Mainstream or “neoclassical” economics is built on a model of individual economic decisions by consumers and producers. The core of that model is a set of curves that relate demand for some commodity (by consumers) and supply (by producers) to price. The combination of those curves is the model of supply and demand (also known as the “law” of supply and demand), and is used by economists to make economic predictions and give advise, and to argue against government intervention in the economy, among others. The model and its two curves are derived by a number of steps from certain assumptions about human economic behavior, preferences, production costs, and so forth. It is nonsense, however, as I will show in the following: many of its inferences are invalid and virtually all of its assumptions are false or even downright absurd.1

One Consumer, One Commodity

Mainstream economic theory starts with a model of an individual consumer. That individual consumer is perfectly rational, perfectly informed, and perfectly selfish. Furthermore, if that consumer derives satisfaction or “utility” – measured in “utils” – from the consumption of one unit of some good, then he will get more satisfaction (and thus more utils) from consuming more, but for every additional good the increase of satisfaction will be smaller (this is called “marginal utility”). Thus, consuming one piece of rare cheesecake will produce, for example, 10 utils, the second piece 8, the third piece 6, and so forth. And by implication, consuming one piece produces 10 utils in total, consuming two pieces 18 (10+8) utils, three pieces 24 (10+8+6), and so forth.

This is, of course, absurd. Consuming many pieces of cheesecake will make the consumer feel sick, resulting in dissatisfaction (i.e. negative utils). If nausea starts to kick in after the fourth piece and overwhelms the enjoyment of eating cheesecake by the fifth, then the fifth piece would carry a satisfaction or “utility” of approximately –25 (or even lower) because that piece would cancel out all previous enjoyment, and every next piece would just lead to more nausea and thus more negative utils (but not as low as –25 as these following pieces would only increase nausea and not cancel out the previous enjoyment as the fifth piece already did that). Figure 1 below compares the utility curve according to orthodoxy – the red line – with a corrected, more realistic utility curve – the dotted blue line.

Figure 1

Somewhat similarly, the consumer’s utility of one new, white Toyota Camry might be, for example, 100 utils. Owning another white Toyota Camry wouldn’t increase the consumer’s satisfaction by much, however, and might actually decrease it if parking space would be rare and/or costly (which it is in many cities). From the third, white Toyota Camrys may very well start becoming a nuisance. Having to own three of them might – especially in a city – be so expensive and troublesome that it would be better to own none at all. Consequently, the utility curve for white Toyota Camrys looks somewhat similar to that of pieces of cheesecake, but with two important differences. Firstly, the line for cars peaks and sinks much earlier than that for pieces of cheesecake and its peak is much higher. Secondly, and much more importantly, half a piece of cheesecake will produce some satisfaction (or some nausea), but half a car is most likely just an annoyingly big piece of trash. Therefore, while the line for cheesecake is smooth, the line for white Toyota Camrys will be flat until the first whole car where it peaks, after which it sinks (because of the increasingly large part of unusable half-a-car), and suddenly peaks again at the second whole car, and then sinks below zero because of the trouble and expenses associated with owning too many (pieces of) cars.

Such a “mountainous” utility curve can also occur for other kinds of commodities. Let’s say that our consumer plays a 47-string concert harp. He’s been using strings of type A, but now wants to try strings of type B. He needs to replace all of the strings to do that, however, so the utility of 1 to 46 strings is pretty much zero. Then, at 47 strings the utility suddenly jumps, after which it continues to climb very gradually (because it is nice to have a few spare strings), until the number of strings gets so large that issues of storage start to play a role and utility starts decreasing again.

The important point here is that utility curves do not have the shape mainstream economists suppose they have. They will not continue to rise infinitely and they may have flat starts, valleys and peaks, and other “weird” features. And therefore, the “law of diminishing marginal utility”, which holds that all utility curves have shapes like the red line in figure 1,2 is false. (On a side-note, the word “law” in economics seems to be an (accidental?) euphemism for “lie”, as virtually (?) every so-called economic “law” is false, and every well-informed economist should be able to know that.)

But what if this single consumer sells off the additional pieces of cheesecake or the additional white Toyota Camrys? Wouldn’t the curve continue to climb then?

First of all, remember that this is a model of a single consumer. He is a consumer, and not a supplier or trader, and moreover, if he is single (not in the sense of “unmarried”, but in the sense of “alone”), then there is no one else to buy the surplus from him. But even if that is ignored, it doesn’t really change much. There is a limit to how many pieces of cheesecake this consumer/trader could sell. One million pieces of cheesecake would quickly turn into a rotting mess with a very large negative utility, and thus the utility decline would merely be postponed. And half cars are still useless, so the utility curve for Toyota Camrys would still have peaks and valleys. In other words, the “law of diminishing marginal utility” is still false.

But let’s ignore this, and move on.

One Consumer, Two Commodities

Let’s say that our consumer likes bowls of ramen slightly more than pieces cheesecake. One bowl of ramen gives him 12 utils, the second 10, the third 8, and so forth. Every combination of pieces of cheesecake and bowls of ramen also gives him a certain satisfaction or utility, shown in figure 2:

Figure 2

The colored lines in the figure are called “indifference curves”. They connect all combinations of goods with the same total utility. The red line connects the different combinations of ramen and cheesecake that results in a total of 20 utils; the blue line 30 utils; the green line 40 utils; and so forth.

If bowls of ramen cost 550 yen each and pieces of cheesecake cost 400 yen, and the consumer has 2200 yen in his pocket, then he can buy either 4 bowls of ramen or 5.5 pieces of cheesecake or any other combination of ramen and cheesecake on the black dotted line (or below that) in the next figure.

Figure 3

The highest total satisfaction the consumer can reach with his money is 40 utils, represented by the green line. And because our consumer is perfectly rational, perfectly informed, and perfectly selfish, he will try to achieve that maximum satisfaction by purchasing 2 pieces of cheesecake and 2.5 bowls of ramen.

If we vary the price of ramen (and ad a few more indifference curves in between 30 and 40 to the diagram), then we can see the effect of price on our “rational” consumer’s consumption.

Figure 4

The top black line is the line for a ramen price of 550 yen per bowl. At that price (and a fixed price of cheesecake), he will buy 2.5 bowls of ramen (as mentioned above). At a price of 630 yen (the second black line from the top), the highest indifference curve he can reach is that of 37.5 utils (the second red curve from the top), which corresponds to approximately 2.2 bowls of ramen (the yellow dot where the black line touches the red curve) (and a bit more cheesecake than before). At 730 yen he can reach 35 utils by buying 1.6 bowls of ramen; and at 1000 yen he can reach 32.5 utils (the lowest red curve) by buying 0.8 bowls of ramen (the bottom right yellow dot).

With this data we can make a new diagram, showing the relation between price and consumption. This is the “demand curve” for an individual, single consumer.

Figure 5

Thus far, this may all seem to make perfect sense, but remember that all of it depends on the utility curves and a bunch of other assumptions. That utility shapes do not have the shape they are supposed to have was already shown above, but there are other assumptions that are (almost?) equally important (and equally nonsensical). For example, the consumer must always be able to make a choice between different combinations of goods (or commodities). And his preferences must be transitive – if he prefers A over B and B over C, then he must prefer A over C. Both these assumptions may be acceptable for an idealized perfectly rational consumer, but psychologists and experimental economists have shown that they do not (always) apply to real people. And the assumption that a consumer can always compare all possible combinations of goods in terms of their utility becomes especially ludicrous if it is taken into consideration that real consumers do not compare two commodities, but – possibly – thousands.

Furthermore, even if we hang on to most of these assumptions, things do not necessarily work out as nicely as economists tend to presume. If we take into account that eating too much ramen and/or cheesecake makes our consumer nauseous and thus lead to dissatisfaction, and if eating ramen and cheesecake in certain proportions also tends to have unpleasant effects, then even with all the aforementioned implausible assumptions, a rational consumer would have indifference curves something like this:

Figure 6

And this seriously messes up the demand curve. Now, at some price levels, there won’t be a single point at which our consumer can maximize his utility, but two, which means that the supposed “curve” is not a curve at all. If we – on top of this complication – relax some of the other assumptions, we cannot draw indifference curves at all anymore (not even weird or irregular ones), and thus no demand curve either.

But, again, let’s ignore all that and move on.

Budget/Income Changes

Thus far, it was assumed that our consumer’s budget is fixed at 2200 yen, but obviously, if his income rises or falls, so does the budget he has available for ramen and cheesecake. With a smaller budget he would end up with a different combination of goods to maximize his utility. Keeping prizes constant, we can draw a number of different lines in figure 3 parallel to the black dotted line to represent different budgets. These line each touch different indifference curves. If we connect the points at which such budget lines for different budgets reach maximum utility, we get the thick magenta line in the following figure:

Figure 7

This line is called an “Engel curve”. It has a rather peculiar shape in figure 6, which is largely due to the fact that the indifference curves in my figure aren’t as nice and smooth as the indifference curves of a perfectly rational, perfectly informed, and perfectly consumer that can (and prefers to) consume infinite amounts of anything are supposed to be. On the other hand, as Steve Keen points out, “Engel curves can take almost any shape at all” (Debunking Economics, p. 50), so there is no reason why this oddly shaped curve would be impossible. And besides, the Engel curve in this figure is rather ordinary compared to what you’d get if you’d try to fit an Engel curve into figure 6.

An Engel curve shows how a consumer’s spending pattern changes with a change in income. For reasons yet to be explained, mainstream economists assume that Engel curves are straight lines, which implies that spending patterns do not change when income changes. If our consumer gets richer, he just gets more of the same things, but in the exact same proportion (and the same if he gets poorer). This is obviously absurd – spending patterns change very much with a change in income/budget, but this too, we’ll have to ignore to move on.

Multiple Consumers, Multiple Commodities

No economy consist of a single consumer, so a demand curve for a single consumer is rather useless – we need a demand curve for an economy as a whole, or in other words, for all the consumers in that economy together. To get that demand curve, you’d have to find the individual demand curves of all consumers in the economy, and add up the total number of bowls of ramen consumed at each price level. If there are a thousand consumers in our economy and they have pretty similar preferences, then the market demand curve could – supposedly – look something like this:

Figure 8

There is a rather nasty complication, however. The derivation of the individual demand curve depends on a fixed price of cheesecake and a fixed budget. Regardless of the price of ramen, our single consumer still has only 2200 yen to spend. That’s fine in the single consumer model, but that becomes very implausible in the case of whole economies. If the price of ramen goes up or down and/or if the consumption of ramen rises or falls, that will influence the income – and thus the budget – of some people in the economy and those people are consumers too. In other words, changes in price and changes in consumption change the incomes and budgets of some consumers, and therefore, in an economy with multiple consumers, budgets cannot be fixed.

Furthermore, there is another kind of income effect that comes into play when three or more commodities are taken into account. For example, if the consumption of one of those three commodities cannot easily change, but its price can, then a change in the price of that commodity will effectively raise or lower the consumer’s income, and thereby his budget available for the other two commodities. (Think of a raise in rent, for example. Moving house may not be a (short term) option, so such a change would effectively just decrease a consumer’s income and budget.) Or if a commodity takes up much of a consumer’s budget but is really considered inferior by that consumer, then a decrease in the price of that commodity may make it possible for the consumer to buy better alternatives, and thus lead to a decrease in consumption of the inferior good. (For example, if due to poverty all you can afford is one kind of bread, if that bread gets cheaper, you have to spend less on it, but because of that you can buy less of it and spend what you save on better food.) Hence, while a demand curve is supposed to slope downwards (i.e. rising prices mean falling consumption and falling prices mean rising consumption) it can be the other way around.

When such effects and the role of income are properly taken into account it turns out that a demand curve “can take any shape at all – except one that doubles back on itself” (Steve Keen, Debunking Economics, p. 52). In a paper published in 1953 W. Gorman proved that the only way to get demand curves with the shape mainstream economists believe they have – that is, downward sloping lines similar to the individual demand curve in figure 5 – is by assuming that the Engel curves of all consumers are parallel, straight lines.3 These are interesting assumptions. As mentioned above, assuming that Engel curves are straight lines is assuming that spending patterns do not change with income, which really only could be the case if there is just one commodity available. Assuming that they are parallel for all consumers is assuming that all consumers have identical preferences, which really only could be the case if there is just one consumer. These are obviously absurd assumptions. If the only way a continuously downward sloping demand curve (or line) can be derived is by assuming that an economy consists of a single consumer consuming as single commodity, then this demand curve has little (if anything) to do with a real economy. In a real economy (even ignoring most of the problems mentioned in previous sections), a demand curve can have almost any shape.

But let’s ignore all that as well and move on.

One Producer (Monopoly)

Let’s assume a downward sloping demand curve (but remember that every step in its derivation turned out to be invalid and that every assumption it is based on is false). If there is only one producer, then there are only two things that matter: the market demand curve for whatever it is producing, and the costs of production per unit.

Economists divide production costs into two different kinds: fixed costs and variable costs. Fixed costs do not depend on how many units are produced – they include the costs of buildings, machinery, tools, and so forth. Variable costs depend on the level of production and include labor and resources (inputs or ingredients). It is assumed that variable costs rise with the level of production: the more units one attempts to produce at the same fixed costs, the more labor (and/or other variable cost factors) is needed. Therefore, the curve of variable costs per unit of production rises. The curve of fixed costs, on the other hand, starts very high, then drops steeply, and almost flattens out. (The first units (bowls of ramen, pieces of cheesecake, cars, or whatever) are expensive to make due to fixed costs (buildings, machinery, and so forth), but the more units are made, the less these fixed costs matter.) In the following figure, the red line represents fixed costs, the blue line variable costs, and the black dotted line total average costs of production.

Figure 9

It was already shown in the 1950s, however, that this figure doesn’t look like a real average cost curve at all.4 95% of managers reported that there is no significant rise in the variable costs. In fact, for the vast majority of firms, the variable cost curve appears to be nearly or completely flat. This will turn out to have important implications.

Profits are total income minus total costs. Total costs is just the number of units produced multiplied by the average production costs (represented by the black dotted line in the above figure). Total income is the number of units produced multiplied by the price at which they can be sold. We’ll use the demand curve derived above, so if the producer sets the price at 900, it can sell 1000 units (bowls of ramen in the above example). The more units (bowls) are produced, the lower the price has to be to sell all of them.

The following figure shows total costs (red line) and total income (green line). The difference between those two is total profit (blue line). The maximum profit our producer can reach is a bit over 881 thousand by making 1973 units and selling them at 681 each.

Figure 10

The two dotted variant lines show what happens if the nonsensical assumption of rising average variable costs is discarded. Then total production costs (red dotted line) are linear and the profit peak (blue dotted line) occurs later. The maximum profit is now almost 1 million by selling 2488 units at a price of 588.

Multiple Producers (Competition)

Supposedly, the situation changes drastically if there are many producers of the (exact!) same good. Economist assume that in typical markets there will be very many producers, and that because of that none of them can influence the total supply or price. All of them, therefore, have to accept the market price, because if one of them would sell above the market price, it would lose all its customers to its competitors (and consequently go bankrupt). And if it would sell below the market price it would decrease its own profits. (There’s something very fishy about this, but we’ll get to that soon.) Given the average production costs curve shown above (in figure 9), for every price level, we can calculate the maximum profits a producer can reach at that price. The following figure shows these maximum profits.

Figure 11

This is the supply curve for an individual producer. If the market price is 600, the producer can reach maximum profit by producing – and selling! – 2458 units. The supply curve starts at a price of 230. Below that price, the producer cannot make a profit (at any production level).

To derive this supply curve, it is essential to assume that the producer can sell everything it makes (because that is what calculated income depends on). This assumption is defended in the same way as the assumption that individual producers cannot influence price or total supply: there are so many producers in the market that the production of each of them is just a drop in the ocean of total supply, and therefore, every firm can sell everything without influencing prices or total supply. This assumption is obviously absurd from a mathematical point of view, but we’ll get to that in a few more paragraphs.

The total supply is simply the sum total of individual supply curves of all producers in the market, which obviously implies that total supply (at each price level) depends on the number of producers. In fact, given that it is assumed that all producers are identical, the total supply at each price level is just the individual supply multiplied by the number of producers. Hence, the total supply curve for a market with 10 producers would look exactly like the above figure, but with one zero added to each number on the Y-axis.

If the total supply curve is drawn in one figure together with the total demand curve (figure 8), then the point where the two lines cross, is where the market is in equilibrium. At that price level all units produced will indeed be sold. In the following figure the red line is the market demand curve, and the blue line is the total supply curve. Slightly more than 4000 units will be produced, and sold at a price of 327 each.

Figure 12

What the figure doesn’t show is the number of producers in this market. That number is 2. If there would be 3 producers, then the two lines wouldn’t cross, because the three of them would together produce almost 5000 units at their lowest production levels (i.e. the lowest point on the supply curve), which would reduce the market price to well below the minimum they need to cover their costs.

But, hey, … wait, … weren’t there supposed to be many producers in this market? Two isn’t very many…

This is an obvious problem for the theory indeed, but there is an easy way to address this problem – just lower fixed costs. If fixed costs are extremely low, then even small production volumes (relative to the market) are profitable and there is room for very many producers. Unfortunately, for the vast majority of industries, the fixed costs are much larger (rather than smaller) than in the example thus far, so we’d have to – again – sacrifice realism to save the theory.

The assumption, then, that there are very many producers is not a plausible assumption in most industries. (And keep in mind that they must be producers of the exact same commodity.) But the derivation of the total supply curve depended on some other assumptions as well: because of the large number of producers, none of them can influence price and/or total supply, and rising average variable costs of production apply. The importance of the latter assumption cannot be overstated. If it is assumed that a producer can sell whatever it produces, then there is a maximum profit only if average production costs rise and at some point overtake income. If average production costs flatten out – as they do in reality (see above) – then profits keep rising with production. And if there are no maximum profit levels at various price levels, then there is no supply curve – then, regardless of the price level (and still under the assumption that all production is sold), the profit-maximizing producer sets its production level at infinite. This is obviously absurd.

Furthermore, even if there are very many producers, it is mathematically impossible that a change in production level of one of them (without compensation by the others) would not affect total supply and price. Even if there are 100 producers, and one of them increases its production only slightly, that changes total supply by a tiny little bit, and thereby price by a tiny little bit, and that tiny little bit affects everyone in the market. Economists argue that these “tiny little bits” are in fact so tiny that they can be ignored, but that is a rather dubious claim, especially if it is taken into consideration that in most markets there won’t be very many producers.

Given the model we have now it is rather easy to simulate the effect a single producer could have on the market. Let’s assume that in this market somehow a situation has evolved in which there are 10 producers, each making 250 units (and thus, 2500 in total) and selling them for 585 per unit (which is the price level at which all 2500 units will be sold according to the demand curve). All 10 of them make a rather modest profit of 8850 (or 8710 if rising average variable costs are assumed). Now, if one of them would suddenly decide to quadruple its production, then total supply would rise (from 2500) to 3250, which would reduce the price to 458. The firm that raised its production now makes a profit of 208 thousand (or 203 thousand if rising average variable costs are assumed), but the other 9 firms (that didn’t change their production level but are affected by the same change in price) now lose approximately 23 thousand each. (And even a much smaller production increase would increase profits of that firm, while reducing profits of all others.) Hence, a profit-maximizing firm would raise its production to put all of its competitors out of business and then establish a monopoly (in which it could have maximum profits). But because – supposedly – all firms in the market are profit-maximizing, they will all try to do that, leading to massive overproduction, a collapse of the price, and all of them going bankrupt.

That’s not what happens, however, and the main reason for that is that production costs work a little bit differently. (Well, not a little bit, really.) Variable costs are more or less flat – that has already been mentioned – but fixed costs aren’t as “fixed” as they are supposed to be either. So-called fixed costs are the costs for buildings, machinery, and so forth needed for a certain maximum production level – that is, fixed costs are related to a certain production capacity. If a producer runs at 100% capacity, it produces the maximum number of units that can be made with the buildings, machinery, and so forth available. There is no point in adding more labor if you are at 100% capacity, because those extra workers would just stand by and watch. And there is no point in trying to put more resources/inputs in the machines if those are already at 100% capacity either – you can’t make a machine run faster than its maximum speed. Typically, fixed costs are very high and aren’t earned back until a producers is well over 50% capacity (and it may be much closer to maximum capacity), and because of that, the vast majority of firms run at capacities between 80% and 100%. If you are already close to 100% capacity, then a production expansion is possible only by spending “fixed costs” to open another production line, which also would have to run at close-to-full capacity immediately to be profitable. That means that increasing production requires a large investment, which can be earned back only if the full jump in production can be profitably sold. Profit levels are generally insufficient to make such investments, and banks and other investors are only willing to loan the money if they are sufficiently confident that the production expansion will pay off.

There are various other complexities, but those matter little right now. The important thing is that every assumption that underlies the derivation of a supply curve has been shown to be false. Production curves aren’t even remotely similar to what economists imagine them to be. There isn’t enough room for many producers in most markets. And even if there would be many producers, the actions by any one of them would affect total supply and price. The implication is that there is no way to derive supply from price, and therefore, that there is no such thing as a supply curve.

The importance hereof cannot be overstated. Mainstream economists claim that maximum economic efficiency and maximum welfare are reached in a free market. The free market would automatically gravitate towards the point where the (non-existing) supply curve crosses the (equally problematic) demand curve, leading to a situation with the highest production and lowest prices that is economically feasible. But if there is no supply curve, then this just doesn’t follow. Then it actually turns out that maximum economic efficiency and maximum welfare are reached in a situation of democratically controlled monopoly – that is, socialism – because a monopoly minimizes fixed costs and thus can produce at the lowest cost. (We’ll still have to assume that the demand curve has the standard shape, however, and as shown above that assumption is groundless as well.)

But let’s ignore all of that and move on.

The Market as Savior

So here we are; we have consistently ignored reality by making absurd (and provably false) assumptions, by confusing small amounts with zero (in the derivation of the supply curve), and by making one invalid step after the other (in the step-wise derivation of the demand curve, especially), but it has paid of: we now have our two curves. So, what can we do with them?

Well, … nothing, might seems to be the obvious answer, as these curves have nothing to do with reality, but economists see things a bit differently. It doesn’t matter that all of the assumptions the theory is based on are false because all scientific theories are based on false assumptions, they argue, and all that does matter is whether the resulting theory is a useful tool. This is really mainstream economics’ last line of defense, and it fails as miserably as everything that came before.

A physicist might assume that friction doesn’t matter when predicting the effects of gravity on a cannon ball, for example. In that case, she is making two kinds of assumptions at once. Firstly, she restricts her theory to cannon balls and similar objects. And secondly, she assumes that for those objects the effects of friction are negligible. Economists make similar assumptions, but what they don’t seem to realize is that those assumptions imply that their theories only apply to perfectly rational, perfectly informed, and perfectly selfish, profit/utility-maximizing beings (in the same way that the physicist’s theory is restricted to cannonballs), and that for such beings certain effects can be ignored (like friction for cannonballs). But perfectly rational, perfectly informed, and perfectly selfish, profit/utility-maximizing beings don’t exist, and economists thus restrict the domain of their theories to nothing – their theories have no application.

Whether mainstream economic theories are useful is debatable as well. They are not useful to understand how an economy works because they assume things that do not and cannot exist in a real economy (such as a supply curve). And neither are they useful as tools for prediction, as the predictions of mainstream economists are consistently wrong. (The consensus among mainstream economists just a few months before the crisis of 2008 erupted was that there would be continuous economic growth and no more crises, for example.) Economic theories may be “useful” in another way, however, but let’s return to our question first: What can we do with these theories? What can we do with our fictitious supply and demand curves?

Rather neat stuff, supposedly. We can “prove”, for example, that the market is always right and that governments can only get it wrong. Here is how that is supposed to work. Suppose that the government decides that ramen shops need to be supported and towards that end implements a minimum price for ramen of 400 yen per bowl. If the supply and demand curves in the previous figure apply to the ramen market, then we can add a third line to that figure as follows:

Figure 13

The black dotted line is the minimum price of 400. The total supply is where that line crosses the blue supply curve: 4370 bowls of ramen. The total demand is where the black line crosses the red demand curve: 3600 bowls of ramen. Hence, we now have an overproduction of 770 bowls (if those extra bowls are actually produced). Consumers have to pay more for their ramen than before, and are thus unhappy. The two producers see their profits rise from 183 thousand to 306 thousand each, however, and are very pleased. But because there are very many more consumers than producers, overall welfare decreases. The same applies to virtually any other kind of government interference in the market: whatever the government does, it will decrease overall welfare. By treating labor as a commodity mainstream economists believe that they can prove that a minimum wage increases unemployment, for example.

All of this is nonsense, of course, it depends on fictitious supply and demand curves that bear little (if any) resemblance to reality. But this also reveals why these fictitious curves are so pervasive anyway: they serve an ideological agenda. They serve a pro-market and anti-government agenda that favors deregulation and small governments. They serve an agenda that benefits the financial sector and large corporations and that harms almost everyone else. It is in this sense – and only in this sense – that mainstream economic theories are “usefull”: they are useful to serve the interests of the global financial and industrial elite.

I’ve written about the latter kind of “usefullness” before here in this blog and in my little book/pamphlet The Hegemony of Psychopathy (and the relevant section is also excerpted here), and so have many others. The fact that mainstream economics is an ideology (or religion) rather than a science is not the topic of this post, so I won’t repeat (or even summarize) those arguments here. (But I want to add one thing: the above shows that mainstream economics is a surprisingly fragile ideology. Its main result – the religious belief in the market as savior – depends entirely on a false assumption about the nature of production costs. Discarding that assumption while keeping the rest of the fallacious theory would turn it into an argument for socialism instead.) Instead, what I want to suggest here is to stop ignoring reality, and to reject fallacious reasoning and false assumptions. In other words, what I want to suggest is to throw out mainstream economics.

Then What?

If we throw out mainstream neoclassical economics, then what should replace it?

Many theories. In the social sciences and humanities there often are multiple theories to explain the same phenomenon or to make predictions about the same kind of thing, and that is a good thing. If you want to make informed decisions then you need reliable predictions of their effects. A single prediction by a theory with a miserable track record is not a good basis for informed decisions, but if many different theories (that all have made successful predictions before) point in the same direction, then you have something you can work with. Furthermore, science can only advance if multiple theories can “compete” in tests and predictions, and if there is an open inquiry into why some theories fail and others succeed. A future economics must be pluralist.

On a side note, I’m not an economist so I won’t contribute to such a future economics and I don’t have anything useful to say about what kind of theories and/or approaches are good candidates, except that I think that the foregoing makes sufficiently clear that economic theory cannot be built on a model of individual behavior – the scale of economic theory must be social. Nevertheless, I used to be an economic geographer (before (re-) turning to philosophy), and I still have a certain fondness for economic theories that deal with multiple countries or regions (such as Frank Graham’s improvement of Ricardo theory of comparative advantage – see this post), as well as for (temporal) simulations that take complexity into account (and I use such simulations for some of my philosophical research), so if I ever return to my old discipline or something like it, then it would probably be with some kind of model that combines these interests. But even then, it would be more like a toy model to satisfy my own curiosity than as an attempt to contribute to economics. So, this certainly isn’t an answer to the “Then what?” question.

My answer to that question is just a bunch of keywords. A future economics must be pluralist (rather than enforcing a single orthodoxy), realistic (in its assumptions), methodologically social (rather than methodologically individualist), empirical (i.e. confirmed by reality as known from history and/or revealed by experiments), and modest (in its predictions and policy advise). In other words, it must be the opposite from mainstream economics in virtually every respect. Unfortunately, I’m not optimistic about the possibility of achieving that, but that is another topic.5

If you found this article and/or other articles in this blog useful or valuable, please consider making a small financial contribution to support this blog 𝐹=𝑚𝑎 and its author. You can find 𝐹=𝑚𝑎’s Patreon page here.


  1. For a much more thorough analysis, I strongly recommend: Steve Keen’s Debunking Economics (Revised and Expanded edition; London: Zed Books, 2011).
  2. Or more accurately, the “law of diminishing marginal utility” holds that every next unit of a commodity will have a positive value in utils, but a smaller value than the previous unit.
  3. W. Gorman (1953). “Community Preference Fields”, Econometrica 21.1: 63-80.
  4. W.J. Eiteman & G.E. Guthrie (1952). “The Shape of the Average Cost Curve”, American Economic Review 42.5: 832-8.
  5. I will probably address this at some point in the “Crisis and Inertia” series in this blog, but it is also a key topic in my aforementioned book The Hegemony of Psychopathy.