Ch. 5 The Seen and the Unseen

What was stated in the previous chapter about optionality and choice has another dimension, especially as relates to the wealth-generating powers of exchange. As we have already discussed, both parties to a voluntary exchange must expect to be better off or the exchange would not happen. In other words, exchange for consumer goods must by definition create value because it increases overall want satisfaction among consumers. But such exchange also causes ripple effects through the economy’s production apparatus, since carried out exchanges signal where consumers see real value; obversely, not made exchanges signal where consumers find the value offered is insufficient – in relative terms. What this means is that the totality of exchanges affects how entrepreneurs anticipate that they will be able to satisfy consumer wants; to put it differently, it affects where and how entrepreneurs anticipate to make profits and thus the potential reward for different undertakings. Exchanges in this way effectuate adjustments to production plans as entrepreneurs revise their expectations, which changes the quantity demanded in the market and therefore affect prices of the means of production, which in turn reveal the new social valuation of the means of production and thus allow entrepreneurs with lesser ability to properly adjust their production plans.

These endogenously caused adjustments produce a productive structure that is responsive to change and, as changes happen frequently and responses take time to complete, is best thought of as a process in constant flux. While a production process itself may not be easily changed, the productive capability of the market overall thus remains responsive to the changes revealed by actions taken by consumers as well as the changes in how entrepreneurs anticipate what consumers will value at future points in time. Another way of saying this is that the market is in constant disequilibrium, that is, it is in constant search for and adaptation to bring about a better (more valuable) allocation of productive resources, rather than in stable equilibrium. Changes to the market happen both from within production, primarily because of what actors learn as a result of Smithian specialization under the division of labor, Ricardian comparative advantage, and Schumpeterian innovation, and in the form of changing consumption patterns as consumer preferences change. The latter change in response to novel offerings in the market (that is, in response to innovative product offerings or changing prices) and in response to consumer experience as well as seemingly arbitrary changes in fads and fancy.

The interconnectedness of decentralized market production, and thus the ripple effects that result from changes, suggest that the highly responsive market system can also be manipulated. For instance, by temporarily boosting demand for a certain type of product (or producer) or by prohibiting goods and services or production techniques that are valued by consumers (whether rightly or wrongly), can have a major change on what is produced. This is why it is important to properly analyze the effects of regulation on the market, and identify the ripple effects as correctly as possible. As many variables change at the same time, this is a very difficult task and it is likely impossible to perfectly predict the outcome. This does not mean that we are blind to the future and therefore ignorant of what will be. The exact state of the future is unknown and may even be unknowable in the present, but it is not unimaginable. This means that we should be careful in making predictions, but can make educated guesses of what to expect.

More specifically, we can imagine the outcome of certain actions by tracing the likely ripple effects and therefore estimate the changes in terms of shifts in emphases: if the price of apples surges, we should expect more investments in apple-growing (as higher prices should mean higher profits) and increased demand in substitute goods (pears, oranges). We cannot say, however, that if the price of apples increases by 10 % we predict the demand to go down by 8% and the demand of pears and oranges to go up by 3 % and 6 %, respectively. These numbers may be estimated using how consumers have acted in the past, but we don’t know what choices and tradeoffs consumers will face if the apple price goes up – so the numbers themselves are not in valuable as predictions or even indications of what “will” or could be. They are, in fact, closer to arbitrary, since the future is very unlikely to be an exact copy of the past.

To get an idea of what to expect from the future, we can reason about what a certain change will lead to in terms of the resultant ripple effects. It is important to be careful, however, not to fall victim to fallacious reasoning by arbitrarily omitting important variables or focusing on only one side of the issue at hand. A common such fallacy was masterfully explicated by the 19th century French economist Frédéric Bastiat in his parable of the broken window, and is therefore often referred to as the “broken window fallacy.” Bastiat here illustrates the error of following the ripple effects that do happen following a certain change – what is seen – but failing to acknowledge what would have happened – what is not seen.

As we noted above, but will now g into a little deeper following Bastiat’s reasoning, a change brings about changes that in turn bring about changes. For instance, if Adele from our previous chapters thinks it is a great idea to grow wheat on the land between her apple trees in the orchard, and anticipates that this would have no noticeable effect on her apple-growing business, then the added wheat will lower its price in the market. The miller will therefore be able to get his inputs (the wheat) at a lower price than before, which means he can offer wheat flower to Bart at a lower price – so Bart can then bake more bread or take time off to do something he gets more satisfaction from than baking. This effect is what we observe – the seen, as Bastiat has it. But there’s another effect: something that would have happened but now might not happen because of it (the opportunity cost of this new action, as it were) – the unseen. Perhaps the farmer who produced the wheat the miller ground into flour was planning to expand his business by acquiring more land or investing in machinery. As prices fall, this is no longer a viable option; it is not an opportunity for increased return, but would seem to be a loss at the present prices. What we do not get is as important as what we do get when we analyze specific changes in the market.

Bastiat’s example makes this point very clearly, which is why it is still referred to in research and scholarly debate as well as assigned reading for students.

That Which Is Seen

Bastiat tells the story of a shopkeeper whose son has happened to break a glass pane.[1] Furious, the shopkeeper notes that the window needs to be replaced. His no-good son has, through his careless actions, caused a cost that needs to be covered. The shopkeeper has lost what it will cost him to replace it, an amount that he likely had intended to not pay the glazier. But now he has to. How does Bastiat analyze this? He begins with what is readily observed and that is easily understood:

Suppose it cost six francs to repair the damage, and you say that the accident brings six francs to the glazier’s trade – that it encourages that trade to the amount of six francs – I grant it; I have not a word to say against it; you reason justly. The glazier comes, performs his task, receives his six francs, rubs his hands, and, in his heart, blesses the careless child. All this is that which is seen.

Indeed, it is easy to see that this analysis is correct: the glazier increases his income thanks to the shopkeeper’s son’s carelessness. With this income, the glazier may be able to pay his employees a higher wage, take time off to spend with his family, or invest in new tools to improve his business. So the positive effect of this trade, which of course makes both the shopkeeper (who values to have a window that is not broken) and the glazier (who happily trades a new window pane and labor for the 6 francs) better off, continues through the economy. 6 francs may not be enough to cause far-reaching ripple effects in the economy, but consider if there were more sons br9eaking windows or an earthquake that destroyed all windows in a whole town. This would have a much greater effect, and we can trace the likely impact this would have on the economy by applying sound economic reasoning. We do not, of course, know exactly what people would choose, so each step away from fixing the broken windows would be more uncertain about where the value actually ends up. Just like we saw above, the glazier can find many uses of his increased income, and whoever gets the value in the next stage depends on the glazier’s choice. To be able to follow how the value spreads therefore quickly becomes a hopeless task if we wish to make exact predictions.

Some try to calculate the total effect on the economy by using standard estimates of how much of the increased income the glazier will spend, how much of that money the person who earns it will spend, and so on. Such “multiplier” effects are commonly used when evaluating the effects of policy. Questions asked are, for instance, “what is the real effect on the economy if the government invests $20 billion in infrastructure?” or “how many jobs are created if taxes are cut by 2 %?” To come up with an answer, we must assume that we know approximately how much of increased income is spent in each stage, either as an estimate or based on historical analysis of empirical data. So if the glazier will use part of his new income to buy ice cream, the ice cream seller uses part of that new income to buy fresh milk from the farmer, the farmer in turn invests in a better breed of cows, the breeder… and so on. Say the fraction spent of any additional income is 50%, then the glazier will spend 3 francs on ice cream, the ice cream maker will pay 1.50 francs to the farmer, who pays 0.75 francs to the breeder. It is obvious, then, that the real effect on the economy as the shopkeeper buys a new window frame from the glazier for 6 francs is much in excess of 6 francs. In our example, which stops with the breeder, the 6 francs have an effect of 11.25 francs. The ripple effects through the economy therefore amplify the effect of any investment or trade, the magnitude of which is what the “multiplier” is intended to estimate.

The “multiplier” is based on the concept of economic activity, that is on the production and exchanges that are made within an economy and for economic purposes. Increased activity, as it includes both voluntary trade for mutual benefit and production of goods and services (which, if cold, are valuable), is supposed to approximate value creation and therefore is directly linked with economic growth. For this reason, the “multiplier” effect is used to support stabilization policies in line with the theories of macro economist John Maynard Keynes, which state that the government should use fiscal policy to attempt to stabilize business cycles. According to Keynesians, government should increase taxation and withdraw subsidies during booms to make sure the economy doesn’t “overheat,” and similarly invest in infrastructure improvements and education and other social goods during slumps for the purpose of increasing economic activity.

The argument is also used by some economists to suggest that everything about war is not dire. Imagine a city devastated by bombings during a time of war. It is easy to see that as soon as the bombing stops, the city will experience immense economic activity to rebuild and restore all that was broken and destroyed in the war. This is not to say, of course, that wars are necessarily good – only that there may be an economic upside following the destruction and devastation of war. Or so the argument goes.

So we can see the direct relevance of Bastiat’s parable for public policy, both in terms of fiscal policy intended to stabilize the supposedly natural “mood swings” of a market economy and as an analysis of post-war booms. But Bastiat doesn’t tell the story about the broken window to show how great it is that the shopkeeper’s son broke the window. No, he uses it to illustrate that focusing only on the seen – that is, the effect of the shopkeeper needing to get a new window – is in fact a fallacy. There is more to the story that is essential for the analysis: what is not seen.

That Which Is Not Seen

Bastiat’s point is that a proper economic analysis of the broken window, and therefore the analysis of the real effect of breaking the window, needs to take into account the alternative outcomes. In other words, Bastiat argues that there is an opportunity cost to the broken window. Were this not the case, then we would a lot better off by smashing people’s windows. In fact, considering the analysis of post-war booms – wouldn’t occasional wars be a really good idea? At least from the point of view of economic growth, bombing stuff would increase economic activity, set the “wheels in motion,” and thus create both jobs and value. Not so, says Bastiat, because we haven’t considered the opportunity cost – and that’s the proper way of figuring out the real economic effect. Continuing the parable of the broken window, he addresses the issue of opportunity cost as follows.

[If] you come to the conclusion, as is too often the case, that it is a good thing to break windows, that it causes money to circulate, and that the encouragement of industry in general will be the result of it, you will oblige me to call out, “Stop there! Your theory is confined to that which is seen; it takes no account of that which is not seen.”

It is not seen that as our shopkeeper has spent six francs upon one thing, he cannot spend them upon another. It is not seen that if he had not had a window to replace, he would, perhaps, have replaced his old shoes, or added another book to his library. In short, he would have employed his six francs in some way, which this accident has prevented.

Indeed, the opportunity cost to breaking the window (or, really, to replace the broken one) is the other value that is foregone by this action. This is the reason why it makes no sense to smash windows to create income for glaziers – it doesn’t make society better off. It makes the glazier and whoever the glazier then trades with better off, but at the expense of whoever would have received the income. Yes, at their expense. But how does this make sense – haven’t we already argued that the market is not a zero-sum game?

Yes, we have. But imagine the alternative scenario where the shopkeeper’s son had not happened to break the window. Then the shopkeeper would have a whole window and the 6 francs. If he would have used that money to buy shoes, then he would have a window and the shoes, and we would have the “multiplier” work its way through the economy through the shoemaker rather than the glazier. The major difference, of course, is the window: the multiplier effect, assuming we use a standard fraction for consumption/investment, is the same for the 6 francs whether or not the window is broken. The difference between the scenarios, from the point of view of the economic organism, is the window.

The difference is greater if we consider the individual level rather than the system as a whole. The shopkeeper has reason to be furious, because he just lost 6 francs because of his son’s careless action. Had the son behaved, the shopkeeper would have a window and the shoes, but because of the broken window the 6 francs are spent to undo the damage. Destruction, of course, is not a means to become rich. This suggests an answer to the implied question about war above: is war beneficial? No, of course not. Wars destroy, and destruction is a loss. If we for a moment disregard the suffering and death experienced by people affected by war, the economic effect of war may be a post-war boom through increased economic activity and therefore increasing GDP statistics (economic growth). But if we apply Bastiat’s lesson on wars, we immediately realize that there was value in the form of houses, roads, infrastructure, and supply chains that were destroyed in the war. The reason we see increased economic activity is that people have lost their homes and need to rebuild them quickly to have shelter, so they might work day and night for a while just to restore what they used to have.

What we don’t see, of course, is what these people would have done had they instead kept their homes and if the infrastructure and supply chains were intact. They may not have worked as many hours, simply because they wouldn’t need to, but they would start from a much higher level in terms of prosperity. So whereas the economic activity after a destructive war may increase, it increases primarily because the value that has been lost must be restored. It is not actual value creation, but value restoration. All this work would not be needed had the war not destroyed the value that was already created. This means they would have more options had the value not been destroyed: had they not lost their homes, they would have plenty of optionality because the opportunity cost would be relatively high for many alternatives. But since they are without shelter, the value of any other action than rebuilding their homes is so much lower that it makes no sense to even consider it. The need for shelter is so pressing that leisure or even sleep is not an option, so they may choose to work day and night to restore what was destroyed.

Note that the take-away here is that there is an opportunity cost to economic activity as well. Had there not been destructive bombing of their city, the inhabitants would be relatively richer, they would be presented with several alternatives of similar value – and the value of leisure may be one of them. At some point, of course, we are satisfied with what we have accomplished and value time off higher than more time working. This is also an important tradeoff, and therefore relevant to our discussion on choice and optionality – and the cost of choosing, or opportunity cost. But with your home flattened by bombs, the value of restoring it is so much higher, relatively speaking, than leisure that most would not even consider it. The relative value of leisure is very low when you and your family have no place to live.

Destruction and Optionality

As we have seen above, and that we learned with the help of Bastiat’s discussion on the seen and unseen, destruction has different implications depending on one’s level of analysis – and where one looks. For the economic system, the effect of the broken window is approximately the value of the window. The “multiplier” effect acts in both the seen and the unseen, and unless we know that destroying the window will lead to a much higher fraction of consumed/reinvested income in the chain of actions that begin with the glazier than the chain of actions that begin with the shoemaker, then both effects are approximately the same. Of course, for this to be the case, we must also assume that the replaced window pane is sold at the same cost as the shoes. Nevertheless, it would be erroneous to assume that there is a “multiplier” effect on one side but not the other, so even if they are not exactly the same they will to some extent balance out.

But if we instead look at the individuals involved, and therefore focus on each person, then it becomes obvious that the shopkeeper’s son has cost his father 6 francs by breaking the window. This is a loss that the shopkeeper will have to cover by either accepting lower profits, raising prices, or cutting costs in his business. In Bastiat’s example, the shopkeeper is assumed to forego part of his profits and, consequently, what he intended to use that profit for: purchasing new shoes. Perhaps those shoes were for the son, who has now by acting carelessly indirectly caused a loss upon himself by not getting the shoes he was promised or hoped for.

Yet this is only the direct effect on the individual level. We must also consider the indirect effects. As there are indeed similar “multiplier” effects that arise from paying 6 francs to either the glazier or shoemaker, very different people get this money. If you are the shoemaker and I am the glazier, whether the shopkeeper’s son breaks the window makes a huge difference to us. In this case, it is the difference of 6 francs’ worth of sales: either you sell the shoes (because the shopkeeper has a window without holes in it) or I sell the window pane (because the son has managed to force a rock through the one the shopkeeper had). Either you get the extra income or I do, which of course affects our choices. Without the extra income, I (the glazier) wouldn’t buy ice cream, which means the ice cream maker might not buy the better milk from the farmer, which means the farmer might not invest in a better breed of cows, and so on. Of course, this chain of events is initiated because the shopkeeper’s son destroys the window and his father therefore loses 6 francs. In the original situation, before the son breaks the window, the father values the six francs more than a window (since he has no windows that need to be replaced) but less than a pair of shoes. After the window is broken, however, his preferences have changed. Or perhaps we should say: they’ve been forcefully rearranged, since he’s lost a value that he enjoyed. After the window is broken, replacing it becomes necessary or much more urgent and it is therefore worth more to him than the shoes. As he only has the one 6 francs and cannot buy both the windows and the shoes, he might no longer value the shoes more than the 6 francs (for the simple reason that he needs the window, and knows that he can get one for 6 francs).

So we see how the preferences held by economic actors change all the time, even as a result of rather banal things like a broken window. This is what entrepreneurs have to deal with, and why they cannot do better than trying to imagine the future and estimate whether there will be – for wannabe glaziers – enough broken windows to profit from such an endeavor. The shopkeeper’s preferences changed quite dramatically, at least from the point of view of the glazier or shoemaker, as a result of the boy’s rock throwing.

The alternative to a broken window should seem a lot better to the observer. Not only is there a fully functional window in the shop, but the shopkeeper buys a pair of shoes as well. So we get the “multiplier” effect as in the previous example but our starting point is at higher wants satisfaction: with a whole window. The chain of transactions following the shoe purchase could therefore seem much more positive and beneficial, and in a sense this is the case. But it is the case only because we focus on the window. The “multiplier” is, as we noted above, approximately the same. But, of course, the revenues will be received by completely different people. The shoe maker may buy additional or better leather from the hunter, who might invest in a better trap to catch more beavers, and the trap maker might use his additional income to buy ice cream. So maybe the ice cream maker is a little worse off in this particular chain as compared to the other one, but he still makes a sale. Yet for the trap maker, the hunter, and the shoe maker it makes a difference whether the shopkeeper’s son throws rocks.

We cannot, of course, blame the glazier for wanting to sell window panes just like we cannot blame the ice cream maker for wanting to sell ice cream. As long as the glazier doesn’t pay the shopkeeper’s son to break the window, the exchange between the shopkeeper and the glazier is voluntary and for mutual benefit: there is nothing fishy going on, no fraud or theft. So there are no moral implications of the “multiplier” effect and therefore of the ripple effects through an economy. But it does affect what choices people are able to make, and who are able to make them. This is why it is important to understand the “flow” of goods and values through an economy. Without understanding that “one thing leads to another,” we cannot trace or assess the effect that choices and exogenous forces have on an economy – because we don’t really understand how the market works. We need to look at both the seen and the unseen, and understand the overall process within the market.

[1] This section relies on and summarizes the argument originally expressed in Frédéric Bastiat’s essay “That Which Is Seen, and That Which Is Not Seen” (Ce qu’on voit et ce qu’on ne voit pas) from 1850. The reader is recommended to read the full argument in the original essay, which is available on many sites on the world wide web.