Statsguy applies game theory to Fed policy
Today I’ll offer a rare guest post. I invited Statsguy to do this after reading a comment he left on my reply to Steve Waldman a few days ago. Because the post is rather long, I won’t indent the whole thing as I usually do when quoting others. Instead, everything after this sentence was sent to me by Statsguy, except for the final paragraph, where I’ll offer a few comments:
“If everybody could suddenly be made to believe that income would be plentiful, everyone would spend freely and fulfill the expectations of plenty. The world is a much more pleasant place under the second set of expectations than the first. And to switch between the two scenarios, all that is required is persuasion. The market-monetarist central bank is nothing more than a great persuader:” – Steve Waldman
“Speak softly but carry a big stick” – Slogan, Teddy Roosevelt
Many observers – including advocates of NGDP targeting – are questioning exactly how much action the Fed will need to take in order to hit an NGDP target. In particular, the recent conversation between Scott Sumner, Steve Waldman, and Nick Rowe has raised questions such as:
- How big will the Fed’s balance sheet need to be in order to target NGDP?
- What will the Fed need to do to interest rates?
- What kind of assets will the Fed need to buy to hit its target?
The answers to these questions depend on whether one approaches the topic from the perspective of comparative statics (simple equilibrium conditions) or game theory, as well as one’s understanding of the primary drivers of our current economic predicament and expectations about other policies (e.g. bank capital requirements).
The happiest interpretation of the Market Monetarist view was expressed by Steve Waldman above. This view sees monetary policy as a social device that sets focal (Schelling) points in cooperative but unstable games. For example, driving on the right side of the road. In the basic “driving game”, no one cares which side of the road they drive on, so long as everyone drives on the same side. Since there are two sides, there are two stable Nash equilibria. In this type of game (without communication), all it takes is ONE individual to announce their intention to drive on the right side of the road for everyone (who is rational) to drive on the right side of the road. Why? Because if my prior is that all other people would randomly choose a side with 50% probability, but one person would drive on the right with 100% probability, then I am _slightly_ less likely to get into an accident if I drive on the right. The happiest part of this is that everyone makes the same calculation, and so when morning comes around everyone is driving on the right side of the road. If monetary policy is a driving game, then solving it is a simple coordination problem. The Fed barely needs to do anything other than yell “let’s all do X together!”, and the flashmob follows.
However, there are many other types of games. One is called Deer Hunter. In this game, a bunch of hunters form a circle to trap a deer, and each gets a share of the kill. This is costly for each hunter (time and effort to show up), but the share outweighs the cost. If even one hunter doesn’t show up, however, all that cost is wasted (everyone loses a little). Succeeding at this game is just slightly harder – the Fed could set a time and a date for the hunt, but each hunter actually needs to believe that every OTHER hunter would also show up. This game now requires persuasion. The persuasion is not that hard, however. For instance, if every hunter agrees to come on a specific time and date, then the event will happen because everyone has an incentive to do what they say. In this game there are two Nash equilibria. If I believe that no one will come, then my best payoff is not to show. But if I believe everyone will show, then my best payoff is to show. There is no incentive to free ride, but credible coordination is costly.
Finally, there’s prisoner’s dilemma, which we all know well. If we’re all playing prisoner’s dilemma, my ideal outcome is that everyone cooperates but I defect. This is the game that Keynes was talking about when he discussed the “paradox of thrift”. Savings is virtuous for the individual, but if too many individuals do it in excess, it harms everyone. In this game, the only Nash equilibrium is for everyone to defect (everyone to save in excess). If I know everyone else will save, I will save, and if I know everyone else will spend, I will still save. Now, it’s that latter part which is a point of contention. If I know that everyone else will spend, then why would I save? I should want to invest, to take advantage of the high rate of return. However, if everyone tries to invest instead of saving, the return on investment could drop negative (below the value of holding currency, assuming no inflation), so we will have excess holding of currency and insufficient investment (and both future and present consumption).
Now, one obvious question to ask is why we can’t just restore the previous equilibrium, in which everyone expected higher consumption and therefore everyone invested more and consumed more? Moving back to that old equilibrium would simply require persuasion, which means we obviously aren’t playing prisoner’s dilemma, right?
In the “not so happy” version of the world, we must posit that there has been a shift in preference for savings or expectations of the probability of success for future investments. Perhaps a Minsky Moment, or a sudden cultural awakening to the value of savings. The old equilibrium no longer exists, at least not until we’ve gotten some financial deleveraging or discovered a new source of cheap oil. Even if reality has not changed, perceptions or preferences have. The future is uncertain and savings in currency represents a claim on the future production of other people. In a complex society such as ours, savings in a non-inflating currency is the single “best” way to preserve future consumption. By “best”, we mean lowest risk, because currencies are defined against a basked of goods (which individually fluctuate in price). The variance of the currency with respect to the price of any individual good is thus lower than the variance of the relative price of two goods. Moreover, currency is storable (with the exception of inflation).
In this “not so happy world”, the Fed needs to do more – it needs to force consumption and/or investment by forcing everyone to cooperate. This aspect of the Fed’s job is like forcing everyone to pay taxes. My “ideal state” (assuming govt. expenses are fixed) is that everyone pays taxes and I do not, but given a choice between everyone paying taxes and no one paying taxes, I’d rather everyone pay taxes. This is the classic Hobbesian solution.
But how does the Fed get everyone to “cooperate”? It does so by changing the payoff matrix – by altering the outcome so that if I “cooperate” by consuming or investing while everyone else saves in currency, I’m still better off than if I save. If this threat is credible, then the Fed should not actually have to enforce the threat. The game with two Nash Equilibria now collapses to a game with one equilibrium. This is like Nick Rowe’s example of Chuck Norris clearing out a room. Done well, Chuck doesn’t have to hurt anyone, just crack his knuckles. In this type of situation, the Fed’s balance sheet hardly changes at all. Indeed, it could actually shrink as the real interest rate declines and the Fed needs to tighten to reign in an overheating economy. This is the “sort of happy” world, where the Fed doesn’t need to actually beat anyone up, just glower at them.
So is this “sort of happy” world the real story? Again, that depends on several things, including:
- Will the Fed’s threat be credible? After 2008, and the Bank of Japan, some (including Paul Krugman) have expressed significant doubts. If the bank’s threat is not credible, capital markets will test it, and the cost of proving that credibility could go up. In practice, Market Monetarists think this argument is overstated. Scott has repeatedly called out the Bank of Japan publicly declaring that it was targeting little to no inflation, so it got what it said it wanted. Recently, many questioned whether Switzerland could stop the seemingly unstoppable rise of the Franc – those who questioned the SNB with currency instruments lost a lot of wealth.
- Even if the threat was credible, why would the promised actions of the Fed have an effect on real economic activity? Put differently, what is the “transmission mechanism”? If the Fed simply accumulates massive amounts of federal debt by expanding its balance sheet, banks can keep holding ever greater reserves. Indeed, we’ve seen some evidence of this. Scott would note that the Fed’s decision to pay a positive IoR (interest on reserves) certainly didn’t help, although the real effect was small in comparison to the size of bank losses on real estate assets. In extremis (Fed imposes a negative IoR), some argue the banks can hold vault cash and individual citizens can put money in the mattress (banks might charge you a fee for keeping a checking account!). If the transmission mechanism is not real and obviously believable, then the Fed may take ever-more-distortionary actions and discover they aren’t having an effect because there is no mechanism for that effect to occur. What’s worse, even if the transmission mechanism is real but is NOT obvious (and believable), the Fed would find that it needs to pay a much higher price to achieve its NGDP target. In this case, the Fed’s plan would work, but it would need to follow through on the threat. (In this situation, people believe Chuck will try to beat people up, but that his fists won’t cause “transmit” damage, so Chuck actually has to start breaking some bones just to prove that his fists in fact do cause injury.) Just to be clear, all market monetarists believe that the primary mechanism of action is indeed expectations, but they also believe that even if expectations don’t function as expected there is a crude backup mechanism that will still work. (Otherwise, we’re bluffing, and in public! The emperor has no clothes…)
- So what is that mechanism? That differs depending on who you ask, but one mechanism (proposed by Scott) is simply the cash balance mechanism. There is some amount of cash that the public (including our trading partners) would be unwilling to hold, and there is also a limit to the speculative investments that the public is willing to hold as a carry trade against the currency. If the public held that much cash (or speculative investments) at any point in time, they would rather they had spent some of it earlier. Knowing this and that the Fed is intent on pumping that much cash into the economy, the public will spend it before the Fed actually injects as much cash as it threatens. This is a form of recursion. In time period T+1, the Fed will add X cash to the economy – more than people will want to hold, unless they drop below Y today. There are lots of ways to take advantage of this. The Fed could establish any threshold it wants for current reserves (or NGDP) and hit that target… IF everyone believes the Fed is really willing to pay the cost. But if they believe it’s willing to pay the cost, it won’t have to pay. The Fed can also set targets 30 years out, and recursion impacts the present. IF people don’t want to hold X in time T+30, then they will be under some value Y in T+29, and to get below that they’ll need to be below Z in T+28, etc…
- Even if the threat is credible and the transmission mechanism is real and believable, what if households and firms just want to hold more cash in equilibrium? Or, alternatively, what if they want to deleverage and pay off their debts? Is that bad? Or, similarly, what if the government is requiring banks to raise more equity and use less leverage (by tightening the Basel accords to increase cash-to-asset ratios, or by regulating loan standards to limit the moral hazard created by federal deposit insurance or TBTF backstops)? Again, since preferences for holding cash have shifted, everyone would like to hold more cash (even if the prior equilibrium could be restored, which it can’t precisely because everyone wants to hold more cash). Think of this as a general decrease in risk tolerance. In this case, the Fed could still force people to leverage up by making draconian threats to the payoff matrix in the “savings game”. However, this could prove costly and lack political credibility if the magnitude of the threat was too severe. Alternatively, the Fed might need to partially follow through on it’s “threat” by carrying out a plan to inject cash into the economy, and then everyone will be happy. People won’t actually spend the money, they just want to have it (perhaps as a form of insurance). In essence, the Fed is giving everyone a Teddy Bear. Inflation won’t go up (unless preferences or risk perceptions again change), but everyone will feel safer. And individually, they will be safer (just as if they bought insurance against private hazards like job loss or health decline), but collectively society has no more resources to absorb a massive shock (though it will probably have more resources after 30 years of higher trend investment). The Fed’s exposure to risk increases, but individual household exposure decreases as does the exposure of the fiscal authority (which is quickly becoming tapped out as debt-to-gdp figures soar).
I personally think this last case – a sort of hybrid solution – is the most plausible and desirable. It becomes even more plausible if one takes into account the heterogeneity of preference for cash (liquidity) in the population and among firms, which can soften some of the knife-edge solutions that multi-player game theory with homogenous payoff matrices can yield.
Thus, the Fed probably will need to inject some cash into the economy in order to allow the private sector to deleverage to its comfort level and increase personal savings without crippling sovereign governments with high debts. More importantly, the Fed will need to continue to drip cash into the economy over a prolonged period (particularly as the entire world population ages and shifts to more risk averse investments). However, if the Fed clearly lays out its plan to do so in order to hit a steady NGDP trajectory, then the total amount of cash it needs to inject in the economy will be much smaller. Even better, the Fed would avoid the cyclical disruptions that have bedeviled us for the last 3 years as the Fed responds to economic threats ONLY AFTER it has definitive proof that a recessionary environment has taken hold.
There are certainly a lot of complexities beyond what is discussed above, however all market monetarists would seem to agree on the following:
- There is no question the Fed can hit an NGDP target.
- The cost of hitting that target is vastly lower if the target is declared and the mechanism is communicated clearly and credibly.
- Aggregate demand will respond, at some level, to nominal targeting, and it will respond immediately because it MUST respond eventually.
OK, I’m back. Excellent explanation, but a few small quibbles. Vault cash is part of reserves, so negative IOR on excess reserves would make no sense unless vault cash was included in total reserves data. I think the final comments about the likely increase in demand for liquidity are quite interesting, but I have a slightly different take. I foresee lower than normal real interest rates in the coming decades. (Or maybe I should say I am just reporting TIPS market predictions.) But whether this leads to more demand for cash depends entirely on the impact on nominal rates, which depends on the Fed’s inflation (or NGDP) target. For instance, if we targeted NGDP growth at Australian levels (I’m not recommending that—but there are much worse fates, including our current fate) then we’d avoid the zero rate trap and monetary base demand would remain well-contained, at around 5% of GDP. Super-neutrality of money is often overlooked by non-monetarists (most notoriously the MMTers.) Admittedly, it’s very likely that the Fed will keep the 2% inflation target, which means Statsguy might well be right. But I wanted to point out that it’s not just about “fundamentals,” it’s also about trend inflation, which we can control. And finally, I’d recommend Lars Christensen’s piece on Switzerland as a supplement to Statsguy’s comments. (Also look at the graphs in the post Lars links to.) Lars shows that when the Swiss National Bank was trying to stimulate with a bad policy (no explicit target) it expended enormous resources and had a somewhat spotty record. As soon as it set an explicit ceiling on the SF (of 1.2 SF per euro) they had great success, without having having to spend much at all on intervention. That’s a wonderful example of the Chuck Norris effect in action. I think the massive August intervention by the SNB is a good way of understanding how I can argue money is tight, despite all the QE. They were doing a lot, and it probably had some effect, but it’s much more difficult than if you have a credible policy objective. I hope the Fed studies Switzerland as they consider more explicit communication. How explicit should they get? How about XXX-rated? Nothing to hide.
Tags:
4. November 2011 at 08:46
Scott and Statguy…excellent stuff.
4. November 2011 at 08:49
“What kind of assets will the Fed need to buy to hit its target?”
Who says the fed should have any assets?
Are these assets actually savings vehicles?
4. November 2011 at 08:51
Nice bit of reasoning. Bravo, Statsguy.
4. November 2011 at 08:51
It was ok up until here:
“But how does the Fed get everyone to “cooperate”? It does so by changing the payoff matrix – by altering the outcome so that if I “cooperate” by consuming or investing while everyone else saves in currency, I’m still better off than if I save.”
It is very easy, the Fed just prints money and gives it ONLY TO the savers until they say WTF, I’m going to spend.
The point of saving is that you think current prices / payoff are to high.
One last note on Prisoner’s Dilemma…. you wildly miss this.
The point in a capitalist market system is that defectors who defect first correctly get to buy the cheap assets of the losers and then the game begins again.
If you defect to early, you lose out on the investment growth.
If you defect to late, there is no chair for you when the music stops and you get gutted.
This is a PERFECT game.
The problem is the artificial stupid rules recently set up to try and keep the gutting of fish from happening.
Stats, you just don’t like to see people lose.
4. November 2011 at 08:51
This is very good.
Deer Hunter is a version of JJ Rousseau’s Staghunt, I think, only with more than 2 players?
One quibble. Last bullet. Remember, the US is on some sort of equilibrium path now, by definition. There must be *something* balancing the deflationary effects of deleveraging and/or saving and/or hoarding. Otherwise NGDP would already be zero. All the Fed needs to do is throw *any* weight behind those balancing forces to make things a bit better.
I didn’t express that very clearly.
4. November 2011 at 08:57
To be a *little* snarky. In the Deer Hunter Game, as we’re all forming a circle around the deer, I hope we all aim well…otherwise, the collateral damage might prevent *any* of us from enjoying the roast venison.
4. November 2011 at 09:08
Scott
Great stuff.
For what it’s worth, though, I would point out that Statsguy’s final solution (his “most plausible and desirable” solution) bears a striking resemblance to the variation I offered on Matt Yglesius’s cash-injection plan a few weeks back.
No quibbles… I’m happy to hear it, and I think he did a great job describing WHY this is the most plausible and desirable solution.
4. November 2011 at 09:12
It seems to me the “market monetarists” are just another version of spoil the rich, trickle down, supply side economics.
“Savings is virtuous for the individual, but if too many individuals do it in excess, it harms everyone.”
What if an entity “dissaves” in the proper way so that all “individuals” can save?
“Now, one obvious question to ask is why we can’t just restore the previous equilibrium, in which everyone expected higher consumption and therefore everyone invested more and consumed more?”
because of the debt and wealth/income inequality ; If you want the lower and middle class to consume more, you should have to pay them more assuming they want to consume more.
“In a complex society such as ours, savings in a non-inflating currency is the single “best” way to preserve future consumption.”
including being able to retire?
4. November 2011 at 09:45
Two points/questions for Satsguy:
1) MM’s imply the Fed can “change the payoff matrix” by promising to hit the level target. That is, a temporary, modest rise in NGDP/inflation would do the trick. This sets up an asymmetric payoff distribution: if one holds cash and the Fed succeeds, the negative payout to holding cash is limited to the modest NGDP gap; if the Fed fails, the positive payout is comparatively quite large in a severe deflation scenario. Why would actors move away from cash given that asymmetry? This implies the Fed must introduce a symmetric payout distribution distribution: it must, by seeming “irresponsible”, create the possibility of large losses to cash holdings. Unfortunately, such perceived “irresponsibility” also creates the risk of unanchoring inflation expectations.
2) Does the Fed really inject “cash” when it buys securities? It effectively pays for these with excess reserves. Excess reserves are, by definition, not a medium of exchange. Thus, swapping short term liabilities for risk assets could just as easily be accomplished by the Treasury — in effect, it is a fiscal action, not a monetary one.
4. November 2011 at 10:56
> However, if everyone tries to invest instead of saving, the
> return on investment could drop negative
Huh??
> Put differently, what is the “transmission mechanism”?
> If the Fed simply accumulates massive amounts of federal
> debt by expanding its balance sheet, banks can keep holding
> ever greater reserves
I see this a lot and don’t understand it. Maybe someone can explain where this goes wrong: Fed purchases give base money to banks, yes, but also create demand for bonds and therefore give money to government. In fact, primary dealers are required to sell, so the effect is rather direct. Government spends this money rapidly (20% GDP) and because the Fed retires its bonds and even refunds the coupons to the government, this is also ‘base money’ — inflationary. I assumed this was the QE2 transmission mechanism. It seems to make more sense than stock market wealth effects (advanced by Scott).
> Knowing this and that the Fed is intent on pumping that much
> cash into the economy
…Still requires a credible transmission mechanism to get the cash out.
4. November 2011 at 11:34
David Pearson, note I’m making the same point…. you can’t scare them to action, but you can bribe them.
Print the money, give it to the Tea Party (the savers), and keep giving it to them, until they decide to start spending.
The Fed is afraid of the Tea Party, they are not afraid of OWS.
4. November 2011 at 12:39
At some point we need to formalize the discussion with a model. Statsguys is walking down paths laying out detailed points, but every step of the way I’m reluctant to follow him without a specific model in mind. Statsguy, do you have one in mind? Does one even exist?
4. November 2011 at 12:52
“Knowing this and that the Fed is intent on pumping that much cash into the economy, the public will spend it before the Fed actually injects as much cash as it threatens. This is a form of recursion.”
That’s how I see it working too. In financial markets, we call it front-running: getting information about a trade before execution and moving in front of it.
Periodically, the front runner will stop believing the front runnee and will require proof of front runnee purchases before they’ll front run again. Chuck Norris will always be tested.
4. November 2011 at 13:29
@ Fed Up: At the zero bound, if not buying assets what other action would the Fed take/threaten to hit its target?
@ Morgan: the game is only “perfect” if the point is to concentrate wealth based on luck/timing, without regard to maximizing overall growth by reducing volatility.
@ Nick: Yes, it’s multiplayer Stag Hunt, simplified with no rabbits to chase.
I get what you mean about the current situation needing to be an equillibrium of some sort, though not completely. There’s also this notion of an economy hitting “take off”, where we cross a threshold so that an economic expansion accelerates endogenously. In that sense, whether or not we are above or below a separation line in a phase diagram could be binary.
@ Brett: Yes, it’s similar to Yglesias’ cash injection plan, but even before that to Joe Gagnon’s plan
@ Fed Up: the impact of monetary policy on wealth distribution is open to debate (I’ve taken your side at times), but would you not agree that (all things equal, including the monetary targets) removing volatility and uncertainty from a monetary regime is better for the poor than a high volatility regime with the same target?
@ David: If the Fed sets a target of X, and some people doubt the credibility, this could create an asymmetric risk/payoff. However, it depends on what the “median dollar” in the marketplace thinks, since if people bet against the Fed this will increase the payoff to betting with the Fed. This could limit the asymmetry. In any case, any asymmetry could require the Fed to actually start taking action to prove its credibility (which is one reason why level targeting is helpful). On the other side, there are people who think hyperinflation will hit, so the Fed will probably need to take real action from time to time.
Whether Fed purchases of debt are fiscal or monetary right now (given treasuring is running > 1trillion deficit) seems somewhat semantic. Note that when the Fed gives cash to banks to buy treasuries, banks anticipate this and buy treasuries to sell to the Fed – this bids up prices for treasuries for current owners, who may then sell. Cash does get injected. However, with a credible target, the amount of cash that actually gets injected is less important than the impact on asset valuations.
@ Carl: this should say “if everyone tries to invest instead of saving IN CURRENCY”
I don’t think I fully understand all the transmission mechanisms as well, but it seems irrefutable that – even in the worst case scenario – the crude cash balance mechanism _must_ work. How much monetization the Fed is actually conducting is hard to assess (do we count money that the treasury WOULD have paid if the interest rates had been higher?). At a minimum, interest remitted to the treasury should certainly count as monetization, but that’s just not a huge amount of money in any given year (though expected monetization is more). It does seem that the impact of the _threat_ of monetization is more important than actual or even expected monetization, and that the threat of monetization operates largely through asset prices. However, it’s uncouth in official circles to publicly state that the Fed is stabilizing asset prices.
Hopefully this post is useful and did proper justice to everyone’s positions.
4. November 2011 at 13:41
For everyone’s amusement:
http://www.bloomberg.com/apps/news?pid=newsarchive&sid=aLepjRCRQ3go
My favorite:
“Chuck Norris doesn’t mark-to-market. The market marks to Chuck Norris.”
“Chuck Norris doesn’t supply collateral, only collateral damage.”
4. November 2011 at 15:08
StatsGuy said: “@ Fed Up: At the zero bound, if not buying assets what other action would the Fed take/threaten to hit its target?”
And, “@ Fed Up: the impact of monetary policy on wealth distribution is open to debate (I’ve taken your side at times), but would you not agree that (all things equal, including the monetary targets) removing volatility and uncertainty from a monetary regime is better for the poor than a high volatility regime with the same target?”
Good questions. For the first one, get more medium of exchange (liability) into circulation with no asset attached for the fed (the assets should be out in the private sector). This might require “removing” medium of exchange from certain entities who might react by driving up prices of real assets without a corresponding increase in production and/or raising interest rates.
For the second question and imo, having an all “currency” economy with no private debt and no gov’t debt would remove a lot of the “high volatility” because now the amount of medium of exchange can’t rise with increasing debt levels and fall with debt repayments and debt defaults. That would also eliminate this:
savings of the rich = dissavings of the gov’t (preferably with debt) plus dissavings of the lower and middle class (preferably with debt)
I believe it should be:
savings of the rich plus savings of the lower and middle class = the balanced budgets of the various levels of gov’t plus the dissavings of the currency printing entity with “currency” and no loan/bond attached
That means both the rich and the lower and middle class experience “real earnings growth” by their budgets and the currency printing entity experiences “negative real earnings growth” by its budget, which it makes up for with “currency” printing.
I want real AD and real AS to grow by about the same amount. If real AD is below real AS, then real GDP grows by real AD. No problem. Note I’m assuming real AD is NOT unlimited. I’m also NOT assuming real AS is at the same level or above real AD all of the time. In a third world country, real AD is most likely above real AS until enough capacity is built up.
4. November 2011 at 17:26
Frankly this was one of the most ridiculous posts I’ve seen on this sight. If solving economic problems was just about collectively forcing people to cooperate, why not just go around the relatively powerless Fed and just pass laws about things like holding more than a certain amount of money? Why not defeat inflation with price controls? Why not just ignore 300+ years of economic thinking and go back to assuming you can create utopia by writing laws? After all, it works so well in Cuba, North Korea, the USSR, and 1960s China. I hope I don’t see more gimmicky posts like this on here and Scott gets back to looking at the problems in a serious manner.
4. November 2011 at 17:28
“Vault cash is part of reserves, so negative IOR on excess reserves would make no sense unless vault cash was included in total reserves data.”
The way it was explained to me is that vault cash is not part of excess reserves. See:
http://www.interfluidity.com/v2/2110.html
Comments:
25, 35, 36, 58, 68, 97, 98, 101 (especially), 102 (especially), 103, 105, 106, 107
4. November 2011 at 20:36
A very interesting post. I have a minor side remark.
By ‘save’ Statsguy seems to mean *increase one’s holdings of “cash”* (= currency + checking-account deposits?). While I don’t like the definition, I won’t quibble about words. But I would like to have a term for J. B. Say’s notion of “saving”: *refraining from consumption*. Let’s call that ‘Say-saving’. When Statsguy writes:
“[W]hat if households and firms just want to hold more cash in equilibrium? Or, alternatively, what if they want to deleverage and pay off their debts?”
he is confusing the two notions: “saving” (increasing cash-holdings) and “Say-saving.” Deleveraging would usually involve *Say-saving*, but it has no likely correlation with “saving” in Statsguy’s sense. Indeed, to pay off a debt one would normally use cash, thus depleting rather than increasing his own holdings. But he might be expected to make up the depletion by scrimping on consumption–by extra Say-saving. (Other, less likely possibilities: (a) cutting back on his normal investing activities; (b) not cutting back on anything, except his holdings of cash.) However, there is no reason to expect someone who was paying down debt to end up holding more cash; deleveraging has nothing to do with Statsguy’s “saving.”
4. November 2011 at 20:44
Philo: “saving” is just another word for investment unless you save actual currency. If you’re paying off a loan, your lender has to just hold onto the currency or else there’s been no reduction in consumption.
4. November 2011 at 21:46
Statsguy,
A suggested test to see if something is monetary policy is as follows:
“Would the policy action have any different outcome or mechanism if the Treasury carried it out instead of the Fed?”
In the case of asset swaps for Excess Reserves, these are no different than asset swaps for T-bills. Each would have a similar impact on asset prices. Each does not create “money”. Each would create a contingent tax liability/asset for the taxpayer.
The Fed’s action would be slightly different in that it would reduce the liquidity risk borne by the private sector. This is because the Fed’s balance sheet bears no liquidity risk (due to its irredeemable liabilities). This is the rationale for the Fed’s LOLR function. However, it is clear that asset purchases are being proposed not for liquidity purposes but to raise asset prices — a fiscal action.
4. November 2011 at 23:52
Alex Godofsky,
Reduction of mortgage debt destroys money, particularly if it’s through default. It works something like this.
Buy mortgage -> Repo mortgage for a split of the interest – say 3% -> use money to buy another mortgage-> repo that for 3%. Rinse and repeat. Do it 30 times and you’re earning 90% for only risking total financial destruction, from which you personally can walk away. The velocity here is frightening.
“The average daily trading volume in the repo market was about $7.11 trillion in 2008, compared with the New York Stock Exchange, where the average daily trading volume in 2008 was around $80 billion”
http://research.stlouisfed.org/publications/review/10/11/Gorton.pdf
Repayment or default unwinds leverage structure. This creates a demand for better quality money to cover the costs of deleveraging. If this money is not available, you’ll see liquidity problems and a form of tight money.
Our problem is the giant money supply bubble blown up by the alternative financiers. Lehman was leveraged 30 to 1. The German banks are leveraged 31 to 1 last I heard.
The trick is to fix the economy without letting the financiers inflate another bubble (and, of course, to survive the rest of the deflation of this one. It ain’t over yet. We’ve still got Europe and China to go [google china ghost cities].). Financial engineering is much more profitable than mere banking.
5. November 2011 at 04:36
Statsguy: “@ Nick: Yes, it’s multiplayer Stag Hunt, simplified with no rabbits to chase.”
It’s *hares*, not rabbits!
5. November 2011 at 06:25
@ David: So, in that definition, actions are monetary only in so far as they pertain to liquidity risk? In that definition, are interest rate adjustments (away from the zero bound) fiscal?
@ Peter, Alex, Philo: I’ll agree that deleveraging does not _necessarily_ have the same impact as everyone saving in cash (or currency), but in practice it almost always has the same short/medium term impact on asset prices and AD. If saving in currency, the money is removed. If saving in bank deposits, the money could technically be redeployed by bank lending, but in a liquidity trap with excessive reserves, the money might as well be removed. If deleveraging, one is taking money from elsewhere (selling one’s own assets, selling labor, or taking from savings/currency) and paying off debt, which in a zero bound condition (with excessive reserves) means removing money. It does not _necessarily_ imply reduced consumption at the individual level, but at the aggregate it almost always would (without new money entering through fiscal/monetary action or an export boom). A meaningful savings/investment gap is implicitly tied to a liquidity trap(otherwise more savings = lower rates = more investment).
If one believes the static equilibrium story, then the goal of monetary policy in an LT is to lower the real interest rate.
If one believes the game theory story, then the goal of monetary policy in an LT is to _increase_ the rate by threatening to take actions that would _really_ reduce the rate.
These are very different understandings of the impact of monetary policy.
5. November 2011 at 07:04
Stats, you miss it entirely…
“Morgan: the game is only “perfect” if the point is to concentrate wealth based on luck/timing, without regard to maximizing overall growth by reducing volatility.”
This is what I said….
“The point in a capitalist market system is that defectors who defect first correctly get to buy the cheap assets of the losers and then the game begins again.
If you defect to early, you lose out on the investment growth.
If you defect to late, there is no chair for you when the music stops and you get gutted.
This is a PERFECT game.”
Under that system, according to everything Scott and company believe, it CANNOT accrue wealth to the top.
The “luck / timing” means that over the long term, no one is lucky every time.
Look, you mis-applied Prisoner’s Dilemma, under our system WELL TIMED DEFECTION (getting out of the market) is meant to be rewarded by buying up cheap assets when the losers are forced to liquidate.
That is our basic structure. It is the game that EVERYONE is supposed to be playing.
Don’t set up a analysis that isn’t true tot he rules. The rules were not followed, find game theory analysis that focuses on proving that flaw: the rules were not followed.
5. November 2011 at 09:46
Thanks Lars.
Fed Up, It’s convenient for them to have some assets when they need to reduce the base. That way they don’t have to constantly ask the Treasury to give them some T-securities.
Brett, You’ll need to remind me of your proposal.
Fed Up, I’m not sure I followed his saving arguments.
Regarding vault cash, the definitions change over time. But surely if the Fed went to negative IOR it would include vault cash, otherwise the plan would make no sense.
5. November 2011 at 11:21
Statsguy,
No, actions are monetary policy if they reduce liquidity risk or directly influence demand for bank reserves or currency. Changing the FFR accomplishes the latter. Swapping Excess Reserves for assets does not.
The Fiscal action of raising asset prices might indirectly create bank reserve demand, just as any successful fiscal action might.
Excess reserves are not “cash” or “money”. Under an FFR targeting regime, the Fed stands ready to supply all bank reserves demanded by the system at any given FFR target rate. The pre-existence of ER’s, no matter how large, is irrelevant to the supply of the medium of exchange.
5. November 2011 at 16:05
@David – so, under that definition, the main instrument for implementing monetary policy at the zero bound would be a negative IoR? I’m challenged by the notion that direct monetization of debt, which most people would consider a monetary action (creating more money), might not be defined as monetary. At a higher level, I’m not sure I see how liquidity risk is really separable from asset pricing, seeing as classic asset pricing models directly incorporate the risk premium. I think I get your point, but it sounds more like intent rather than effect (action X does A and B… but if you intended to do A it’s a fiscal action, whereas if you intended to do B it’s monetary).
5. November 2011 at 17:00
Yes, the liquidity risk portion of capm can be extinguished by the Fed: the rate and beta risks are passed back to the private sector through contingent tax liabilities. In liquidity crises, liquidity risk premia spike and QE has great effect. Other times, liquidity risk is a small part of overall risk.
Fiscal deficits can increase demand for total credit and therefore for bank reserves. If the Fed meets that demand at the zero bound, it is providing bank reserves — i.e. “money”. If the fiscal deficits do not increase total credit and create demand for bank reserves, then QE is not monetization, as Scott himself has pointed out. It is merely swapping a s.t. gov’t liability for a risk asset. This could just as easily be accomplished by the Treasury.
8. November 2011 at 17:30
Scott.
You asked for a reminder of the proposal I offered–which was really a modestly-tweaked version of a proposal offered by Matt Yglesius. (As Statsguy points out, Matt’s proposal was actually preceded by a similar proposal by Joe Gagnon.)
Under Matt’s version, the Fed buys $300 billion in bonds and lights a match to them. The Treasury takes that $300 billion and sends a $1,000 check to each man, woman and child in America. And both the Fed and the Treasury announce up-front that they intend to keep doing that until unemployment falls below a certain level.
The variation I offered was to drop the unemployment target and replace it with an NGDP level (and/or a we-need-to-make-up-ground growth rate) target.
At first you seemed to dislike the idea, but in our later discussions you seemed to warm to the plan in principle, although you thought all this talk of helicopter drops would make the inflation-worriers absolutely go off their rockers.
Your thinking, I believe, is that you would be happy with just the Fed just embracing NGDP targeting and using its well-established tools. At the time, your implicit argument seemed to be that you believed these well-established tools would work on their own.
Perhaps, though, Statsguy’s arguments have convinced you that true cash injections might be necessary after all.
I will take this opportunity to offer one more reminder–in the hopes you won’t forget as events unfold in the coming years… The big issue I am worried about is the historical relationship between a growing economy and growing HH_Debt/GDP.
In previous discussions, you said your guess is that NGDP level targeting would allow economic growth to occur without requiring growth in HH_Debt/GDP. I am less confident.
My point is that with massive amounts of “excess” household debt in the system, having a mechanism to inject cash into households’ pockets may be very important for two reasons.
The first reason is what Statsguy’s game-theory analysis laid out: Households that are over-indebted and have taken significant hits to their wealth probably need to have more cash in their pockets before they are going to provide the demand that will get NGDP targets back on track (and the main thing they want to do with this cash is to put in their stash of cash holdings so it shouldn’t be unduly inflationary).
The second reason I think it is a good idea is because, if it does turn out that an expanding economy DOES rely on expanding HH_Debt/GDP (even under an NGDP level targeting regime) sooner or later the Fed is going to want a mechanism that allows it to inject purchasing power into the system in a way that does not require a simultaneous expansion of debt.
I am more than happy if we take all of this one step at a time. Let’s first get the Fed to fully embrace a stance of NGDP level targeting. Once that happens, then they will be forced to start working through their quiver of tools.
I believe (and Statsguy argues above) that the process of working through those tools may ultimately lead to direct injections of cash. In my book, such an outcome is vastly preferable to standing by and allowing the traditional (long and painful) processes of debt destruction to play out.
10. November 2011 at 19:43
Continuing this and from:
http://www.themoneyillusion.com/?p=11510
“Scott Sumner
5. November 2011 at 12:53
Fed Up,, Money is neutral in the long run, it doesn’t affect inequality in any important way.
I don’t understand the problem with steady 5% NGDP growth-so what if it required a continual increase in M?
Ask the averge guy making $20,000 if they’d prefer the lifestyle of someone making $200,000. Believe me, there are and always will be plenty of unmet wants. We heard the same argument in the Great Depresison, which later looked silly.
*****
Fed Up
6. November 2011 at 22:45
“Fed Up,, Money is neutral in the long run, it doesn’t affect inequality in any important way.”
I’ll need some more explanation there, but at first glance, I don’t believe that is going to be right. Some of it will come down to the definition of “money”, a term I would like to see done away with because there are too many definitions of it.
And, “I don’t understand the problem with steady 5% NGDP growth-so what if it required a continual increase in M?”
That is an excellent question, and one I was going to basically ask in another way. First, I’m going to consider M to be medium of exchange because I have bought things with currency and/or demand deposits and sold things for currency and/or demand deposits out in the real economy. I have never bought or sold for central bank reserves although they may “go along for the ride” with a demand deposit if it ends up at another bank.
There is nothing wrong with a continual increase in medium of exchange per se. It depends on how that happens, specifically if it is coming from more and more demand deposits from loans (making it debt). If it is, you now have to consider the possibility that the amount of medium of exchange can fall from debt repayments (principal) and debt defaults when the economy needs it to be at least flat if not rising. I believe that also means wealth/income inequality is more likely. You get accounting wise:
savings of the rich = dissavings of the gov’t (preferably with debt) + dissavings of the lower and middle class (preferably with debt)
You get budget wise:
the rich have positive real earnings growth, the gov’t and the lower and middle class have negative real earnings growth, which they make up for with more debt
I’m also going to make a Minsky like argument. You get periods of stable NGDP growth, but in reality the debt is making the situation more and more unstable eventually. In other words, if more and more debt does not cause price inflation or is being used to prevent price deflation, there is an imbalance in “someone’s” budget somewhere (negative real earnings growth and more debt).
Next, if you create more medium of exchange from currency/demand deposits with no bond/loan attached, you eliminate the possibility of debt repayments and debt defaults. I believe that also means wealth/income inequality is less likely. You get accounting wise:
savings of the rich + savings of the lower and middle class = the balanced budget(s) of gov’t + dissavings of the currency printing entity with currency/demand deposits and no loan/bond attached
You get budget wise:
the rich have positive real earnings growth, the lower and middle class have positive real earnings growth, the gov’t has balanced budgets, and the currency printing entity has negative real earnings growth, which is makes up for with currency/demand deposits and no loan/bond attached
It seems to me the Minsky argument goes away because there are no imbalances in anybody’s budget somewhere.
More later!
*****
Fed Up
10. November 2011 at 19:33
More on “I don’t understand the problem with steady 5% NGDP growth-so what if it required a continual increase in M?”
If I’m stating your argument correctly, you basically believe the period from about 1990 to 2007 or 2008 was a stable NGDP growth path. Then something went wrong or was not handled correctly in 2007 or 2008. My argument is that the period from about 1990 to 2007 or 2008 was an UNstable NGDP growth path. It all comes down to whether someone believes that too much debt can cause medium of exchange problems that build up over the years.
To do that, you would need to take a step down from macroeconomics to microeconomics and budgeting. The budgets of these three(3) entities all are big enough to affect the macroeconomy. The three(3) entities are the gov’t, the rich, and the lower and middle class. The gov’t is one(1) entity so it acts the same way, and the rich basically act as one(1) entity because they are trying to get richer. If enough of the lower and middle class act the same way, it can be considered one(1) entity. If the budget of the entity is big enough and enough of the entities act the same way, then the entity can affect the macroeconomy. So since about 1982, it has been negative real earnings growth for the lower and middle class and more debt. Eventually, that entity goes bankrupt.
Lastly, a problem that was caused by negative real earnings growth can’t be solved by more negative real earnings growth.
“Ask the averge guy making $20,000 if they’d prefer the lifestyle of someone making $200,000. Believe me, there are and always will be plenty of unmet wants. We heard the same argument in the Great Depresison, which later looked silly.”
Although it is different for different people, there is some point ($75,000, $150,000, $200,000) where people have enough. Warren Buffett is a good example. Your policy seems to be to affect the workers indirectly by making the rich richer. When the economy goes from mostly supply constrained to mostly demand constrained, that is probably not going to work. You are probably going to need to affect the labor market directly.
“Believe me, there are and always will be plenty of unmet wants.”
Then why do people voluntarily retire (meaning I don’t believe you)?
10. November 2011 at 20:07
Brett, I still think conventional tools are way better. What I might have said is that if we try conventional tools (zero IOR, a higher inflation target, plus $10 trillion in bond purchases) and it doesn’t work, then we could go with your plan.
Fed up, I define money as the base, which is not really debt. So it can rise at 5% a year forever, I don’t see the problem. If people are accumulating too much debt, due to moral hazard, that should be addressed through regulation, not monetary policy.
10. November 2011 at 20:18
“Fed Up, I’m not sure I followed his saving arguments.”
What can be done with medium of exchange that has been earned?
1) consume
2) invest (national income definition)
3) buy a financial asset (saving and looking for a return on that saving)
4) hold the medium of exchange (saving)
I believe what you are saying is that overpaying for an asset will cause some entity to stop saving and spend in the present or will cause some entity to dissave and spend in the present. That may not to probably won’t happen with wealth/income inequality, too much debt, and an economy that has gone from mostly supply constrained to mostly demand constrained.
And, “Fed Up, It’s convenient for them to have some assets when they need to reduce the base. That way they don’t have to constantly ask the Treasury to give them some T-securities.”
Skipping our disagreement about monetary base and imo, that asset should be something I’m going to call “intangible” for right now for the reduction. Next, I don’t want assets purchased because they could go down in value. The savers (many of them excess savers) will now start driving up prices of things like oil with NO INCREASE IN FINAL DEMAND. That means the producers won’t produce more. The fed will now say we have to issue more debt to get the people out of things like oil to save. In the end, it will be make the rich richer and the lower and middle class will be stuck with the higher taxes and/or benefit cuts.
And, “Regarding vault cash, the definitions change over time. But surely if the Fed went to negative IOR it would include vault cash, otherwise the plan would make no sense.”
If I’m remembering correctly, SRW had a comment about getting around a negative IOR on vault cash here:
http://www.interfluidity.com/v2/2110.html
10. November 2011 at 20:34
“Fed up, I define money as the base, which is not really debt. So it can rise at 5% a year forever, I don’t see the problem. If people are accumulating too much debt, due to moral hazard, that should be addressed through regulation, not monetary policy.”
The problem is the demand deposits from debt are medium of exchange that can be defaulted on or destroyed with debt repayments. That means the amount of medium of exchange can fall when it needs to stay at least the same or rise. It seems to me what you want to describe is an economy with no debt (which I would like to see). It is the debt that is creating time differences between spending and ASSUMED earning and allowing for wealth/income inequality that is the problem.
The too much debt is not from moral hazard per se. It is from the lower and middle class not being paid enough (negative real earnings growth). That is why the debt has not caused price inflation.
10. November 2011 at 20:54
“Fed up, I define money as the [monetary] base [currency plus central bank reserves], which is not really debt.”
[] are my additions.
Are you sure about that statement? Look at the fed’s balance sheet pre-crisis. Hint: what does debt look like on a balance sheet?
12. November 2011 at 14:17
Fed Up. I don’t understand your 4 categories–saving and investment are two sides of the same coin, not different categories.
Regarding base money being debt, that only with IOR, which is a recent development. It’s carried on the books as a liability for historical reasons–it used to be redeemable into gold. No longer.
I don’t see a problem with “debt” per se.
12. November 2011 at 21:18
The way saving and investment was explained to me.
I make $2,000 per month. I only spend $1,800 per month. I save $200 per month in medium of exchange. I put the $200 per month “under the mattress” (hold the medium of exchange), or buy a CD, or buy a bond, or buy a stock. All those are not investing by national income definitions. I buy some equipment to start/expand a business to produce something with the $200 per month. That is investing by national income definitions.
Regarding base money being debt, I get a mortgage at a bank. The bank has medium of exchange (demand deposits) as a liability and the loan/bond as an asset making the total transaction debt. The same idea applies to the central bank. I believe all the central bank’s liabilities (currency and central bank reserves) were “backed” with the assets of gov’t bonds. That makes it debt. Think about the accounting of how new currency is created (not the central bank trying to force it into existence).
13. November 2011 at 09:51
More about base money being debt, just because the liability of the central bank/a bank pays 0% interest does not mean debt is not involved.
13. November 2011 at 11:10
Fed up, You said;
“I make $2,000 per month. I only spend $1,800 per month. I save $200 per month in medium of exchange. I put the $200 per month “under the mattress” (hold the medium of exchange), or buy a CD, or buy a bond, or buy a stock. All those are not investing by national income definitions.”
If the stock you buy finances new investment, it is investment. If you buy a stock from another person in the secondary market, your increased saving exactly equals his dissaving.
S&I are just two sides of the same coin. If it doesn’t seem that way you are making an error.
13. November 2011 at 14:46
“If the stock you buy finances new investment, it is investment.” OK.
“If you buy a stock from another person in the secondary market, your increased saving exactly equals his dissaving.”
OK, but what happens if the stock price goes down?
Also, what happens if I hold currency “under the mattress”, hold a demand deposit in a checking account from a bank that I expect to have 1-to-1 convertibility to currency (FDIC insurance), or “buy” a new CD from a bank that I expect to have 1-to-1 convertibility to currency (FDIC insurance) along with guaranteed interest? It seems to me in all three(3) cases, the velocity of that amount of medium of exchange goes to zero(0).
“S&I are just two sides of the same coin.”
I believe you are referring to S=I. If so, is that correct for a barter economy?
And more importantly, can having a medium of exchange somehow cause a disruption of that identity?
13. November 2011 at 19:35
Fed Up, I don’t have time to go into all this in detail, but yes, S=I in any economy, monetary or not. Saving is simply defined as funds diverted into investment projects. So they must be equal.
14. November 2011 at 19:20
“Saving is simply defined as funds diverted into investment projects.”
If I take my $200 per month that I save and hold demand deposits with FDIC insurance or currency, I don’t see any investment or anything else happening. I believe that will also be true for a CD from a bank with FDIC insurance. I’m thinking a CD from a bank is a daily renewing demand deposit at a fixed rate. I’ve ignored the interest payment, if any.
The way it was explained to me is:
current account deficit = gov’t deficit plus private deficit
With that equation, there is a stock of medium of exchange that flows. Saving in the medium of exchange or medium of exchange like can disrupt the flow.
With S=I, the same idea applies. There is a stock of medium of exchange that flows. Saving in the medium of exchange or medium of exchange like can disrupt the flow.
Actually, “current account deficit = gov’t deficit plus private deficit” is a disaggregation of S=I.
15. November 2011 at 12:38
That equation merely says that the current account equals national saving minus national investment. It’s still true that S=I at the global level, but not in each country.
16. November 2011 at 18:14
I’ll see what else I can find about currency and central bank reserves actually being debt and saving in the medium of exchange or medium of exchange like.
17. January 2012 at 10:13
On real AD NOT being unlimited:
http://finance.yahoo.com/blogs/daily-ticker/finally-rich-american-destroys-fiction-rich-people-create-152949393.html
“The customers of most companies, Hanauer points out, are ultimately the gigantic middle class “” the hundreds of millions of Americans who currently take home a much smaller share of the national income than they did 30 years ago, before tax policy aimed at helping rich people get richer created an extreme of income and wealth inequality not seen since the 1920s.
The middle class has been pummeled, in part, by tax policies that reward “the 1%” at the expense of everyone else.(It has also been pummeled by globalization and technology improvements, which are largely outside of any one country’s control.)
But, wait, aren’t the huge pots of gold taken home by “the 1%” supposed to “trickle down” to the middle class and thus benefit everyone? Isn’t that the way it’s supposed to work?
Yes, that’s the way it’s supposed to work.
Unfortunately, that’s not the way it actually works.
And Hanauer explains why.
Hanauer takes home more than $10 million a year of income. On this income, he says, he pays an 11% tax rate. (Presumably, most of the income is dividends and long-term capital gains, which carry a tax rate of 15%. And then he probably has some tax shelters that knock the rate down the rest of the way).
With the more than $9 million a year Hanauer keeps, he buys lots of stuff. But, importantly, he doesn’t buy as much stuff as would be bought if that $9 million were instead earned by 9,000 Americans each taking home an extra $1,000 a year.
Why not?
Because, despite Hanauer’s impressive lifestyle “” his family owns a plane “” most of the $9+ million just goes straight into the bank (where it either sits and earns interest or gets invested in companies that ultimately need strong demand to sell products and create jobs). For a specific example, Hanauer points out that his family owns 3 cars, not the 3,000 that might be bought if his $9+ million were taken home by a few thousand families.
If that $9+ million had gone to 9,000 families instead of Hanauer, it would almost certainly have been pumped right back into the economy via consumption (i.e., demand). And, in so doing, it would have created more jobs.
Hanauer estimates that, if most American families were taking home the same share of the national income that they were taking home 30 years ago, every family would have another $10,000 of disposable income to spend.
That, Hanauer points out, would have a huge impact on demand “” and, thereby job creation.
It’s time we stopped mouthing the fiction that “rich people create the jobs.”
Rich people don’t create the jobs.
Our economy creates jobs.
We’re all in this together. And until we return to more reasonable tax policies that help the 99% instead of just the 1%, our economy is going to go nowhere.”
I keep trying to say the problems of the U.S. economy are wealth/income inequality, too much currency denominated debt, and going from mostly supply constrained to demand constrained.
NOTICE the ONLY 3 cars. Car demand NOT unlimited.
***
Also,
http://financiallyfit.yahoo.com/finance/article-113833-11747-3-from-a-100k-plus-income-to-less-than-35k-and-happier-than-ever
‘From a $100K-Plus Income to Less Than $35K and Happier Than Ever’
“Fortunately, we had no mortgage or car loans. Before my layoff, we consolidated our credit card debt and opted for a small home equity loan.”
The blessing of having very little or no debt so there are no unrealistic assumptions about future wage income.
And, “Re-Frame Your View
Losing my job was not the heartbreak I thought it would be. Not working in an office environment has lowered my income but also lowered my outlay. I no longer have enormous bills from dry cleaning. I no longer go out to eat because I don’t have time to cook. I now employ not one, but two crockpots and have cut our grocery spending by cooking that way.
Being a freelance writer and speaker now, I spend more time at home with my family. We may have lost a lot of income, but we have gained a lot of family togetherness. And that, to us, is priceless.”
Demand for eating out NOT UNLIMITED.
I don’t have the link, but there was one story of a lady who because of the recession stopped eating out so often. She lost 30 pounds. Hopefully, she continued eating more often at home.
Demand for eating out NOT UNLIMITED.
17. January 2012 at 17:13
Fed up, Real AD is a meaningless concept. What matters is nominal AD and real AS. If we can produce the goods, the Fed can assure enough demand to buy the goods.
9. August 2013 at 13:53
For reference:
zyxzyxooxmi