Sticky prices, leverage, and Pascal’s wager

In Keynesian / quasi-monetarist of explanations of depression, sticky prices play an essential role. If prices were not sticky, a deficiency of expenditure would just lead to a reduction of the price level, and nothing very bad would happen. There are (at least) two channels by which sticky prices can harm production:

  1. Sticky relative prices distort patterns of economic activity, preventing the economy from achieving the optimal level of production. Following a sharp change in nominal expenditure, the sluggishness with which some prices adjust leaves activity badly distorted, and so observed real production falls relative to the expenditure-stabilized trend.
  2. Sticky absolute prices allow changes in nominal expenditure to affect levels of economic activity more directly. Suppose we set all relative prices correctly, and then fix them in stone. Now, if we scale up the willingness of all agents to part with money, we might not observe a decline in aggregate production, but markets would fail to clear and we would observe shortages. If we scale down aggregate expenditure, we would observe a glut of capacity and a fall in production (as measured by transactions).

I’m interested here in the second channel. [1] Except under politically imposed price controls, we rarely observe what absolute price-stickiness would predict in an expenditure boom — production at capacity but shortages at offered prices. The relevant case is asymmetrical. Absolute prices adjust upward easily, but they are “sticky downward”. They do not fall.

A while back I had a post that described the price of extinguishing old debt as “the stickiest price”. After a wonderful comment exchange with Nick Rowe and others, we came, I think, to some agreement that sticky nominal debt contracts were both like and unlike sticky goods prices in important ways. However, I’ve recently come to think that, besides the direct but distinct distortions associated with rigid nominal debt, indebtedness might be an important source of downward stickiness in the prices of goods and services.

The argument is a form of Pascal’s wager. Suppose that I own a firm which generally operates at capacity. The firm is leveraged in the expectation of achieving a certain level of nominal income, out of which my debt will be serviced. Should I fail to service my debt, I will face outcomes that are very dire. Perhaps my firm will be out of business, perhaps I will have to surrender the firm to creditors. Perhaps I’ll manage to squeak by after a very radical downsizing that allows me to service my debts but destroys the long-term value of the firm. Let’s refer to any of these catastrophes as “bankruptcy”.

Suppose there is a shock to nominal demand, and people become less willing to part with money. I have two choices. I can cut prices to maintain my expected volume of sales, or I can leave prices alone. In the first case, I condemn myself to bankruptcy with certainty. I was already operating at capacity, so there is no hope that an increase in volume will save my bacon if I reduce prices.

If I do not cut my prices, my expected level of sales will fall due to the recession. On average, I will still be bankrupt. But I could get lucky. In any economic environment, sales are a fickle random variable. It is possible, if I stick with my old prices, that sales will prove robust despite the dowturn. So the rational thing for me to do is to refuse to adjust my prices and hope for the best.

Now this is a perverse outcome, from an economic perspective. Considered without regard to financing, my firm fails to maximize expected profits by failing to adjust its pricing. It instead maximizes the value of the right tail of the profit distribution, because as the owner of a leveraged firm, the right tail of the distribution is all that I have claim to. Not reducing prices is a form of screwing creditors, but I don’t care. As the owner of a highly leveraged firm operating near capacity, I will be disinclined to reduce prices.

This tale of an overleveraged entrepreneur would be insignificant, if it were an idiosyncratic occurrence. One overextended entrepreneur might refuse, but her less leveraged competitors would cut prices, and observed market prices would fall. But suppose our entrepreneur is in an industry where intense leverage is the norm. As Hyman Minsky famously pointed out, if an industry is competitive and at least some players are not foresighted about risk, levering up in good times ceases to be optional. More levered firms gain a cost-of-capital advantage that permits them to undercut financially conservative rivals over what may be prolonged periods of tranquility. So we might expect to competitive forces to drive whole industries into similar capital structures. And empirically we do find this — firms in general choose wide varieties of capital structures, but within industries, capital structures are more alike.

In highly leveraged industries then, we’d expect downward price stickiness. Following a negative shock to nominal expenditures, we would observe production losses, but not in the form of evenly distributed cutbacks. Instead, some firms would seem to thrive despite the weather, while others are forced into bankruptcy. Perhaps the firms that survive would be the “best” firms, and certainly differences in quality and ex ante leverage would affect the distribution of outcomes. But even among perfectly identical firms, if the distribution of sales is stochastic, we’d expect “consolidation” to occur. Firms that are lucky early in a depression survive. It gets easier to stay lucky as time goes on. The failure of competitors eliminates supply, helping to support your sticky price (which becomes less sticky as you retain earnings to delever).

From a macroeconomic perspective, this account suggests that, even putting aside systemic fragilities introduced by cascading bankruptcies and financial accelerators running in reverse, financial leverage leaves an economy vulnerable to depression through a price rigidity channel. This strikes me as relevant to our current situation. Policymakers have effectively guaranteed the debt of highly interconnected borrowers and successfully eliminated the threat of cascading defaults. But if my account is correct, reducing leverage at “Main Street” firms may be at least as important as ensuring the stability of interconnected financials. Policymakers have put tremendous effort into ensuring the continuous availability of credit to firms that wish to expand. But promoting debt-financed expansion may be self-defeating, if it reduces the ability of the economy to adapt to fluctuations in nominal expenditure by making prices sticky. [2]


Notes:

  1. This will be an intramural point. Footnote 2 is more interesting.

    I think that explanations of the business cycle based on relative price stickiness ought not be classified as Keynesian or monetarist at all. Relative price stickiness is really a recalculation story of the sort favored by Arnold Kling (and to which I am also sympathetic). If you think of markets as calculators of equilibria, and that after a large shock computing a new equilibrium takes time, then there must be some sort of friction that prevents the computation from being instantaneous. Sticky prices offer one plausible source of friction.

    I won’t speak for Kling, but I think that some proponents of recalculation-ish theories would object to this characterization, because they view economic calculation as something deeper than a rejiggering of relative prices. They’d focus instead on inspired entrepreneurship, creative destruction, entirely new practices and products. I agree with all that, but when making up models we do have to reduce a variegated and multicolored world to symbols, and modeling recalculation as a laborious price vector computation is more expressive than it first appears. For example, we can imagine a space of potential new products whose prices begin at infinity and adjust downward with difficulty, as we learn by doing or as a stochastic function of entrepreneur effort.

    It might seem odd to expel relative-price-stickiness-based explanations from the Keynesian pantheon. After all, aren’t New Keynesian models almost defined by incorporation of relative price stickiness? Well yes, and they use relative price-stickiness to achieve monetary non-neutrality. However, at risk of stepping on toes (which is really not my intention), I think that by construction New Keynesian models are poorly suited to the analysis of extreme business cycles. New Keynesian models, like their Real Business Cycle progenitors, are usually characterized in log-linear approximation around a long-run equilibrium. Even if we believe the models to be perfectly correct, the conclusions we draw from log-linear approximations become less and less reliable as variables depart from equilibrium values. Log linearized models, if they are useful, are useful at describing near-equilibrium dynamics. If extreme business cycles involve severe departures from the presumed equilibria, or worse yet, if they involve multiple equilibria so that the economy might be durably drawn away from the presumed steady state, common New Keynesian models just aren’t helpful. To invoke Hyman Minsky again (via Steve Keen), if you want to answer questions about extreme business cycles

    it is necessary to have an economic theory which makes great depressions one of the possible states in which our… economy can find itself.
    Perhaps I am overharsh. I am certainly no expert on New Keynesian macro, and I’d be delighted to learn that I am wrong. But the New Keynesian models I have encountered simply don’t live up to Minsky’s very sensible criterion. Monetarist and Old Keynesian models, though hydraulic and not-microfounded they may be, incorporate in their design the possibility of durable and severe depressions. [ back ]

  2. The story I’ve told isn’t particularly novel. It complements commonplace accounts of why unemployment occurs in depressions. In theory, firms could simply cut employee hours and wages rather than fire people in response to a downturn. But they don’t. Employment adjusts on the “extensive” margin of layoffs much less than it adjusts on the “intensive” margin of reducing work or pay. This is often attributed to employee morale. It is better, the story goes, to have a small workforce of happy people than a big workforce of bitter people.

    But consider the same situation from employees’ perspective. Families are often highly leveraged. Even when they are not explicitly in debt, many families take on operating leverage. That is, for many families, the fixed costs of ordinary living (e.g. rent, day care, food) approach their total household income. With high leverage (whether explicit debt or operating leverage), a reduction of wages and hours translates quickly to financial and personal crisis, and ultimately to a disruptive reorganization of living arrangements. A cutback in wages and hours may leave families unable to afford their mortgage or their rent, and force a move to less desirable digs or “doubling up” with family, usually after a lot of confusion and juggling and bills and shame and collection agencies.

    Getting fired will do all that too, of course. But if firms hold wages steady and cut back by firing workers, then some workers will avoid the reorganization entirely. When firms cut wages or hours, all highly leveraged workers must reorganize. If the severity of crisis is not so different for those who are fired versus those who see pay cuts (a big if), then workers rationally prefer a layoff lottery to universal pay cuts. Their reasoning would be identical to that of leveraged firms who hold prices steady and take their chances.

    So firms find layoff lotteries to be better ex ante, because employees prefer them, and ex post because only the happy winners remain with the firm. A perennial suggestion among reformers is that we substitute some form of work-sharing for cyclical unemployment, so that the burden of downturns is evenly shared instead of falling disproportionately on an unlucky few. That sort of reform only makes sense when household leverage is generally modest. [ back ]

Update History:

  • 14-May-2011, 3:15 a.m. EDT: Fixed some typos and awkward sentences in Footnote 1. No substantive changes.
 
 

72 Responses to “Sticky prices, leverage, and Pascal’s wager”

  1. Luis Enrique writes:

    I think it’s a great idea to think about price stickiness in the context of negative shocks, rather than trying to theorise price stickiness in general, in normal times.

    How does your idea about the role of leverage differ from the idea that firms might not be able to reduce prices in recessions simply because input suppliers (including labour) are not reducing prices, so if they did reduce prices they’d start making a loss and pretty soon face bankruptcy whether highly leveraged or not?

  2. JKH writes:

    SRW,

    That makes sense.

    Price/revenue deflation increases the risk of loss.

    The risk role of equity capital is to absorb loss.

    Less equity capital means less ability to absorb loss (without bankruptcy).

    Therefore, leveraged firms are less willing to offer deflated prices.

    Of course, the effect of capital structure is a continuum rather than binary. A firm financed entirely by equity is only slightly less exposed to bankruptcy risk than the same firm with a small amount of leverage. In the same sense, the propensity to offer deflated prices is also a continuum – it doesn’t suddenly jump when equity financing hits 100 per cent.

    I’m trying to understand this thought:

    “Now, if we scale up the willingness of all agents to part with money, we might not observe a decline in aggregate production …”

    And I’m not sure about the Pascal wager translation.

  3. Luis Enrique writes:

    [excusing me for indulging in thinking aloud here …] is there some mileage in the idea that the sequencing of price cuts matters?

    for example, if a negative shock hits, and the people at the bottom of the inputs/production chain (land owners, resource owners, labour suppliers, and maybe debt owners) immediately cut their prices, then those above them (intermediate goods producers, final goods producers, retailers) could at least afford to cut their prices without risking bankruptcy.

    But if everybody down the chain tries to hold up prices waiting for those at the top of the chain to cut prices and then to turn round and try to negotiate lower prices from suppliers, do we get stickiness that way? Of course suppliers, if they hold up prices, will find themselves facing quantity shortfalls so they question is why don’t they cut prices to deal with that, but if they have suppliers (and debt owners) standing behind them holding prices up, they can’t afford to for the same reason the agents above them can’t. So who is at the bottom of the chain holding things up – debt owners, land owners (maybe?) and workers?

    In your story, if the first thing that happened when a shock hits is that debt owners worked out they were going to face defaults from firms going under, and that if they cut the nominal price of debt (forgave some debt) then defaults woudl be lower and the real value of their (no nominally lower) loan repayments would be maintained by subsequent price cuts, things would work out differently. Thinking of Steve Keen’s ideas – does this mean that high debt levels cause recessions, or merely that they create price stickiness when they happen?

    I’ll probably realize this is a dumb idea after hitting submit. Could downward price stickiness in recessions be a co-ordination problem?

    Of course, this idea is hard to square with the fact (I think) that corporate profits have been high throughout the recession, so firms ought to have been able to cut profits without going anywhere near bankruptcy.

  4. Luis Enrique writes:

    rats – should be “now nominally lower” not “no” in brackets of 3rd para above.

    (also I realise that the idea firms cannot cut prices because wages a sticky is not a new insight!)

  5. David Merkel writes:

    This is similar to my contention that depressionary conditions arise from a rise in debt based finance, particularly when debts are used to finance other debts, at high levels of leverage, as is true with most financial institutions. Debts become particularly sticky when ordinary liquidation of debts would cause additional institutions to become insolvent, perhaps creating a cascade.

    The solution is banking system reform, which will be fought by the banks, because it will lower their profitability and influence in the economy. This is my most recent short post on the topic:

    http://alephblog.com/2011/05/05/on-systemic-risk/

  6. john writes:

    As a small business owner who is part of a network of such owners in affiliated industries revolving around construction, I can attest that in fact in my industry, outside of large firms and union labor, wages have not been sticky. All of my employees and most of my peers employees have taken pay cuts. These cuts reduced operating costs to the extent that they could be leaving contracted costs like rent and debt as the largest fixed problem for the business. Prices for our services were very competitive even in the boom so there has been little leeway to reduce price. Thus debt service out of a reduced revenue flow has translated into liquidation of capital that will continue until either we get stochastically lucky or throw in the towel. A two to three year recession was our worst case plan. Obviously we screwed up on that on.

  7. Indy writes:

    The key connection I draw in my mind is between the bit about involuntary conformity to an industry-wide capital structure (and an asymmetric ratchet-effect in the direction of levering-up) and the bit in Note 2 concerning household leverage. Households are also in a kind of competition not just in the keeping-up-with-the-jones’s status-signalling conspicuous consumption, but also for “scarce resources” such as bidding up the price of homes in class-segregated neighborhoods with “good schools”.

    I’m reminded of the counter-productive rat-race argument of Elizabeth Warren’s The Two Income Trap where gains in productivity or household income by a few pioneers get eaten away in an inflated cost-of-decent-middle-class-living as more people dive into the workforce and pretty soon it’s no longer a choice if you want to live that life.

    One of the ways people people bid up prices into outrageous bubbles (including in education), of course, is by taking on increasing levels of leverage (and therefore risk and fragility to even small shocks which turn into crises). I think a lot of the economic history of the 00’s could be told in terms of this narrative of households gradually ratcheting up leverage in their “involuntarily uniform industry capital structure”.

  8. Detroit Dan writes:

    Outstanding post, on second reading. (The first time through I was unimpressed.)

    With regard to footnote 1, I believe that MMT / functional finance advocates are Post-Keynesians who reject the idea that a general equilibrium with full employment is inevitable in the long run. This paper by David Colander discusses the unfortunate significance of the Neo/New Keynesian acceptance of long-run equilibrium, and the preferable alternative of functional finance. Excerpt:

    With multiple equilibria the long run and the short run are interconnected but neither dominates the other. Short run considerations play a role in choosing which long run equilibrium we arrive at and, hence, short run policy can affect the long run equilibrium. As a practical matter, since many of the long run paths are unknowable, practical policy will often focus on the short run…

    Let me now turn to the potential theoretical role of functional finance in a multiple equilibria economy. Above I stated that functional finance’s theoretical role was to influence the equilibrium selection process of the economy…Why should the government have any role in guiding the economy to an equilibrium? … The answer: With multiple equilibria there is no presumption of global optimality of the equilibrium chosen by the market…there should be no general presumption that the private economy, given its institutions, arrives at an equilibrium preferable to one achieved with government guidance…

    If one can show empirically that government has a comparative advantage in adjusting to short-run fluctuations, then it can have a positive effect… Here enters Lerner’s and my macroeconomic externality argument. The effects of individual decisions on the aggregate level of spending and inflation are too small for individuals to take into account separately; it only makes sense for people to take those effects into account collectively. Since people have collectively organized
    through government already, the logical agent to internalize those externalities is government…

    Notice I have said ‘potential role.’ Theory will not prove or disprove a role for government; it simply provides a framework for thinking about the issue. Built into the rules of functional finance is the implicit assumption that government has a comparative advantage in achieving steady state stabilization. Keynes and Lerner gave plausible reasons why government did have a comparative advantage; they did not prove it did have a comparative advantage. That position can only come from historical observation.

    In my view, the past 60 years have shown both the limits and strengths of functional finance…

  9. Lord writes:

    Ex ante, the businessman doesn’t know whether it is a matter of relative or absolute prices but would assume it is a matter of relative prices because that would be the most common, occurring more or less continuously, and the appropriate adjustment would be to adjust volume rather than price. Even in the case of absolute prices, his relative price may outperform, so it would be unnatural for him to change his behavior initially even if non leveraged. Leverage just reinforces this. This also explains why recovery from financial crises is so difficult. Offers to increase leverage when already over leveraged with reduced demand fall on empty pockets. Instead what is needed are ways to reduce leverage.

    I agree, NK models are too symmetric and equilibrium focused to be very useful.

    (Until the last bankruptcy law, it was easy for bankrupt competitors to compete and actually have the advantage over non-bankrupt firms as their existing debt was frozen.)

  10. Luis — You are always very welcome to think aloud here. That’s what this space is for!

    So, to discuss whether reluctant input providers could be the source of price stickiness, we have to consider two things:

    1) Even holding input prices constant, we’d still expect a firm with an upward sloping marginal cost curve to reduce prices following a demand shock.

    and

    2) That said, if input prices fell, a firm can reduce prices while maintaining the ability to earn profit sufficient to service debt. So flexible input prices would, as you intuit, encourage some price flexibility despite leverage. So sticky input prices are certainly part of the problem. But it gets recursive: why are input prices inflexible?

    I want to argue that up and down the supply change, leverage broadly defined contributes to price stickiness. At the very bottom, there are workers carrying operating leverage who can’t afford their housing, health care, and day care if they take large salary hits. The firms they work for, though, are also leveraged, and have a hard time dealing with lower prices (especially given that they cannot cut wages). Those firms look for efficiencies, they cut workers and work those who remain harder, and that savings allows for some margin of flexibility to cut prices. But not so much. And so on up the supply chain.

    Your suggestion that the problem is a coordination failure is interesting, and certainly plausible. Here’s how I’d characterize it. Suppose that the full supply chain is leveraged, but each element has a certain amount of slack. If all input providers were to adjust simultaneously, each would be no less likely to suffer bankruptcy. It might be optimal for all input providers to form a grand treaty, splitting the expected profit loss in a manner that ensures each link in the chain can service its debt.

    However, arranging such a multilateral agreement might be very hard. Absent a multilateral cooperative arrangement, each participant’s best bet is to negotiate for their own best pricing. So, it becomes a kind of prisoners dilemma, where the optimal coordination plan is not a Nash equilibrium. (If I am one input supplier of many, and if other input providers as well as my own suppliers are accepting cuts, I hold tight and try to capture an expanded profit margin. But if several input providers try this, our customers won’t have slack to cut final goods prices. Knowing that, “altruistic” price cuts won’t happen, as each input provider assumes that any cut it offers will likely be split among other input providers and the parent firm without affecting final goods prices or the likelihood of bankruptcies.)

    This explanation is an extension of the story told in the post. It still suggests that leverage would create price stickiness, and that the more levered firms are, the stickier prices are likely to be. It adds a possibility, though, that some means of coordinating loss sharing could increase price flexibility if firms are not so leveraged that they go bankrupt no matter how they allocate a loss. Ultimately the degree of acceptable price reduction would be bounded by the leverage of the full supply chain, but in absolute dollar terms, that implies more flexibility than any single firm is capable of providing.

    Not a dumb idea at all, and it has interesting policy implications. Coordinating prices smacks of “cartel”, but in this case, vertical coordination along supply chains (rather than horizontal coordination among competitors) leads to better, not worse, pricing for consumers.

    The story I’m telling argues that high debt levels are a factor in recessions via a new channel. It is complementary with more conventional stories about the relationship between debt and recessions. For a Keynesian recession to happen, you need both a negative shock to expenditure (i.e. aggregate income) and price stickiness. Most stories about the relationship between debt and recessions, including Keen’s I think, argue that debt impairment is responsible for the aggregate expenditure shock. I’m arguing that it can also account, at least in part, for the price stickiness.

    I’d take measures of aggregate corporate profits during the recession with a grain of salt. As I understand it, high corporate profits have been disproportionately a function of high financial sector profits. Repairing bank balance sheets through earnings has been policy objective. Directly and indirectly, overtly and covertly, large government deficits and financial sector profitability are likely related.

    My argument suggests that even if the banking system is fully stabilized, leverage at the “leaves” of the graph of financial connectedness make prices sticky. So you’d need substantial deleveraging in nonfinancial sectors, either via repayment from high nonfinancial profits or via reorganization in bankruptcy. Eventually we get there, one way or the other. But we have endured subtrend production for some time, and might continue to endure it for some time more.

  11. anon writes:

    “I’m interested here in the second channel. [1] Except under politically imposed price controls, we rarely observe what absolute price-stickiness would predict in an expenditure boom — production at capacity but shortages at offered prices. The relevant case is asymmetrical. Absolute prices adjust upward easily, but they are “sticky downward”. They do not fall.”

    Are you taking imperfect competition into account here? Most goods are sold with a markup over marginal cost, so production at capacity is quite rare. Upward price stickiness would tend to reduce the firm’s markup and push price closer to MC. This would tend to cause an economic expansion, as it implies a second-order increase in profits for the individual firm, but a first-order increase in real wages and profits at all other firms (via the effect on the price level).

    You can see this as either an expansion in aggregate demand, or as something which pushes the economy closer to the perfect-competition ideal, by setting price closer to marginal cost.

  12. anon writes:

    Edit: for “second-order increase” above, read “second-order decrease”. My bad.

    As for why we never see production at capacity during an economic expansion, there may be a number of reasons: (1) Perhaps the industries which work closest to their capacity limits also have more flexible prices, so they raise their prices and never hit capacity constraints. (2) If capacity is constrained, expanded supply cannot push the price level lower by the above mechanism. The central bank then sees increasing inflation pressures and tightens money, which halts the economic expansion.

  13. JKH — I fully agree it’s about continua, not binary choices. An all equity firm’s incentives are to do what conventional economics prescribes following a demand shock, which (for firms under monopolistic competition with increasing marginal costs) means reducing price and accepting smaller profits and expected sales.

    If you buy my story, a leveraged firm will have an extra constraint: If it’s debt service cost is K hold prices to at least K / Qmax, so that if it gets lucky, it will be able to service its debt. Exactly what price it chooses will be a function of its estimated distribution of uncertain sales, but K / Qmax is a minimum bound. K is a measure of leverage, so K / Qmax is proportionate to leverage. If a firm is levered so that K / Qmax is greater than price an unlevered firm would choose after a demand shock, relative price inflexibility is guaranteed.

    Even if the firm is not that levered or the capacity constraint is not that hard, as long as sales are uncertain — if the distribution of sales conditional on price is broad (assuming a symmetrical distribution), equityholders will choose higher prices at the margin than an unlevered firm would choose to maximize the fraction of the distribution of outcomes outside the bankruptcy zone. Probably the breadth of the distribution of sales is itself dependent on price (at a high enough price, expected sales approach zero with very little variance), so the marginal effect will depend upon how quickly the price uncertainty collapses. I think (but I am not certain, I really should math it out) that the marginal effect on price of any amount of leverage would be weakly positive. That is, a firm with any amount of leverage would choose a price no smaller, and perhaps larger, than an unlevered firm would choose.

    But all this is an elaborate way of agreeing with you. Price inflexibility — the gap between the price a levered firm chooses and what an unlevered firm would choose — should be a pretty continuous function of leverage, I think, as long as we assume that the distribution of sales around its mean changes smoothly with price. The case I described — a firm with fixed capacity levered to the point where debt service required sales at full capacity at the pre-shock price — is an extreme case intended to make the effect very clear.

    Re “Now, if we scale up the willingness of all agents to part with money, we might not observe a decline in aggregate production …”

    That’s a case of me writing defensively, sacrificing clarity to avoid being technically incorrect.

    Initially I had written something like “Now, if we scale up the willingness of all agents to part with money, we would not observe a decline in aggregate production”, which seems natural. But then it occurred to me that demand needn’t scale linearly with income, so increasing willingness to spend might invalidate the correct relative prices I had just fixed in stone. If the shock does invalidate relative prices, we’ll produce a subobtimal mix of goods and the real value of production might be less than it would have been without the shock, even though everybody is spending more money. It might depend on how purchases were rationed: Since they would not be rationed by price, maybe some people would consume luxuriously while others would starve, leading to a reduction of net welfare. It might also depend upon the reason why expenditures increase — if the risk preferences of agents change so that maintaining financial equity is less valuable, maybe that shifts preferences and optimal relative prices.

    Anyway, I realized I could not safely say that an expenditure boom would surely be attended by production at capacity, so I chose the mealymouthed phraseology that you flagged.

    Pascal’s wager has been interpreted in a variety of ways, but one popular interpretation is this: Suppose you have to wager whether or not God exists. If God does exist, and you believe, you go to Heaven. If God does exist and you don’t believe, you go to Hell. If God doesn’t exist, you go nowhere no matter what you believe.

    Under these “payoffs”, it always makes sense to believe in God, regardless of any objective probability that the belief is correct. If you believe and you’re wrong, you lose nothing. But if you disbelieve and you’re wrong, you lose everything.

    For a leveraged firm, your choice is that you can have faith that sales will hold up at a price high enough to service your debt, or you can disbelieve.

    If you believe and you’re wrong, you’re outcome is exactly the same as if you disbelieved and dropped your price, you lose nothing.

    If you disbelieve and you’re wrong, the valuable firm you might have kept goes bust, you unnecessarily lose everything.

    Regardless of the objective probabilities, you have nothing to lose and everything to gain by believing.

    I think the interpretation I gave of Pascal’s wager is considered vulgar. Rather than Heaven and Hell in the afterlife, I’ve been told that Pascal’s wager is really about the quality of this life, that one who believes in God lives more fully in this world and is no worse in death for being wrong. In any case, the structure of the argument is the same: You should believe a thing if the implications of believing will never be bad for you and, with nonzero probability, might be good for you.

  14. David — Unsurprisingly, I like your piece on systemic risk, and mostly agree with it.

    I’m trying to say something complementary, but a bit different here, in that I’m arguing that even if you eliminate the systemic risk associated with leverage (and exposure size, and interconnected, etc) inside the banking system, leverage at “Main Street” firms has a secondary effect that can bite in a downturn. The two points are very complementary, because what drives a systemic slowdown in expenditures is usually some pathology of the financial system (defined to include the Fed as well), so eliminating the systemic risk that you describe might render the effect that I’m describing innocuous. Also, as a side effect of the changes you suggest (and I support), businesses in general would end up with fewer extreme-leverage capital structures, as the financial system would become discriminating rather than psychotically pushy about extending loans. The effect I’m describing probably wouldn’t matter very much in a world in which your reforms had been adopted.

    But in a world in which leverage is overextended and we do experience credit cycles, pathologies and fragilities inside the financial system are, I claim, only part of the problem. Even if we fully recapitalize those from the goodness of our taxpaying little hearts, the aftermath of a credit boom may leave Main Street firms and also households with leverage-skewed incentives that may be contribute to a recession until the debt is repaid (and/or operating leverage is reduced).

  15. John — So, your account is interesting, because I think economists would find my story more controversial with respect to firms than with respect in workers. It is pretty common to argue that wages are sticky because workers inability to deal with nominal pay cuts is so intense, they effectively “prefer” a layoff lottery. It’s not as common to argue that firms set prices stickily because of the debt they need to service.

    In your experience, wages are not sticky, workers have been able to deal with pay cuts without their being so disruptive that they resisted at all costs. But at the firm level, it sounds like you are experiencing what I’m trying to describe, a cost structure set in the boom that doesn’t permit you to reduce prices and survive, so all you can do is stick with your prices and hope you get enough business to keep you afloat.

    It’s always nice to have ones little blog theories sort-of confirmed by people’s experience, but in this case, I am very sorry to hear about what you are experiencing. You aren’t alone in having screwed up, and in a boom, and even with the benefit of hindsight, it might have been impossible to remain competitive iif you tried to operate so conservatively that you could ride out a five year contraction. Of course people have to take responsibility for their businesses, but we call it “systemic” risk for a reason. At least in part, your business was caught up in currents that would have been difficult or impossible to resist. So I hope you cut yourself some slack. And I hope you do get lucky before you have to throw in the towel.

  16. Indy — I agree with pretty much everything you’re saying. No fun.

    The Elizabeth Warren Two Income Trap connection was intentional. One of the links in Footnote 2 is to the isbn.nu page for that book.

  17. Detroit Dan — Your impression of the post was probably more accurate the first time through.

    I enjoyed and and am very sympathetic to the excerpt from the Colander paper. I look forward to reading the whole thing.

  18. Lord — The thing about leverage is that it’s primary effect (if you buy my reasoning at least in outline) is on absolute prices. A business must ensure its pricing is such that, if business is good, it can service its debt. That sets limits on the prices it can choose, regardless of how the prices of other goods and services in the economy move. Of course, by putting a floor under the absolute price, leverage also ensures that relative prices get misaligned in a downturn (when prices in less levered industries are deflating while prices in highly levered industries stay stuck.)

    The point about bankrupt competitors is important. Bankruptcies can have do, totally opposite effects, depending on how you expect them to work out. If overleveraged firms liquidate or if they are acquired, that reduces capacity and supports high prices for “lucky” survivors. We observe “consolidation” — skillful and/or lucky firms win both market share and pricing power as competitors drop out.

    However, if firms reorganize, bankrupt firms effectively lose their leverage and can suddenly afford to reduce prices. This undercuts the ability of survivors to hold prices at levels high enough that “winners” can stay solvent. Price stickiness collapses, which forces an industry price structure that competitors of the bankrupt firm cannot support. So we observe contagious outbreaks of bankruptcy. The only way to compete is to lose the leverage, and the only way to lose the leverage is in bankruptcy court.

    Airlines, I think, have gone through this kind of cycle. They accumulate too much financial and operating leverage, struggle and squeak by for a while, until one goes bankrupt and others can’t compete until they go bankrupt as well. And so they do.

  19. anon writes:

    I’ve often wondered why firms don’t issue debt that is parameterized on some indicator of local business conditions, or conditions within the industry, or some combination of these. By way of example, many small firms cannot issue stock for credibility reasons, yet much of the variability in profits is explained by shared factors which can be measured fairly accurately. ISTM that this would effectively reduce the firms’ operating leverage.

    Of course, there are many plausible reasons why this does not work out. Perhaps the firm’s debt exposure would be too large in good times, and the firm would not be trusted to repay its debt. Or it may be that business owners tend to overconfident about business conditions, so they want to retain exposure to that kind of variability.

  20. anon — I’m thinking of firms under imperfect or monopolistic competition.

    As I described to JKH above, firms operating at capacity is an extreme case that is helpful for communicating how leverage leads to price stickiness. But it’s not actually necessary to the argument, we can relax that assumption. Holding symmetrical and constant the shape of the distribution of sales conditional on price, adjusting prices downward increases the mass of the profit distribution inside the bankruptcy regime. It is unrealistic to hold the shape of that distribution constant, but even allowing for plausible variation (the distribution of sales narrows around a mean of zero as price increases), at the margin the effect of leverage is to reduce price adjustment, prevent the price from reaching the price an unlevered firm would set.

    Your point about the dynamic effects of an expenditure boom when prices are upward-sticky is interesting. Eventually marginal cost rises to price and the willingness to produce is exhausted. If this happens to a subsector of the economy, I can see your point that everybody else’s real wages rise and that might be stimulative. Even if it is general, firms are more profitable, and as long as the price level holds, everybody’s real income rises. Still, there might be shortages and the distribution of non-price-rationed output may be weird, which might have less benign effects.

    One last thing. The question of the capacity of firms under monopolistic competition is a bit tricky. In a stylized sense, we tend to draw smooth graphs. We see P greater than a gently curving MC, which implies the firm would be willing to offer higher than expected quantities should demand appear.

    But firm’s marginal cost curves aren’t exogenous. That is, an entrepreneur can write down a business plan in which she estimates costs and demand, including an unrestricted marginal cost curve that slopes smoothly upward. She then chooses the quantity she expects to actually produce, based on marginal cost and her downward sloping demand curve, setting MC = MR. Now she determines P, the price she will sell at, reading it off the demand curve at the quantity she has chosen.

    Finally, she revises her proposed plant expenditures, based on the quantity she has anticipated as profit maximizing. She decides her estimate of the demand curve is good, so there is little chance that quantity demanded will be anywhere near the point where MC goes to P. So she cuts corners and purchases plant that cannot support all that extra capacity. Of course, tautologically, she will still produce until MC rises to P. But effectively, by cutting corners, she recasts the MC curve from one that is gently upward sloping to one that has a sharp step at the capacity limit she has chosen. In this way, it is possible for a firm under monopolistic competition, earning hand over fist at the profit-maximizing price, to be near capacity even though price is well above marginal cost. To reduce initial and fixed costs, firms may be organized so that MC rises to P very quickly above the expected level of sales.

    [Note: I edited this a bit after posting, reposting at about 8:10 a.m.]

  21. anon again — I wonder this too. It seems to me that hybrid securities of the sort you describe, that offer equity-like risk-sharing but well characterized and enforceable payouts, should be popular. If we are both right, there must be frictions preventing these products from their success.

    Your explanations are sensible. As you suggest, markets wouldn’t clear if entrepreneurs underestimate the risk they’re exposed to relative to potential funders, and risk sharing might encourage too much leverage so the net effect would be to create rather than mitigate risk.

    Some other frictions: the informational costs of deviating from convention, especially if these instruments are variable and bespoke; regulatory issues (would SEC approval be costly to obtain?); mandate restrictions that prevent institutional investors from owning what is not conventional equity or debt.

    But ultimately it seems an area ripe for financial innovation. I tried my hand at this kind of thing once here.

  22. paine writes:

    the system just needs more insurance eh ??

    a payments grid with full insurance mandated
    participate or not

    uncle makes it an option you can’t refuse by a subsidy thru premium payment participation

    ultimately all risk should be held at maximal spread
    ie the level of the whole society

    ie uncle as reinsurer of all policies

    moral hazard ??
    what lesser creditor has uncle’s ability and capacity for generosity ???
    justice has its maximal chance if uncle is the final creditor

    of course this isn’t cor[porate capitalism

    so

    forget about it

    ——————

    happy fewest job force
    is indeed corporate optimal

    the secular increase in the reserve army
    is an artifact of social marketeering
    corporate style

    example
    no wage insurance which would solve the wage cut or hour cut alternative
    to the job axe
    but if corporate operations can extract higher margins with bigger reserve armies …..

    take th apparently benign german job time account
    despite the recent job kill free swoon in output there
    that indeed the system might have contributed to
    like the rest of the hartz reforms
    this strikes me as fishy if not over the full cycle then secularly

    these job time accounts remove the over time premium and back load it as an exit tax paid directly to the exiter
    to avoid the exit tax at time of job hack corporates are incentivized to reduce hours or wage rates
    usually hours

    this seems systemically to pin hours higher and participation lower over the long haul

    as does company pay ins toward job attached benefit plans
    so long as the company pay ins are not pro rated by hours worked

    the over time premium oughta be higher if anything
    and jobs oughta pay wages with no benefits
    –uncle could scoop this bene stream up
    and set up an earned income credit system to provision the stream in part

    and of course wage theft thru recorded hours games enters here double force

    ahh so much to do so little corporate incentive to do it

    we need an october

  23. paine writes:

    “firms under monopolistic competition with increasing marginal costs”

    assume increasing marginal costs ??
    is this the magic trick here?

    what if its almost never the case particularly in a demand squeeze
    what if demand constrained firms always face declining or at least level variable costs and thus

    your operating at capacity needs to go poof

    and this ??

    ” Holding symmetrical and constant the shape of the distribution of sales conditional on price, adjusting prices downward increases the mass of the profit distribution inside the bankruptcy regime. It is unrealistic to hold the shape of that distribution constant, but even allowing for plausible variation (the distribution of sales narrows around a mean of zero as price increases), at the margin the effect of leverage is to reduce price adjustment, prevent the price from reaching the price an unlevered firm would set.”
    strikes me as a very intense tap dance
    unpack it dear sir
    in particular
    “adjusting prices downward increases the mass of the profit distribution inside the bankruptcy regime”
    not for sure if sales increase
    ie violate ..and why not..your stricture :

    “Holding symmetrical and constant the shape of the distribution of sales conditional on price”

    and with unit sales flex capacity
    we get both q and p movements at the firm level
    and alas ambiguity as to firm level revenue change eh ??
    first off the line price cut marketers often gain share here …no ??

  24. david writes:

    On Fn2: isn’t this the implicit-contract theory of wage rigidity? Or, at least, inherits all the empirical problems of said theory (namely, that job creation remains high during recessions, even the GD – just not high enough to absorb job losses; thus suggesting that costs of reorganization cannot be prohibitive).

    Applying the implicit-contract intuition to the shareholder/manager interaction, rather than just the employer/employee interaction, is new to me though.

  25. paine writes:

    “the effect of leverage is to reduce price adjustment, prevent the price from reaching the price an unlevered firm would set.”
    this key insight and by itseelf worth a post
    runs aground for equally keen reasons you nicely sight

    the competitive advantage of maximal leverage

    sunk cost are to be fogotten in operation eh

    and fixed costs have this added sunk like dynamic aspect you can’t liquidate em
    especial in market contraction conditions

    best practice utilize all fixed factors
    but the incubus of debt used to buy and build these fixed factors
    cripples the marginal pricing option
    ie dramatic price cuts
    by spoiling the market for other participants
    as well as unsettling your creditors and holders of your trade receivables
    that have other firms in their payment stream that will face dramatic unit sales drps if they don’t respond

    the flabalanche of weak participants into payment problems quickly insues

    again market failure abounds

  26. paine writes:

    the burden of more and more arbitrary assumptions
    build as you thicken your one firm decider scenario towards
    market like complexity

    btw have you ever run a firm thru these sorts of tempest ??

  27. Scott Sumner writes:

    Steve, I’m having trouble understanding why firms would not want to maximize expected profits, just because they have some debt. Firms frequently change prices, presumably because they think it will help them maximize profits. How does this change when there is a drop in AD? Even if it were true that price cuts lead to lower revenue (and that’s not at all clear), firms should want to maximize profits, not revenue.

    If the New Keynesian model can’t explain the Great Depression, that’s a plus not a minus. There’s only been one Great Depression in the US; if the NK model predicted that sort of outcome, we’d wonder why they didn’t occur more often. You’d expect an outlier like the Great Depression to be caused by multiple shocks–both real and nominal. And of cour eit was.

  28. Ken Rhodes writes:

    Steve wrote: However, if firms reorganize, bankrupt firms effectively lose their leverage and can suddenly afford to reduce prices. This undercuts the ability of survivors to hold prices at levels high enough that “winners” can stay solvent. Price stickiness collapses, which forces an industry price structure that competitors of the bankrupt firm cannot support. So we observe contagious outbreaks of bankruptcy. The only way to compete is to lose the leverage, and the only way to lose the leverage is in bankruptcy court.

    …and I think (once again): One more feather in Ford’s cap, and one more reason why I will not buy a car from GM or Chrysler. Come to think of it … how the heck did Ford survive? Like the monk in the Xerox Super Bowl commercial, I look up and whisper “It’s a miracle.”

  29. Ken Rhodes writes:

    Scott, it is definitely true that many firms frequently change prices to maximize profits. However, it is also true that other firms make their prices sticky for reasons, not only of “inertia,” but also for the P.R. value of “consistency.” Those firms frequently think of themselves as “in it for the long haul,” making decisions based on long-term planning, even if they are counter to short-term profits.

    This has frequently been stated by successful entrepreneurs as a reason they were reluctant to relinquish control to public stockholders, who tend to emphasize the most recent quarterly earnings and the most recent daily stock price fluctuation.

  30. Lord writes:

    Nowadays, the objective is often to sell the company to bondholders ahead of bankruptcy with the sophisticated making sure they are senior enough and have enough security to withstand it while the unsophisticated are wiped out. Profit is only one goal of companies or they would all leverage to the moon. There is also survival or to extract all value ahead of time so survival is not a concern. Companies would have maximized profits before; leverage shifts the equilibrium away from lowering prices.

  31. Steve Roth writes:

    Steve, as so often your insight makes me think, “Why isn’t this debt-causing-price-stickiness idea — so obvious once it’s laid out clearly — standard fare in economic models and textbooks?” (Answer: because most economists don’t understand much less model how debt works.)

    Scott Sumner: “why firms would not want to maximize expected profits, just because they have some debt”

    The decision is not between more or less profit (that would be a rather easy choice…). It’s between a range of probabilities, where the upside is marginally better, but the downside is disastrous.

    As a serial entrepreneur, I’ve had this what’s-the-potential-upside/what’s-the-potential-downside discussion with multiple business partners, innumerable times over decades. In every case, there has been an unabrogatable assumption to those discussions: Protect the Downside!

    (Contrary to the popular belief, successful entrepreneurs aren’t so much comfortable with risk as they are comfortable with *evaluating* risk. In the sense of risking the worst, I would suggest that *successful* entrepreneurs are in fact congenitally and profoundly risk-*averse.*)

    The greater-fixed-debt-cost scenario that Steve explains increases the possibility of business death if you lower prices, relative to the possibility if you hold fast on pricing. And given the demand shock (and resulting reasonable higher expectation of continuing slack demand), the marginal upside is less likely and/or smaller.

    Multiply those probabilities by the higher “value” of avoiding business death (relative to the value of marginally higher profits), and for an entrepreneur the choice is clear.

  32. Steve Roth writes:

    Oh and Steve, without going into it at length or working out all the implications, I would like to suggest that the best implementation of the “multilateral agreement” you suggest would be a small increase in the inflation rate (and reliable expectations of same), to 3 or 4 percent.

    Holders of financial assets (creditors) would undoubtedly confute price stability with inflation (expectation) stability to paint a resulting picture with dire results (and those results would indeed be born out to some extent for the value of their financial assets), but holders of real assets would undoubtedly see through and point out that confution…

  33. paine — didn’t we used to hang out together, under different names when the days were longer, over @ Brad Setser’s? i see your doggerel, now and again at different places, and i smile. welcome here.

    i’m not arguing for more insurance here, although that is an argument one could make. i wouldn’t make it, at least not very easily, because i detest the moral hazard associated with insured leverage. but i’m not intending any policy prescriptions at all here, not even tacitly (although i do off-handly criticize policies that fail to address leverage).

    i don’t know enough about German labor laws to criticize them or to laud them, although I am told that they avoid the problem I point out by having the state subsidize full sharing during job sharing. creating exit taxes for firms that downsize does seem like it might harm participation over the long term, for boringly conventional reasons. my reasoning above would suggest that even workers would dislike this, if they are leveraged to their income so that wage cuts and job losses both lead to severe crises, precisely because it encourages employers to cut wages and hours divide the world into lucky winners and sloughed off losers. but again, this all disappears if the state subsidizes the wage loss associated with wage/hour cuts.

    Re: “firms under monopolistic competition with increasing marginal costs”

    So, I’ve been a bit muddled. Let’s think it through, I’m trying to navigate between the circumstances I imagine (rightly or wrongly) as typical and the circumstances I really require for this story to work. I was imagining firms under monopolistic competition with increasing marginal costs, but do I require that? I don’t think that I do. For the story to work, I require a demand and cost structure such that

    1) an unleveraged firm would maximize expected profit by cutting prices in response to demand;

    and

    2) sales and therefore profits are uncertain at a given price, and there is a zone of bankruptcy that drives a wedge between expected profits to the enterprise and expected profits that can be appropriated by shareholders

    I think that under perfect competition, this requires an increasing marginal cost curve. Otherwise, I’d not drop prices in response to falling demand, condition one fails.

    Under monopolistic competition, though, the MC curve helps me to set a quantity, not a price. The quantity I choose falls with demand as marginal revenue intersects with MC. Playing around with math and linear downward-sloping demand curves, I can show that with constant MC, a leftward or downward or inward shift of the demand curve or a steeper downward slope implies falling expected prices, but a rotation, where the downward steepness of the curve increases without affecting the price at which zero are demanded does not. That rotation is weakly but unambiguously a reduction of demand (at any price, the quantity I want is unchanged or diminished), so there are possible demand reductions that would not provoke me to lower price. I can come up with price-increasing, weakly demand reducing rotations under monopolistic competition and increasing MC as well.

    So I think I’ll fall back a little, and just restate my precondition (1). All I need is that the cost firm and the nature of the demand shift be such that the demand shift would induce an unleveraged firm to reduce prices. That encompasses most of what we call falling demand for a firm with monopolistic competition and constant or increasing MC, and firms with increasing MC under perfect competition.

    Re the shape of the distribution: If I hold the shape constant and independent of price, then I definitely drop prices less if I am levered than if I am unlevered. The benefits to shareholders of dropping price diminish as the mass of the distribution in the bankruptcy regime increases. The marginal effect is to reduce price reduction.

    However, if I am not operating at capacity and the width of the distribution increases with falling prices, there is an ambiguity: what I lose by dropping the expected price, I might gain back through an increased probability of getting lucky. We’d have to do the math to compute the marginal effect. But even if we are not operating at capacity, if we are anywhere near capacity that will truncate the right tail of getting lucky, and tilt things towards diminished price flexibility.

    You can posit distributions of volume conditional on price where the effect I’m describing won’t occur, or would even occur in reverse. That is, if dropping prices leads to extremely high variance of sales, then a leveraged firm might be more inclined than an unleveraged firm to do so, as the lengthening right tail more than offsets the downward shift in expected (unleveraged firm) profits.

    So if we are not operating “at capacity”, I have to restrict the shape of the distribution of sales conditional on price, and we’d have to make assumptions and do math to characterize that restriction. But I think in the real world, this isn’t much of a problem. In the real world, firms do tend to operate somewhere “near” their capacity, so the right tail is truncated somewhere. Ignoring that, pretending infinite capacity, the variance of sales, while perhaps increasing as price drops, is bounded. The bankruptcy threshold, expressed as minimum sales to avoid, grows larger as firms drop prices, eating away at any benefit caused by a broadening distribution. So under pretty general and realistic conditions, I think the effect I am describing would take hold.

    No argument on the insidious attraction and crippling cost of leverage…

  34. david —

    So I don’t know so much about the implicit contract theory of wage rigidity, but from what I do understand it is a bit different. Do correct me if my understanding is misguided.

    My understanding of the implicit contract theory of wage rigidity is as follows: Employees are more risk averse than employers, and would be willing to forego a portion of their wages for insurance against cyclic unemployment. Risk neutral firms are indifferent between paying competitive wages in good times and laying people off in bad times, or accepting an appropriate insurance premium P (in the form of discounted wages) in good times and keeping people on the payroll unprofitably in bad times. If the risk aversion of employees is sufficiently large and the distribution of business cycles actuarily foreseeable, the insurance premium employees are willing to offer may exceed P, in which case it is Pareto improving for firms to offer employees insurance contracts. We don’t observe such contracts explicitly, but implicitly, firms sometimes do retain unprofitable employees through downturns, so we might posit an implicit contact.

    Empirically, as you suggest, this theory has a hard time I think, because increasingly, firms do layoff employees in downturns rather than holding them through cycles. As a theory of price rigidity, I think the theory assumes rather than explains it: the theory fixes a constant wage which is below employees’ marginal production in good time but above it in bad times, and assumes a contract in which that wage is held constant. The whole story, of course, is driven by employee risk aversion, both to job losses and to fluctuating income, so in that sense it is complementary to my story. My story is about leverage, which tends makes people more risk averse. But my story actually makes a prediction quite opposite to the implicit contract theory.

    In the implicit contract theory, people dislike being fired or wage variation, so they pay an insurance premium to avoid it. The implicit contract theory predicts bot stable employment and stable prices.

    In my story, there are no implicit contracts, and in downturns, firms will reduce labor costs. The only question is how employees will bear that: they can divide the costs up evenly, or they can let a few unfortunates bear all of the losses and have the rest bear no cost at all. My story predicts they will choose the latter. The implicit contract story predicts neither, that employees will have prepaid the costs of the downturn by working cheaply, and so firms will carry workers through.

    The implicit contract theory does explain wage rigidity, and so it’s intuitive to treat it as a foundation for sticky prices in recessions. But it doesn’t predict unemployment in recessions, which my account does.

    So I think they are pretty distinct arguments. But again, I’ve only a thumbnail acquaintance with the implicit contract theory, and perhaps I am mischaracterizing or misunderstanding it. Feel free to correct me.

  35. paine writes:

    yes i’m the quandom” pinky ”
    could there be two such lisping jack asses with an econ con slant ??

    btw i wish i had your unabashed willingness to rethink

    this was an excellent response much to chew
    my cud runneth over

    but i’m holding steady as we go on one thing

    “In the real world, firms do tend to operate somewhere “near” their capacity”

    nope not my take
    and additional units aren’t even at constant cost
    they get spit out at a falling marginal cost

    but i agree its formally frought with ambiguity
    that i’ll settle for

    to me its always a history not an orbit
    the trick to me is assume contrary to the pure shock school
    where we always fall from at least an initial moment of paradise

    my initial conditions are simply any spot within
    what dick goodwin
    after three high balls
    called the specified systems
    aperiodic forced oscillation channel

  36. paine writes:

    “You’d expect an outlier like the Great Depression to be caused by multiple shocks–both real and nominal”
    lovely specimen of the “it’s all shocks school ”

    internal drivers are verboten
    why that would indicate a fatal flaw at the heart of the “system”

    marx looms up like a nasty neptune on the far horizon

  37. beowulf writes:

    It might seem odd to expel relative-price-stickiness-based explanations from the Keynesian pantheon. After all, aren’t New Keynesian models almost defined by incorporation of relative price stickiness?

    Paul Davidson would point out the odd partis that anyone thinks New Keynesians are Keynesians. :o)

    “In chapters 17 -19 of his General Theory, Keynes explicitly demonstrated that even if a purely competitive economy with perfectly flexible money wages and prices existed (“conceding a little to the other view”), there was no automatic mechanism that could restore the full employment level of effective demand in a money using , market oriented economy… Keynes then spent the rest of chapter 19 explaining why and how a general theory analysis must look at the relationship between changes in money wages and/or prices and changes in aggregate effective demand – an analysis that, by assumption, is not relevant to either a Walrasian system or Samuelson’s neoclassical synthesis
    Keynesianism.”

    http://webcache.googleusercontent.com/search?q=cache:j3BpSXZkq10J:www.aeaweb.org/aea/2011conference/program/retrieve.php?pdfid%3D557

  38. JKH writes:

    SRW,

    Footnote on options context, ref to your response to Paine:

    View (values) equity as an option on total capital, with a strike price equal to debt.

    An unleveraged capital structure maximizes the moneyness of this option (with an effective strike price of zero). It maximizes downside risk protection – relative to a forced event of default/bankruptcy.

    Under any capital structure, leveraged or not, a reduction in prices at least contributes to a decline in the value of total capital. This is borne at first by the equity capital component, whose value shifts in proportion to its option “delta”.

    But if a reduction in prices also results in an increase in capacity utilization, the expected profit distribution widens, and the value of equity increases. Expressed in option terms, this widening is an increase in implied volatility, and the effect on equity value is its option “vega”.

    In that case, the (pricing) delta effect subtracts value, while the (capacity) vega effect increases it.

  39. Scott,

    So my claim is that leverage places a wedge between maximizing expected profit at an enterprise level and maximizing expected profit for shareholders. Since shareholders control the firm, they will maximize their own profit rather than maximize the profit of the enterprise considered independently of finance. As finance-types like to say, a leveraged firm is basically a call option on firm assets: If the assets are valuable, shareholders exercise the option, purchasing firm assets for the price of repaying the debt. If the assets are not valuable, shareholders walk away.

    In an uncertain world, the incentives of an option-holder are different from the incentives of an asset holder. If I own an asset, I benefit by increasing its expected value, and I am indifferent to volatility if I am risk neutral, or actively harmed by it if I am risk averse. If I hold a call option, however, my interests are divided. All else equal, I prefer its expected value to rise. But since I only capture good outcomes and walk away from bad outcomes, I also like volatility or right skewness in the distribution of outcomes. I would be willing to trade off some measure of expected value for a longer right tail in the distribution.

    Let’s start with a simple, very contrived example. Suppose I have a business with $1000 in assets and two strategies.

    Strategy 1: My firm can flip a biased coin, in which case it will lose $1000 with p = 0.6 or win $1000 with p = 0.4.

    Strategy 2: My firm sells a service for a certain 2% ROA.

    The expected value of Strategy 1 is negative, (0.4 x $1000) + (0.6 x -$1000) = -$200. The expected value of Strategy 2 is positive $20. An unlevered firm chooses Strategy 2.

    Now, suppose my limited liability firm is leveraged, it has borrowed $900 of its $1000. Let’s review the strategies, from shareholders’ perspective. Remember, shareholders’ maximum loss is the $100 they’ve invested.

    The expected value of Strategy 1 is negative, (0.4 x $1000) + (0.6 x -$100) = -$340. The expected value of Strategy 2 is $20. Nonrecourse leverage alters shareholders’ optimization problem, and drives a wedge between their choices and the profit maximization choices of an unleveraged firm. This is true even if the leverage is appropriately priced (see below).

    Let’s consider a slightly more complicated example, in which the firm’s problem is to choose an optimal price. Suppose I own a firm that will exist for one period. My production capacity is one perishable unit, which I will either sell or fail to sell. The probability of my selling the unit is a function of price. We’ll set the probability as follows:

    If the price is less than $1, the probability of a sale is 1.

    If the price is greater than $100, the probability of a sale is 0.

    If the price is between $1 and $100, the probability of a sale is (1- price/100).

    The cost of production is independent of price, so we set price to maximize revenue. Without loss of generality, let’s assume zero production costs, so we can talk in terms of profit.

    It’s easy to see that we won’t choose a corner solution. Consider a price of $20. Our expected profit is price x probability_of_sale, or $20 x 4/5 = $16 . That beats the $1 we’d make if we set our price at $1, or the $0 we’d make setting price=$100. So, we find an interior solution, maximizing profit = price x probability of sale = price (1 – price/100). That’s a parabola whose maximum is at $50, and our maximized expected profit is $25.

    Now suppose we are leveraged. We have to pay $10 in debt service to cover the loan, but we never pay if we’d take a loss. Instead we walk away from our limited liability firm.

    Now what does our profit function look like?

    If the price is less than $10, we earn too little to cover the debt, and so walk away and earn 0 profit.

    If the price is greater than $10, our expected profit is (price – $10)(1 – price/100)

    If the price is over $100, we earn nothing and walk away from the debt, 0 profit.

    Maximizing our leveraged profit function, we find that the optimal price is $55.

    Note that the expected value of our assets, if we set the price at $55, is $24.75, less than the optimum of $25. At $55, the expected payout to shareholders is $20.25 and creditors’ expected loss is $1. At $50, the unleveraged maximum, creditors come out even, but shareholders expected payout is only $20. Shareholders gain $0.25 from setting price = $55, while creditors lose a full $1. Rather than maximizing the value of the firm to all claimants, shareholders extract a transfer from creditors and create a deadweight loss.

    If we didn’t have limited liability — if we couldn’t walk away from the debt — we’d behave like an unleveraged firm. Here’s our profit function with unlimited liability:

    If price is less than $1, we earn (price – $10).

    If price is between $1 and $100, we earn (price – $10) (1 – price/100) + -$10 ( price / 100)

    If the price is greater than $100, we earn (price – $10).

    Again, the corner solutions are clearly nonoptimal, and if you maximize [(price – $10) (1 – price/100) + -$10 ( price / 100)], you find yourself right back at the unleveraged firm optimum of $50. Now, shareholders’ expected profit is $15 and creditors break even in expectation. No transfers are extracted.

    The “trick” here is limited liability. The ability to avoid payouts under bad outcomes drives a wedge between the ordinary economic optimum — maximize expected value of the real assets — and the leveraged firm optimum.

    In the real world, putting aside the perverse effects of government guarantees, creditors understand they are providing an option when they offer nonrecourse debt. So they charge an interest premium that is ensures that their net losses are zero. However, if we maximize shareholder profit while imposing the condition that creditors break even, if I’m doing the math right, we find that shareholders choose a price of $56.49, while creditors demand to be paid $2.98 in interest on their $10 loan. So if we assume the relevant probability distributions are known ex ante, the price mechanism can avoid perverse transfers in expected value from shareholders to creditors, but it does not correct shareholders’ incentive to set the price at values other than the unleveraged firm optimum that conventional models assume.

    Does this make sense?

  40. Scott —

    If the New Keynesian model can’t explain the Great Depression, that’s a plus not a minus. There’s only been one Great Depression in the US; if the NK model predicted that sort of outcome, we’d wonder why they didn’t occur more often. You’d expect an outlier like the Great Depression to be caused by multiple shocks–both real and nominal. And of course it was.

    I won’t disagree with you, if you’ll let me add a caveat. Per your reasoning, NK models may be good for understanding states that are not outliers which result from multiple unusual shocks. And they may well be!

    However, when we model, we have to decide what we are trying to understand, and our model should provide insight about that. No model is “true”. We are not looking for one, perfect key to all human behavior. We develop models to understand phenomena.

    If the phenomena we are trying to understand are how an economy behaves when modestly perturbed from a stable steady state, NK models may be awesome. And that may be what we’re interested in most of the time.

    But if we are trying to understand very extreme business cycles, if we are trying to characterize under what circumstances a durable Great Depression might occur and under what circumstances it would not, NK models are terrible. They typically rule out the possible of bad equilibria that do not revert to steady-state. And they are characterized as linear approximations in the neighborhood of a good-equilibrium. By construction, NK models becomes less and less reliable as we consider more severe shocks.

    So, I’m not trying to diss NK models generally (although I’ll confess to not being a great enthusiast either). They may well be great tools for understanding the economy 95% of the time.

    But when we are asking questions like, “will contractionary monetary policy under current circumstances risk a recurrence of the Great Depression?”, Minsky’s admonition binds. A model that doesn’t foresee and characterize the state “Great Depression” cannot help us answer questions about the probability of Great Depressions.

    So when we are trying to characterize extreme business cycles, NK models are not the models to use.

    How about right now? I’d say that’s a judgment call. If you want to argue that the economy is within a zone of sufficient normalcy that NK models are good to reason with, I’ll shrug my shoulders and say “maybe”. But it might be true that we are still in a moment when unusual shocks and distortions are shaping the distribution of outcomes, where we need models that can offer insight into possibilities that in more tranquil periods we can safely ignore. I’d feel better if we at least had some good models of dynamics under extreme circumstances. If nothing else, we might learn the boundaries of where NK models are good enough, and where we require exotica.

    I’m tempted to draw an analogy between NK models and Newtonian physics. Charitably evaluated, NK models might offer good approximations 90% of the time, and be excellent workhorses. That’s true of Newtonian physics. But when we want to design semiconductors or fly spaceships at great speeds, usually negligible quantum mechanical and relativistic considerations become important. And if we didn’t understand quantum mechanics and relativity, we wouldn’t be able to predict when Newtonian mechanics are a good enough approximation, and we’d periodically get ourselves into trouble.

    The best we can hope for is multiple models, each one incomplete and deficient in its own special way. I’m not saying NK models are bad. I’m just saying they are inadequate for helping us reason about very extreme business cycles.

  41. More later… I’m taking a break. Sorry!

    But thanks everybody, fantastic comments all around.

  42. Scott Sumner writes:

    Steve, Thanks for the very complete reply. I do understand the idea that debt creates a sort of principle/agent problem for firms (another side effect of our insane corporate tax code), but don’t see how that relates to nominal price stickiness. I’m sure you can create scenarios where a firm with debt would not cut price during a AD downturn, whereas the same firm w/o debt would. But I have trouble seeing how that could be anything other than a very unusual special case. In any case, I see wage stickiness as a much bigger problem than price stickiness, so I guess I’m not really a new Keynesian.

    I’m not sure exactly what sort of models are currently considered “new Keynesian.” I think the AS/AD model can explain the Great Depression, but only if you assume supply shocks play a major role after 1933. Are supply shocks part of the NK model? I believe they are.

    One point I forgot to mention is that it isn’t just deflation that causes downturns. You can have a steep recession with 4% inflation, if 12% was expected (remember 1982?) I think the sticky-wage model does much better than sticky prices in that case. In any case, you’ve got an interesting idea. As always in macro, it probably depends on the empirical importance of the problem. I’d encourage you to keep pursuing it–certainly don’t be discouraged by anything I said–this is not my area of expertise.

    Ken, I completely agree on price stickiness.

    Steve Roth, I may be wrong but I think Steve’s argument is more than just protect the downside. If you fail much of the burden is borne by debt-holders. If you succeed the upside is yours. That biases firms toward excessive risk taking, unless I’m mistaken.

  43. Dan Kervick writes:

    Well, here are some of my loose hypotheses about downward price stickiness:

    First, there are various ways in which people just aren’t rational. Competitors are stubborn. They don’t like to admit defeat. They are loathe to write off sunk costs.

    The prices firms charge are not just an offer they are making to customers in the marketplace, but a signal they are sending to their own employees in the production, sales and marketing chain. People know that price decreases in a recession are only designed as a way of standing pat or minimizing losses, not as a tool for increasing revenue. If the firm lowers prices, they are effectively saying to their employees, “We expect less of you this year.” It is in fact rational to expect less during a recession, and probably the firm’s bosses do expect less. But they don’t want to tell their employees that. They would prefer to lose money by selling less at constant prices, but keep cracking the whip over the employees, than lose money by lowering prices and keeping sales constant.

    They don’t just cut salaries across the board for various reasons. One is that they simply prefer to fire people and extract more labor from everyone that is left, and the buyer’s market for labor gives them the opportunity to do that. Also, there is a per-employee base cost of employment no matter what the wage-level is, and so one saves more money by cutting employees, not by reducing the salaries of a constant staff by an equivalent amount.

    Also, in times of crisis in a hierarchical system, the chiefs in the hierarchy are all about sending signals that reaffirm their authority and control. It’s not all about quantitative rational assessments of the best means of maximizing economic performance. The competition inside the firm becomes much more intense, power struggles abound, and simple niceness declines. The urge to humiliate and dominate finds more opportunity for expression, and their are fewer institutional checks on the expression of that urge. This is another reason the firm is reluctant to send a signal of reduced expectations by lowering prices.

  44. Ken — Good points all around.

    Your second comment makes me consider a firm like Apple, which is perhaps the ultimate counterexample to my hypothesis. Apple is not a leveraged firm. In fact, it has so much cash, you might think of the firm as negatively levered. And yet it holds prices of some products at a premium, arguably at levels inconsistent with medium-term profit maximization. Apple also holds prices of new products unusually low — new Apple products are a counterexample to my contention that we rarely see shortages of products at offered prices.

    I think you have the right explanation. Apple is interested in maximizing value of a very long horizon. It considers its “consistency” in pricing an asset, and sets its price at a level that perhaps maximizes profit when averaged over several business cycles. That means it leaves money on the table in downturns and demand spikes. It could increase profits by cutting prices in recessions, and sell new goods at a higher premium. But the firm increasingly chooses not to do that, and holds prices within product categories very steady, competing in large part by improving product quality at nearly-fixed prices. Since the firm maintains its substantial cash buffer — really, almost textbook precautionary savings, except that in textbooks firms aren’t supposed to save — it ride out business cycle fluctuations despite near-term suboptimality.

    I don’t think that Apple is unique, although it is unusual and extreme. Various luxury products that are arguably “Veblen goods” (goods whose price is taken as a signal of value) also adopt pricing strategies very different from what an economics textbook would predict, and not because they are financially levered.

    I don’t think this sort of consistent, above-the-fray, premium pricing strategy is widespread enough to contribute very much to macroeconomic price stickiness, but as they say, every little bit helps. I don’t mean to suggest that financial leverage is uniquely or even the main source of price stickiness, but I do think it might be an important contributor.

  45. Lord @ 30 — Your comment is pretty cynical, but I don’t think that you are wrong.

    Economists tend to underestimate the degree to which debt and financial arrangements can be used to engineer transfers. It’s not that they don’t understand — add asymmetric information to financial arrangements and the capacity for transfers becomes obvious. But for all kinds of reasons — you can call it optimism or corruption or faith in the quality of regulation; you can call it giving rationality the benefit of the doubt in a world where information asymmetries are unobservable — economists usually underemphasize that aspect finance.

  46. Steve R — First, thanks for the nice words!

    Inflation, or even just restoring nominal income and hoping for the best with regard to output (Hi Scott!) does address the problem. Though I think the difference is only semantic, I’d describe it less as a resolution of the coordination problem, and more an undoing of the negative shock, a restoring or ex post ratification of the nominal income expectations under which the debt was originally contracted. That sounds a little bit too innocuous: During a credit boom, it’s likely the case that debt was contracted under unreasonable income expectations, and then to stabilize output, policymakers have to ratify (or validate, as Minsky has it) those unreasonable expectations ex post.

    There are big moral hazard issues here: if unreasonable leverage will be always be validated to prevent a fall in output, rational people will herd into leverage booms. As long as one stays near the middle of the pack, one will enjoy more profitability than leverage fuddyduddies in the boom, and your debt will on average prove manageable in the bust, thanks to policy interventions. Trying to adopt a “reliable expectation” of inflation or nominal expenditure growth might not prove credible if leverage grows faster and forces policymakers’ hands.

  47. Paine @ 35 & 36 —

    I appreciate your help at provoking me to rethink!

    Re operating at capacity and falling marginal cost, I won’t try to argue. I just don’t have a good sense. My intuition is kind of as I described to anon in comment 20 (although the story I told, the sequence of events, is very artificial). I think that most firms have some pricing power and operate well below marginal cost, and they invest in some extra capacity in hopes of selling more than they expect. But I don’t think they invest in so much extra capacity, so pretty run up against constraints. I think firms do try to retain an option to expand, so that constraint capacities are rigidities that can be overcome, quickly but not instantly.

    But all of this is speculation. The effect I outline would be real under certain conditions — leverage and capacity constraints increase the effect, the degree to which price reduction widens the distribution reduces it. Capacity constraints aren’t necessary, but they limit any widening of the distribution. Whether these conditions are widespread in reality, I can’t say for certain. Intuitively I think they are, otherwise I’d not have offered the conjecture. But all I can say for sure is that if the right conditions hold, the reluctance of rational shareholders in leveraged firms to drop prices could contribute to price stickiness.

    By the way, if option Greeks aren’t Greek to you, I love JKH’s concise summary of the dynamics in comment 38 below.

    I agree with you about the “pure shock school”. Although I have learned to speak that language and sometimes fall into it, I think economists’ convention to divide the world into equilibrium and shock hides more than it helps. The world is dynamic, shocks are endogenous. They rarely come from outer space, and even when they do, outer space is part of the universe we are characterizing.

    Now when I hear “shock” in economist-ese, I can’t help of thinking about Greenspan’s “notably rare exceptions”. Wonderful, placid equilibrium is the norm, pay no attention to the calamities behind the curtain. As you say, stories in which contradictions build predictably behind an appearance of calm efficiency only to explode into crisis are reminiscent of Mr. Marx, and we must not be reminded of Mr. Marx. We can barely get away with referring to Mr. Minsky. Thank goodness we have Schumpeter. He is safe to talk about, although of course he himself was obsessed with Marx.

  48. beowolf — Thanks for the quote.

    New Keynesians are kind of like neoconservatives.

    In some sense, neoconservatives are related to conservatives. Sort of. But then, in another sense, conservative is almost precisely the opposite of what they are.

  49. JKH —

    This is a great, great way of putting things, much more clear and comprehensible than all the handwaving I was doing. Thanks a ton.

  50. Scott — My understanding might be overnarrow, but I’m using “New Keynesian” to refer to DSGE models that build on the RBC foundation, but layer price-stickiness on top to achieve monetary non-neutrality and other interesting dynamics. And also micro-founded-ness, which for some reason seems to be very important to New Keynesians. Supply shocks are certainly incorporated in the RBC/New Keynesian tradition (after all, in an RBC world, the only shocks are supply or technology shocks).

    I think of IS/LM, and AS/AD models as “Keynesian synthesis”. I’d call that “old Keynesian”, although with caveats because self-styled bearers of Keynes’ legacy squabble amongst themselves a great deal, and before there were the New Keynesians to get upset about, many of the people who are now called “post-Keynesian” strongly objected to the abstraction away from financial arrangements and institutional detail in the old Keynesian synthesis.

    This is my own taxonomy of squabbling Keynesians, and is not intended to be normative. ;)

    I’m also very interested in sticky wages. I don’t mean to suggest that financial leverage is the unique source of price rigidity, just that it might be one important source. (Also, per footnote 2 of the post, I think that sticky wages can be explained in part in terms of operating leverage among workers. But I don’t think it’s everything, as you point out, there’ve been some deep recessions associated more with a sticky growth rate in wages rather than immobile stuck wages.)

    You’ll hardly discourage me. On the contrary, I’m always delighted to find your cameo appearances in the blogosphere, and I’m flattered that you dropped by. I hope that you are enjoying your well deserved but much lamented sabbatical!

  51. Dan —

    I think those are all interesting explanations of price stickiness, subtle and quite insightful. The factors you suggest are hard to operationalize and observe from a macro-ish distance, but that they are hard to measure does not mean they are unimportant.

    Since my schtick is leverage, I want to make a point about leverage and sunk costs. Usually, we say that including sunk costs in ones analysis is a fallacy. What’s past is prologue, one must make choices looking forward.

    But to the degree that past costs are debt finance, they are not irrevocably sunk. I can unsink at least a portion of those costs by defaulting on my debt. I’ll still look forward to make my choices, but if my past decisions were poor that will be reflected looking forward as debt service insufficiently offset by income.

    So, if I can get financing on good terms and face initial costs of uncertain value, I buy myself a nice option to undo my choice to pursue a project if, incorporating the value of sunk costs, the project turns out to be a loser.

  52. […] Sticky prices, leverage, and Pascal’s wager Steve Waldman […]

  53. […] sign for the defendant, but here the jury was being very meticulous about analyzing the evidence.Sticky prices, leverage, and Pascal’s wager Steve WaldmanAntidote du jour (hat tip reader Barbara B). Turn the sound up: Read More at […]

  54. beowulf writes:

    As to overleveraged firms, is the problem too much existing debt or too little available credit? I guess this goes to Wynne Godley’s point about “buffer stocks”, a firm with a large enough credit line can outlast competitors that don’t have the credit (or corporate savings) to buffer its losses during a downtown before the economy turn around and the company starts making money again.

    “It will be argued that there will always be at least one component of the operations of a sector over which that sector has no direct control. Each sector is perforce operating under conditions of uncertainty and hence must take decisions on the basis of what it expects will happen, so there must be at least one flexible component in each sector’s list of options which acts as a buffer and over which it has no control in the very short run. There is, in this model, neither a short nor a long term market clearing mechanism which brings (or which fails to bring) supplies into equivalence with demands with the important exception that prices do clear the market for equities and bonds… As firms do not know what their sales will be, there must once
    again be a flexible element, assumed in the model to be changes in inventories, which act
    as a buffer. But any increase in inventories requires additional finance, assumed to be
    readily available in the form of additional bank loans.” (p. 6, 7)

    http://webcache.googleusercontent.com/search?q=cache:66FFHVc14lcJ:www.ipc-undp.org/publications/srp/TOWARDS%2520A%2520RECONSTRUCTION%2520OF%2520MACROECONOMICS.pdf

  55. Luis Enrique writes:

    Thanks v much Steve

    I like your point that if input providers at the bottom of the chain cuts prices, rather than cooperate and cut prices too, the agents higher un the supply chain would try to maintain prices and capture some extra profit. This is stretching things a bit, but the owner of a nominal debt during a negative shock is a bit like that – if prices are going to fall, the real value of their nominal asset is going to rise, so they’re trying to grab some extra profit by not cutting their “price”. If they forgave some nominal debt (took a x% haircut) they would see the real value of their debt preserved if prices consequently fell x% too. When it comes to simple economic models that include price stickiness, is that why everybody is happy to cut prices becuase real incomes are preserved, because this coordination is assumed to happen and we just leap to the new lower price equilibrium, rather than getting tangled up in prisoners dilemmas?

    somebody else has probably already made this point (I haven’t read every comment) but this is all somewhat moot in the presence of a central bank determined to avoid deflation isn’t it? I’m not sure I fully understand why some models say price cuts would help in a recession, but real world policy strongly disagrees (is it just becuase consumers defer purchases?)

  56. […] – Sticky prices, leverage and Pascal’s wager. […]

  57. […] Sticky prices, leverage, and Pascal’s wager Steve Waldman […]

  58. reason writes:

    Steve,
    Like some of the others, I’m a bit suspicious of the “near capacity” assumption, my guess is that it is worse for your argument than you think. But I think there might be another approach that gets you there – the “financial health” bluff. Firms may well reason that lowering prices (and there is a good chance that their competitors will follow suit) makes them look desperate. In many markets (especially consumer durables and building trades), the ability to remain solvent is of itself a sales point.
    My answer to others such as Scott would be – you are perfectly correct – maximising income by whatever method still makes sense (but information uncertainty is maybe bigger than you are allowing for) – but empirically it is not what happens. And I think that remains a problem for economics as a whole.
    I remember in 1990 in London walking down Regent Street in London and noticing every second shop had a for let sign in it. I wondered – why are rents so inflexible? Why did the owners prefer to lose tenants rather than reduce the rents. After getting something is better than getting nothing. Someone on a blog suggested that the owners are leveraged. (I very much doubted that this was the case in Regent Street – built by John Nash early in the 19th Century – but even that doesn’t explain it – making a loss on a rental property is still better that getting no income at all on it.)
    The only thing I can come up with is the difficulty in negotiating the changed contracts – reducing the rent in one property almost ensures all the other tenants will come knocking. Maybe it is easier to find the new market price with new tenants, than with old. (Something the same may explain what happens to wages).

  59. reason writes:

    Steve,
    your argument about Apple applies even more so to Pharmaceutical companies – and for a similar reason – apple is not a manufacturer but an innovator – at least that is its key ability. Following the business cycle makes little sense for a business that is looking 10 years ahead all the time. (That is why also why it isn’t leveraged – the returns on its products are too risky for borrowing to make sense.)

  60. dblob writes:

    Prices cannot be too sticky, else a new (under-levered?) entrant you will have a competitive advantage.

  61. reason writes:

    “So firms find layoff lotteries to be better ex ante, because employees prefer them, and ex post because only the happy winners remain with the firm. A perennial suggestion among reformers is that we substitute some form of work-sharing for cyclical unemployment, so that the burden of downturns is evenly shared instead of falling disproportionately on an unlucky few. That sort of reform only makes sense when household leverage is generally modest.”

    I don’t think this is correct. Firstly, if work-sharing was a normal response then families wouldn’t be so levered in the first place. And even if they were, then the bank would be more sympathetic to rescheduling their debt. Secondly, most families now have two earners so the jeopardy is doubled. I don’t think MOST workers would prefer a lotterie. I think EMPLOYERS prefer it. I see it like this – employers don’t like sacking workers in good times. They are hard to replace and it discourages workers from joining them. So they do it when they have a good excuse and the workers are easy to find if they find they need replacements for some of them.

  62. […] Sticky prices, leverage, and Pascal’s wager Steve Waldman […]

  63. reason writes:

    dblog.
    Ah – but you are ignoring game theory and dynamics. Nobody enters a stagnant market, and merely the threat of lowering prices can deter new entrants.

  64. […] Leverage Create Price Stickiness? Posted on May 19, 2011 by rsj SRW has argued that leveraged firms are less likely to lower prices as they need to make debt payments, and this […]

  65. RSJ writes:

    I don’t see why the cost of debt should be the binding constraint rather than the cost of equity. Equity is not free, and must be paid for in the same way as debt.

    It’s true that a failure to meet the cost of equity will not result in formal bankruptcy for the firm. But it will result in the replacement of management and the selling off of the firm for scrap. You’ve seen firms with zero debt lay off large portions of their workforce just to meet their earnings targets. Many firms that are current on their debt payments shut down plants and close business units if they are not able to meet their dividend or earnings growth targets.

    I did some number crunching and see no correlation between firm leverage (or firm debt burdens) and the volatility of prices for the firm’s products.

  66. Steve Randy Waldman writes:

    dblog — Echoing reason’s response, I think you are right in the long term, but wrong over a short time horizon.

    My account presumes imperfect competition. If firms are price-takers and there are no barriers to entry, than firms refusing to drop prices make no sales and go out of business.

    But under imperfect competition, firms do achieve sales at a variety of price, but the volume of sales varies, both with price and as a function of random noise. So firms can hold above-optimal prices and sell, and if they are lucky, they can sell a lot.

    If new entrants without leverage showed up immediately, and were able to offer substantially lower prices without fear of bankruptcy, then sure, they’d eat nonadjusters’ lunch. But if there are barriers to entry — nothing insidious necessarily, in many industries it takes a substantial investment of time and capital to organize a technologically competitive firm — then unleveraged competitors won’t immediately rush in and there will be scope for leveraged incumbents to maintain high prices.

    Also, as reason points out, periods following a downward demand shock are the periods in which it’s hardest to organize new entrants, as the industry is deemed to face overcapacity and require consolidation. And in real terms, that may be right. There is a temporary opportunity due to firms’ reluctance to drop prices, but the fall in demand may be durable, in which case it would be foolish to enter the industry. Sure, you’d do well until you’d driven competitors to bankruptcy or their capital structures loosened up. But the physical capacity would remain, and bankrupt firms might reorganize rather than liquidate, so after a transition period of free eats, you might find your new firm is indeed trouble.

    In the long term, if firms for whatever reason maintain uncompetitive pricing, we hope that markets are sufficiently contestable that new firms will enter. But it really depends on the reason that incumbents are uncompetitive. If their technology is obsolete, new entry is smart, you’ll have a durable advantage. But if their technology is state-of-the-art and their only problem is capital structure, your advantage would be temporary and you’d be well advised not to enter (if entry and exit are costly).

  67. Steve Randy Waldman writes:

    reason — Re capacity, this is really calling out for math. Limited capacity is an easy way of communicating that benefits associated with a change in the distribution of sales following a price drop won’t save you. Absent a capacity limit, there are some ways the distribution of sales might change that would save you, and some ways the distribution of sales might change that would not. Really, the firm will compare the expected distribution of EBIT after a price drop with the expected distribution of EBIT holding price firms, and choose the distribution that maximizes the value of EBIT above a bankruptcy threshold.

    My intuition is that for many firms, sometimes but not always due to capacity constraints, the truncated distribution of EBIT conditional on a high price will be more valuable than that at a low price. Capacity constraints would do that, but so would a great deal of uncertainty of sales (volatility increases option value!). So too would high variable costs (even if dropping prices increases sales, it won’t help meet your debt obligations if unit profits are too small). There are a lot of plausible circumstances where the high price truncated distribution of EBIT might dominate the downward-adjusted distribution. Of course, there are plausible circumstances the other way too. So, ideally we should characterize the sort of distributions that would lead to one choice over the other, and try to get some sense of how common adjustment-reluctant pricing should be.

    Alternatively, we can try to examine the issue empirically, as RSJ below has done (although I am not ultimately persuaded by his experiment).

  68. Steve Randy Waldman writes:

    reason @61 — “if work-sharing was a normal response then families wouldn’t be so levered in the first place.”

    I agree with you there. But it is an equilibrium question. If we were in an equilibrium where work-sharing were normal, people would leave buffer to accommodate occasional worksharing, and it wouldn’t be disruptive. But we are in an equilibrium where worksharing is not normal, where people have the expectation of fairly continuous and rising earnings and have leveraged their lifestyles accordingly. How to get from one equilibrium to the other is a challenging question (if in fact we wish to do so). In the current equilibrium, workers are leveraged and do face severe disruption if wages or hours are reduced. (Going forward, perhaps this diminishes, as the recent turmoil induces greater caution. But so far, I don’t see that happening except among the minority that actually experiences severe distress. And many of them remain unemployed and in distress.)

    Similarly, it would be a very different world, in which banks share households’ wage-cash-flow risk and flexibly reschedule payments without tacking on high fees as compensation for uncertainty and delay. Such an equilibrium is hopefully possible, but we are far from it.

    Your point is a good one, that many households are leveraged to two earnings, which basically doubles their chance of losing a lottery and facing a disruptive crisis. I’ll score one for you there.

    But I’ll counter with a point I left out of the main post. Workers don’t face identical odds of losing the lottery, and they evaluate their own chances, sometimes accurately and overoptimistically. We’ll call workers likely to win the lottery “well-situated” and those likely to lose “poorly-situated”. Well-situated workers whose evaluations are accurate and poorly-situated workers who are overoptimistic all prefer lotteries to a shared burden. The only employees who dislike lotteries are poorly-situated workers whose estimates are accurate and well-situated workers whose estimates are over-pessimistic. The firm is less interested in the views of poorly situated workers; they are by definition dispensable. So the ex ante preference for lotteries is shared between firms and some workers, the workers who are ultimately enfranchised in the decisionmaking process. Poorly-situated workers who accurately estimate their chances prefer worksharing, but employers don’t care. Employers do have to placate well-situated but pessimistic workers, for fear of losing or harming the productivity of employees they highly value. But this is probably a small group, and it probably is not so difficult to find ways of relieving them of their pessimism.

    If well-situated workers who accurately estimate their chances tend to aggregate in married couples, then “the households that matter” will prefer lotteries they expect to win.

    Which I think is basically where we are. There’s a conflict of interest between more and less dispensable workers in this decision of whether to fire people or work-share. Firms prefer firing people, and they align themselves with workers whose odds of losing the lottery, while never zero (!), are relatively small. If well-situated people are confident and therefore especially leveraged, their preference for not sharing burdens may be strong.

  69. […] whose excellent blog is windyanabasis, takes issue with my claim that financial leverage is a source of sticky prices. Not only that, but he’s performed an […]

  70. Alex writes:

    Really fascinating post. A couple of points.

    In so far as you’re leveraged, your position actually does change from being the outright owner to being the holder of a call option. On the downside you stand to be wiped out, but on the upside you only repay the nominal debt and get to keep the return on the whole value of the deal (your equity plus the OPM). This is of course why people love leverage so much.

    If you own the business outright, your budget constraint works differently – in any form of imperfect competition, you are not really a price taker with regard to your return on invested capital. The binding constraint is OPEX – you can put up with lower revenues and profitability as long as you can make payroll. And the upshot of this is that you would theoretically be willing to cut prices as much as it takes to keep the business going. Arguably, what’s going on is a sort of smoothing of profits over the business cycle. If you can accept reduced profitability in the recessions, this implies that you have some pricing power in the boom (or that you can hold out on wages and let your margins drift up). In a sense, you can forgo profit in the expectation that you can recoup it down the line (or alternatively, unwind past profits).

    (To be classical, there’s a limit where your return on capital hits the risk free interest rate and at this point you sell. But if the risk free interest is 0.5%? A big recession implies zero bound conditions. Arguably, if you own the business outright you’d hang on rather than accepting peanuts.)

    The calculation is very different, as SRW says, if you’re leveraged. In that case, the limiting case is the level of free cash flow you need to service debt (and in practice, you probably have banking covenants that cause a liquidity failure at some point). At any level below that you lose the lot, so you may as well roll the dice.

    Interesting – leverage brings the excitement of derivatives to the real economy!

    A broader issue – in what way should you *not* be risk-averse? Arguably, risk aversion is rational. Going bankrupt is an unpleasant, shameful, and costly experience, as is unemployment. Only economists think that losing the whole business, the ship sinking and the farm burning down, is not really a problem. (Also, seeing as human beings are risk averse, shouldn’t economics accept this and deal with it rather than trying to stuff them into the homo economicus framework?)

    An even broader issue – if we accept that there are shocks (or that there are system-inherent sources of instability), should we actually be looking at ways of making the system adjust real fast to what are, ex hypothesi, random or at least meaningless events? If the function of the economy is to allocate resources between problems, isn’t reacting to random events fundamentally harmful to this purpose? If you’re responding to random events, you’re behaving randomly. If you’re building a control system of some sort, a major concern is to *avoid* responding to random noise – to filter it out – or if there has to be response, to prevent the system going into a feedback loop driven by the noise. The whole point of the discipline of statistics is to extract trends from noise.

    So the economy is a bunch of interrelated production functions. If nothing changes in those underlying *relations of production, baby*, why should any prices change? A genuinely random shock *ought* to be ignored. Is stickiness actually a good thing?

  71. Alex writes:

    To be clear, I suspect we could do with less leverage, and I wonder what the leverage ratios prevailing in the Mittelstand are. (Firms in Germany are stereotypically bank financed, but on the other hand, companies that are publicly traded often experience pressure to “lever up”.) This may seem to go against my point on stickiness, but I suspect that option-ising businesses doesn’t lead to rational decision-making. Assuming mediocrity on all sides, a lot of the businesses that choose to keep trying and hope to bring it off will fail and a lot of the ones that go to zero might have made it.

  72. […] you are wondering where the shape of the sticky-price expenditure demand curve comes from, see my earlier post on sticky prices. Basically, to generate the expenditure demand curve with price rigidity, I assume that industry […]