the new minimum actually expanded employment. These results were greeted with disbelief and howls of protest. Adie and Gallaway(1995) were so convinced of predictions of the naïve model that they rested their criticism on the fact that Card and Krueger used as survey of employment rather than actual employment records. Neuberg's (1997) criticism was more reasoned, but also did not present new research. A study sponsored by a fast food trade association used pa yroll data from a subset of the fast food restaurants used by Card and Krueger. In their results Neumark and Wascher report that the 19% increase in the minimum wages caused hours worked to fall by 5%. As Card and Krueger have alleged all along, this is a very small response to the increase. The net result of the change would be some loss of jobs, but an increase in aggregate earnings.

Spriggs and Klein(1994) studied the effect of the 1991 increase in the federal minimum wage on fast food restaurants in Jackson, Mississippi, and Greensboro, North Carolina. They report that employment increased among firms that had the h ighest turnover rates. Overall there was no significant decline in employment. While prices rose in some restaurants in Jackson, they did not do so in Greensboro and there was no overall increase. Spriggs and Klein also found evidence that the increase in the minimum wage had a ripple effect across all workers to the extent that as wages of workers at the bottom rose employers tried to maintain the wage distribution.

The most recent fast food case study looks at restaurants in Monroe, Louisiana. Eisenstadt and Chachere(1998)conducted their initial survey prior to the increase in the federal minimum from $4.25 to $5.15 per hour. As a consequnce their study qu eried restaurant owners and managers about their likely response to the rise in the minimum wage. Ex ante, employers report that they expect the minimum wage increase to have minimum effects on employment practices, although do expect to boost productivit y. 80% of the survey respondents would plan price increases to offset the increase in wages.

There are both direct and indirect benefits from the change in minimum allowable wages. The direct benefit accrues directly to the wage earner. A common refrain among opponents to the minimum wage is that it benefits only teenagers in otherwise aff luent families. Although it is not universal in the sense of a broad based minimum wage, it is instructive to use the living wage as a cut-off in counting those who benefit from a minimum wage. Table 3 shows the percent of workers by age and work status t hat would benefit from an increase in the minimum wage from its federal level to the level proposed in the Philadelphia living wage ordinance. Among workers in Philadelphia, 29%, or about 170,000, are over the age of 19 and are currently earning less than the proposed living wage. Using the Current Population Survey it has been shown that low income working people receive most of the wage gains from an increase in the minimum wage: 57% of the gain from the increase from $4.25 to $5.15 accrued to 40% of wo rking families and only one in 14 recipients of the minimum wage are teens in affluent households. The allegation that affluent teens benefit most from changes in the minimum wage doesn't seem to hold much water.

The indirect beneficiaries of a change in local minimum wages may well be taxpayers. Their gains come in the form of reduced reliance by poor families on food, housing, utilities, health, and child care subsidies. Pollin and Luce (19 98) have constructed a table showing the sources of income (see Table 4) of a low income family with and without the benefit of a living wage. Their work shows that without a living wage a family of four receives about $9,500 in local, state and federal a ssistance. Of that amount $994 is in locally funded medical care.

Philadelphia's circumstances are somewhat different. In addition to funding some health care for low income households, the city also provides housing assistance through the Philadelphia Housing Authority. The city provides funding for childcare progra ms, family planning and pre- and neonatal counseling services. All of these represent uses of taxpayer dollars which might be reduced if the affected households earned a living wage. The Philadelphia Gas Works has programs to provide adequate heat to low income households. Other rate payers must subsidize these programs. Subsidies to low income households also have an impact on PGW's ability to return revenue to the city. Although they are private companies, PECO and Bell Atlantic also have programs for l ow income households that are paid for in higher rates for other customers. The extent of these subsidies is unknown at this time. However, the amount must be sizable. At present about 113,000 households in Philadelphia receive cash and in-kind assistance from one source or another. According to the U.S. Department of Housing and Urban Development the fair market rental rate for a one bedroom apartment in Philadelphia is $558 per month. After paying that rent, one of Philadelphia's 53,000 full-time, minim um wage workers would have $266 left over for taxes, transportation, utilities, clothing and food. In order to make ends meet such an individual must either rely on public assistance or get a higher wage. The reduced reliance on public monies in Table 4 t hen represent a lower bound on public savings resulting from a living wage.

Paying for the Living Wage

Economists are fond of saying that 'There is no such thing as a free lunch'. To the extent that there is an employment decline from a higher minimum wage then the increase represents a reallocation of income from some low wage workers to others. If some low wage, low productivity workers are replaced by more productive workers then there may be a further transfer of income from those the program is meant to help. Since durations of joblessness are short and turnover among low wage workers is high i n the U.S., the disemployed will not suffer for long.

The minimum wage increase might also be paid for by adjusting days off, hours of work and other fringes. Since many part-time minimum wage earners are teens and secondary workers, this may not be a very great price. Indeed it may have a salutary effect . If there are reduced opportunities for teenage employment, then maybe more of them will be induced to remain in school.

If there is a negligible employment effect from an increase in the minimum wage, then business cost must rise. Any increase in the cost of doing business will be shared between the firm and the consumer. The shares will depend on the competitiveness of the market place and the sensitivity of consumers to price changes. The burden shouldered by consumers will be spread among different income groups depending on their sensitivity to price changes. This begs the questions of whether the affected low wage workers make goods for the poor or for the affluent.

As the minimum wage rises the covered worker will rely less and less on public assistance to make ends meet. This represents a shift of the burden of ensuring the welfare of the poor from the general taxpayer to consumers. It is a market solution to th e poverty-trap of subsidies.

In 1994 The City of Baltimore passed Ordinance 442. Under the ordinance, effective July 1, 1995, anyone working on a city service contract was to be paid $6.10 per hour. One year later the minimum was to rise another $.50. Weisbrot and Sforza-Roder ick have recently assembled a study of the impact of it living wage ordinance on the city and employers of Baltimore. They addressed a number of specific allegations made by opponents of the living wage. Critics have argued that the living wage will drive up the costs of city of city contracts and that the costs would then have to be passed on in the form of higher taxes. Relying on the naïve economic model outlined above, others have fretted about the loss of employment in the city. In some locales it has been alleged that a local minimum wage with narrow coverage will result in high administrative costs. There has also been the fear that when confronted with higher wage costs, fewer firms will bid for city contracts. Finally, at the level of pure r hetoric, some critics have claimed that a living wage ordinance sends the signal that the municipality is hostile to business.

Using data on contracts affected by the living wage ordinance Weisbrot and Sforza-Roderick found that the average contract price, weighted by its share in the sample, fell by 1.92%. There are two possible explanations for this. First, contractors often submit a lower bid at renewal time. They submit a lower bid since after the first contract award potential competitors know how highly the winning firm valued the contract and can revise their own subsequent bids accordingly. Second, there may have been some productivity gain resulting from lower turnover and training costs that contractors passed on in the form of lower bids.

Baltimore's living wage law required contractors to submit biweekly payroll statements to the city for review. The city's Bureau of Management and Budget Research concluded that the enforcement cost associated with the ordinance increased the average t ax bill in the city by about $0.17 per year. This figure does not include the values of fines levied on covered firms for non-compliance.

To determine the employment impact of the ordinance Weisbrot and Sforza-Roderick interviewed 31 firms with contracts before and after the adoption of the ordinance. No firm in the sample reduced its employment level. Bidding practices were also essenti ally unchanged. 43% of contracts had the same or more bidders and 57% had fewer bidders. The average number of bidders did decline, but it was not a statistically significant decline.

In summary of the Baltimore experience, the cost of city contracts does not seem to have risen, there is no evidence that employment has fallen, and there was no significant change in bidding practices. Although they offer indirect evidence that there has been no negative impact on the perception of Baltimore as a place to do business, Weisbrot and Sforza-Roderick's finding in this regard is still too tentative.

On March 8, 1997 the Los Angeles city council unanimously approved a living wage ordinance and subsequently overrode a mayoral veto. Robert Pollin and Stephanie Luce (UC-Riverside) have turned their initial analysis of the L.A. ordinance into a boo k length research project, "The Living Wage: Building a Fair Economy". In their study Pollin and Luce analyzed the likely impact of three different living wage proposals on the City of L.A. and workers and firms in the city. The narrowest proposal was tha t adopted in Milwaukee. Their projection is that such a living wage in L.A. would affect about 1400 workers employed by 470 firms. The middle case was the ordinance actually adopted in Los Angeles and thought to impact 7,600 workers at 1,000 firms. The fi nal case was a citywide minimum wage.

The punch line in the Pollin and Luce study is that the increase in the cost of production at the covered firms amounted to about 1.5% of total costs. Since profits are typically 10% to 13% of the cost of goods sold, the increase in costs due to a high er minimum wage is not burdensome. They do concede that some firms with high concentrations of low wage workers would experience a much higher increase in cost, perhaps as much as 10% to 30%. Firms in this group include concessionaires at city owned facil ities like airports and stadiums. Such firms can pass on part of the increase in cost to customers whose demand is inelastic.

If the city were forced to absorb all the cost increases of the affected firms then the city budget would have to rise about 2%. In the context of the naïve supply and demand paradigm an increase in cost is shared by the firm and the consumer, in this case the municipality. Therefore, the 2% cost to the municipality is an upper bound. The true figure could be substantially less.

Philadelphia is a poor city. More than a quarter of its citizens live in poverty. In many families the head of the household is a minimum wage worker. Those minimum wage workers have not shared in the productivity growth of recent years. The public assistance safety net beneath them is being profoundly altered. How does a community address the social contract with its poorest citizens? A good beginning, though by no means the only approach, is to raise the minimum wage. But in raising the minimum w age we may be harming the very group we mean to help. In this paper a brief survey was taken of the extent of poverty, some of the living wage proposals, and the economics of the minimum wage.

Neither the abstract reasoning of economic theory nor statistical evidence offer any clear conclusion on the likely impact of a change in the minimum wage. Even if the naïve model of conservative business economists is correct in its predicted emp loyment impact, the likely size of the employment effect is so small as to result in a higher aggregate income for low wage workers. And that is even before we try to consider the spillover effects of the increased spending associated with the higher inco mes. If the naïve model is tweaked a little bit so that it is made to account for induced productivity gains and technological advance, then the predictions of the doom sayers are completely reversed.

With the predictions of economic theory so uncertain, one would hope that empirical evidence could settle the issue. Some controversial research has found that in response to changes in the minimum wage employment has actually risen. Some equally contr oversial research has produced exactly the opposite result. In both cases the employment effects have been very small. So small that when there was an employment decline the increase in the wage rate was more than offsetting.

Given the scholarly work of economists, Philadelphia's living wage ordinance would very likely have no adverse employment effect. Indeed, the aggregate incomes of the targeted group will rise. Because the targeted group and their employers are all conn ected to the City of Philadelphia by contract or financial subsidy, the proposed ordinance would not have the same potentially chilling effect that a more universal ordinance would have.

Comprehensive reviews have been made of the experience in Baltimore and Los Angeles. In both cities the dire predicted consequences of a living wage have not materialized. With the passage of a living wage ordinance covered workers in these cities were able to earn a wage which more closely approached the poverty level than the current federal minimum wage. The increased wage for a small covered segment of the workforce represents a significant improvement in their quality of life. At the same time the re was no flight of business from the cities. There is no evidence that firms no longer wish to bid on city contracts because of the living wage ordinance. There is no evidence that in any meaningful sense the cost of delivering city services has increase d or that the tax rate had to rise in order to pay for the living wage.

The living wage is a first step in the process of rebuilding the economic well being of Philadelphia's lowest paid workers. Taken together all of the evidence suggests that the benefits of a living wage ordinance far outweigh the costs. Philadelphia ca n expect at most a negligible impact on employment. There will be a significant increase in earnings of those covered by the ordinance. To the extent that the living wage spills over into wage increases for those just above the living wage, the city can e xpect further increases in aggregate income of its residents. It should not be necessary to raise taxes to compensate for the higher wage costs. The city can also expect a decreasing reliance on public assistance among the covered workers.

Bibliography

Adie, Douglas and Lowell Gallaway, 1995, "Myth and Measurement: The New Economics of the Minimum Wage," The Cato Journal, Vol. 15, No. 1, pp. 137.

American Banker's Association, 1996, "Minimum Wage: A Tail of Two Cities", Vol. 88, no. 6, Pp. 17.

Bell, D.N.F. and R.W. Wright, 1996, "The Impact of Minimum Wages on the Wages of the Low Paid: Evidence from the Wage Boards and Councils", The Economic Journal, Vol., 106, No. 346, Pp. 650.

Byrns, Ralph T. and Gerald W. Stone, Jr., 1993, Economics, 5^th Edition, Harper Collins College Publishers: New York, Pp. 56-79.

Card, David, 1992, "Do Minimum Wages Reduce Employment? A Case Study of California, 1987-1989," Industrial and Labor Relations Review, Vol. 46, pp. 22-37.

Card, David, Lawrence Katz, and Alan Krueger, 1994, "Comment on Neumark and Wascher: Employment Effects of Minimum and Sub-minimum Wages: Panel Data on State Minimum Wage Laws", Industrial and Labor Relations Review, Vol. 47, No. 3, Pp. 487-496.

Card, David and Alan Krueger, 1995, Myth and Measurement: The New Economics of the Minimum Wage", Princeton, N.J.: Princeton University Press.

Carter, Thomas, 1995, Minimum Wage Laws that Reduce Labor Costs, Atlantic Economic Journal, Vol. 23, No. 2, Pp. 154.

Deere, Donald, Kevin Murphy and Finis Welch, 1995, "Sense and Nonsense on the Minimum Wage", Regulation, Vol. 18, No. 1, Pp. 47.

Eisenstadt, Robert C. and J. Greg Chachere, 1998, "The Impact of Minimum Wage Legislation on Employment Practices in the Fast Food Industry: A Small Market Case", Delta Business Review, (http://cber.nlu.edu/DBR/dbrissu2/).

Garegnani, Piero, 1970, "Heterogeneous Capital, the Production Function, and the Theory of Distribution", Review of Economic Studies, Vol. 37, No. 3, Pp. 407-439.

Harcourt, Tim, 1997, "The Economics of the Living Wage", The Australian Economic Review, Vol. 30, No. 2, Pp. 194-203.

Henderson, James M. and Richard E. Quandt, 1980, Microeconomic Theory: A Mathematical Approach, 3^rd Edition, McGraw-Hill Book Company: New York, Pp. 190-192.

Kim, Taiel and Lowell Taylor, 1995, "The Employment Effect in Retail Trade of California's 1988 Minimum Wage Increase," Journal of Business and Economic Statistics, Vol. 13, No. 2, Pp. 175.

Machin, S. and A. Manning, 1996, "Employment and the Introduction of a Minimum Wage in Great Britain," The Economic Journal, Vol. 106, No. 436, Pp. 667.

Mazur, Jay, 1995, "The Minimum Wage Revisited," Challenge, Vol. 38, No. 4, Pp.23.

Milward, H. Brinton and Heidi Mosbach Newman, 1989, "State Incentive Packages and the Industrial Location Decision," Economic Development Quarterly, August, Pp. 203-222.

Neuberg, Alan, 1997, "Myth and Measurement: The New Economics of the Minimum Wage", Southern Economic Journal, Vol. 63, No. 3, Pp. 832.

Pollin, Robert and Stephanie Luce, 1998, "The Living Wage: Building a Fair Economy", The New Press, forthcoming.

Prasch, Robert, 1996, "In Defense of the Minimum Wage", Journal of Economic Issues, Vol. 30, No. 2, Pp. 391.

Prasch, Robert, 1997, "The Minimum Wage Revisited", Journal of Economic Issues, Vol. 31, No. 1, Pp. 265.

Rebitzer, James and Lowell Thomas, 1995, The Consequences of Minimum Wage Laws: Some New Theoretical Ideas", Journal of Public Economics, Vol. 56, No. 2, Pp. 245.

Schiller, Bradley, 1994, "State Minimum Wage Laws: Youth Coverage and Impact", Journal of Labor Research, Vol. 15, No. 4, Pp. 317.

Schweke, William, Carl Rist and Brian Dabson, 1994, "Bidding for Business: Are Cities and States Selling Themselves Short?", Washington, D.C.:Corporation for Enterprise Development.

Shapiro, Carl and Joseph Stiglitz, 1984, "Equilibrium Unemployment as a Worker Discipline Device", American Economic Review, Vol. 74, no. 3, June, Pp. 433-444.

Spriggs, William and Bruce Klein, 1994, " Raising the Floor: The Effects of the Minimum Wage on Low-wage Workers", Economic Policy Institute: Washington, D.C.

Webb, S. "The Economic Theory of a Legal Minimum Wage", American Economic Review, Vol. 20, No. 10, Pp. 973-998

Weisbrot, Mark and Michelle Sforza-Roderick, 1998, "Baltimore's Living Wage: An Analysis of the Fiscal and Economic Costs of Baltimore City Ordinance 442", PREAMBLE, (http://www.rtk.net/preamble/baltimore.html).

Wessels, Walter, 1993, "The Minimum Wage and Tipped Employees," Journal of Labor Research, Vol. 14, No. 3, Pp. 213.

Young, Madelyn, 1991, "Wage Rate Differentials in the Fast-Food Industry: An Atlanta Labor Market Study, unpublished Dissertation, Georgia State University.