were well secured in making the loan that we did. in a second example in october of 2008, as i'm sure you all know, aig was very, very close to failure. this, again, was the largest insurance company, perhaps amon. let me just talk a bit about that case. aig was a complicated company. it was on the one hand a multinational financial services company with many constituent parts including a number of insurance companies, global insurance companies, but it had a part of the company which was called aig financial products that was involved in all kinds of exotic derivative and other financial markets including credit insurance that it was selling to the owners of mortgage backed securities. so when the mortgage back securities started going back it became evident that aig was in

big trouble and its counter parties demanded cash or refused to fund aig and coming under tremendous pressure. now the failure of aig in our estimation would have been basically the end. it was interacting with so many different firms. it was so interconnected with both the u.s. and the european financial systems, global banks. we were quite concerned that if aig went bankrupt that we would not be able to control the crisis any further. fortunately perspective of lender of last resort theory, aig has taken a lot of losses in its financial products division but underlying those losses was the world's largest insurance company sponsorship it had lots and lots of perfectly good assets and as a sell it had collateral which you could offer to the fed to allow us to make a loan to provide the liquidity it needed to stay afloat. and so to prevent the collapse

of aig, we used aig assets as collateral, and loaned aig $85 billion. obviously a fairly serious amount of money. later treasury provide additional assistance to keep aig afloat. and, again, while highly controversial it was both we thought legitimate in terms of lender of last resort theory because it was a collateralized loan and the fed has been fully paid back and secondly it was a critical element in the global financial system. over time as i said aig stabilized, has repaid the fed with interest. the treasury still owns a majority share of its stock. but it has, aig has beens well. it's been in the process of doing that. le to emphasize what we had to do with bear

stearns and aig is obviously not a recipe for future crisis fst very difficult and in many ways distasteful intervention that we had to do on we needed to do that to prevent the system from collapsing. but fundamentally wrong with a system in which some companies are too big to fail. ifat it knows that it's going get bailed out even putting aside the it' to other companies but even beyond that obviously they have an incentive to take big risks. if the risks big money if they don't pay off the government will save us. that's a situation we can't tolerate. so as i'll describe more next time, the problem we had in

september of 2008 was we really didn't have any tool, legal tools, policy tools that allowed us to let aig and these other firms go bankrupt in a way that would not have incredible damage, create incredible damage on the rest of the system and, therefore, we chose thelers of two evils and prevented aig from failing. with that being said that going forward we wanted to make sure this never happens again. we wanted to be sure the system is changed so that if a large systematically critical firm like aig comes under this kind of pressure in the future that there will be a safe way so let it feel. it can feel and the consequences of its mistakes can be borne by management and shareholders and creditors but in doing so it doesn't bring down the whole financial system. i'll talk more next time about the progress we made

collectively in instituting a system that will, i hope, eventually at least end too big to fail. so finally let me just say a couple words about the consequences of the crisis. we did -- we did stop the meltdown. we avoided what would have been i think a collapse of the global financial system. that was obviously a good thing. but to give you a sense, one thing that i was always sure of and i think the federal reserve was always sure of is that collapse of some of these big financial firms was going have some very serious collateral circumstances. there were people arguing even as late as september of 2008 well why don't you just let the firms collapse. you know, system can take care of it. bankruptcy. we have a bankruptcy code. let them fail. and, you know, we never thought that was really a good option.

particularly if the whole system collapsed we would have had extraordinarily serious consequences. as it was even though we prevented the total meltdown there was still obviously, as you know, very serious collateral impact on not just the u.s. economy but the global economy as well. so following the crys is, even though the brought under control, the u.s. economy and much of the global economy went into a sharp recession. the united states, gdp fell by more than 5%. which was a quite deep version. there's some other statistics. 8.5 million people lost theirnt. very consequential impact. as i said, this was not just the u.s. situation. was, in fact, kind of an average recession. many countries around declines, particularly those dependent upon international trade.

so it was a global slowdown. and as all this was happening, fears0s depression were very real. so nevertheless the great depression was much worse than the recent recession. and i think the view is increasingly gaining acceptance that without the forceful policy response that stabilized the financial system in 2008 and early 2009, we could have had a much worse outcome in the economy. here's a couple of indicators just to close with a couple of graphs here. so this -- i think this is an interesting graph. this shows the stock market. the blue line starts in august 1929 which is the peak of the stock market before the great depression. the red line is the more recent stock prices, it starts in

october of 2007. and then the graph shows you the evolution of stock price in the depression period in the blue and in the more recent period in the red. and the thing which is pretty striking here is that for the first 15 or 16 months stock prices in the united states behaved pretty much in this crisis as they did in 1929. in 1930. but about 15 or 16 months into the recent crisis, which would have placed it in the early 2009 about the time the financial crisis was stabilizing look what happened. in the depression era the stock price kept falling and as i mentioned in the end stock prices lost 85% of their value. in the united states by contrast stock prices recovered and began a long recovery and now are more than double than where they were three years ago.

this is industrial production, measure of output. again, the red is the more recent data, the blue is the depression era data. you can see in this case that the fall industrial production was not quite as severe, quite as fast as in the depression but you ge phenomenon that about 15 or 16 months into the episode, crisis was brought under control industrial production bottomed out and began a period of steady recoveryprsion the collapsed continued for several more years. okay. so that is a very rapid overview of the crisis of '08-'09. in lecture four we'll talk about the aftermath, the recession, how did monetary recession, why recovery been relatively sluggish, what has happened to financial regulation to try to make sure this never happens l

fed taken from this experience? okay. questions? yes. [ inaudible ] >> why do you think these insti [ inaudible ] so much risk -- [ inaudible ] couple of reasons. one reason was simply the fact that firms were probably too confident about house price

increase. house prices were likely to keep rising. and in a prices are rising these are not bad products because people can afford to pay for a year but then they can refinance into something more stable than might be a way to get people into housing. but, of course, the arriving was that house rises wouldn't keep rising and that's ultimately what happened. the other aspect of this was that the demand for securitized products grew very substantially during this period. in part there was a large international demand from europe and from asia for high quality assets. and the ever clever u.s. financial firms figured out that they could take a variety of different kinds of underlying credits whether subprime mortgages or whatever and through the miracles of financial engineering they could

create from that at least some securities that would be high quality, it would be rated aaa in which they could then sell abroad to other investors. unfortunately that sometimes left them with the remaining bad pieces which they kept or sold to some other financial firm. so there were trends in the financial markets, i think, including over confidence about their ability to manage those risks. a belief that house prices would probably keep rising. a sense that they could even after they made those mortgages they could then sell them off to somebody else and that that other person or other investor would be willing to acquire them. there was a big demand for quote safe assets. for all those reasons it was actually a very profitable activity while it lasted and only when the house prices began to fall did it become a big loser. yeah.

[ inaudible ] >> it kind of remind me of the volcker rule because as i understand it the volcker means -- [ inaudible ] so i'm wondering what you think about that. does that seem counter intuitive. >> the volcker rule is part of the dodd-frank financial regulatory reform which i'll talk more about on thursday which the fed and other agencies are tasked with implementing. the purpose of the volcker rule, as you said, is to reduce the risk of financial institutions by preventing banks and their affiliates from doing quote

proprietary trading which means doing short term trading on their own account. so from taking those kinds of risks. now the law recognizes that there are legitimate exceptions to, for why banks might want to acquire short term securities and those include, for example, hedging against risk, but one particular exception is to make markets. to serve as intermediaries who buy and sell in a particular market. that's exempted f and one of th of implementing this rule is trying to figure out how to set a set of standards that allows the so-called exempted or legitimate activities like market making and hedging while ruling out the proprietary trading. that's obviously very difficult. we're working on that. we put out a rule. we got thousands of comments. we're looking at that. trying to figure out how best to

do that. the point you raised liquidity in markets is important. during the crisis, it was much worse, much worse problem than just a little bit lack of trading volume. you had big financial institutions unable to fund themselves, unable to find the funding to support their asset positions, the assets that they held which left them with one or two possibilities, either defaulting because they didn't have enough funding, ortook is assets as quickly as possible which in turn spread the panic. if there's a huge sellers market for commercial real estate bonds, that's going to drive the price down very sharply and anybody else buying those bonds find their financial position being eroded and that causes

pressure on them. i didn't use contagion. contagion like an illness context is the spreading of panic, the spreading of fear from one market, from one institution to another and contagion was a major problem in many financial panics but certainly in this one and that was one of the mechanisms that led the funding pressures to jump from firm to firm, and create such a broad based, broad based problem. daniel. anybody have -- [ laughter ] >> i had a question specifically about global collaboration during the financial crisis. you talked about the g7 in 2008.

specifically as we saw multinational corporations begin to be on the brink of failure, what pressures came from the international community, the decision to bailout aig? >> well there weren't any real pressures. everything was happening too fast. i think, in fact, you know, one area where collaboration was not as good as we would like was exactly dealing with some of these multinational firms, for example there were problems between the uk and u.s. over lehman brothers failure, for example, inconsistencies which causedro creditors of lehman. so, one of the things that we're trying to do the dodd-frank legislation which includes as i mentioned before, includes provisions for safely allowing large financial firms to fail, one of the complexities

there is many of the firms that this would be applied tore multinational firms. maybe not just two or three countries, maybe dozens of countries, and so collaboration with other countries in figuring out how we would work together to help a large multinational firm fail as safely as possible internationally. going on now we tried during the crisis to cooperate and mostly ad hoc way. we were in touch with regulators in the uk and elsewhere. but given the time frames and the lack of preparation, you know, we didn't -- didn't do as much as we would be able to do with a lot more lead time. i think that was a weakness of international collaboration. for the most part, though, countries cooperated in dealing with the financial institutions that were based in their own countries. aig was an american company. we dealt with that.

whereas a company like dexia which was a european by the eur. also, there was a lot of cooperation between central banks and i may have a chance to say a bit more about this. but there were of european banks that used dollars, that needed dollar funding as opposed to euro funding. thundi both because they held dollar they made dollar loans,upport t which is often done in dollars and they needed dollars. the pe provide dollars so what we did was what we called a swap where we gave the european central bank dollars they gave us euros. they took the $s we gave them and lent them to european banks taking off the dollar funding pressure and easing $funding pressures around the world. so those swaps which are still

in existence now because of the recent issues in europe were an important example in october oft before, right as the crisis was intensifying, the federal reserve and i think five other central banks all announced interest rate cuts on the same day. so we coordinated even our monetary policy. about you, so we did our best to coordinate. there were some areas where like working on multinational firms where you know, a lot more preparation was needed, and we are still working on those things cooperatively today. noah? >> i was, my name's noah whittier. on the off balance sheet vehicles that were being used and sort of why they were allowed to, you know, keep that much information off their books. >> well, it has to do with

accounting rules basically. you create this separate vehicle. and the bank might have substantial interest in that vehicle. it might, for example, have a partial ownership. promises to provide credit support if it goes bad or liquidity support if it needs cash. but it doesn't have under the rules that existed at the time, if the amount of control that the bank had on this off balance sheet vehicle was sufficiently limited then according to the accounting rules, it could treat it as a separate -- a separate orgati sheet. and that allowed the banks to get away with somewhat less capital, for example, than they would have had to carry if they'd had all these assets on their own balance sheet. now, one of the many good developments since the crisis is that these rules have been reworked. and many of the off balance

sheet vehicles that existed during the -- before the crisis would no longer be allowed. they would have to be consolidated, which means brought back onto the balance sheet, made part of the bank's balance sheet, have appropriate capital and so on. so sthoes practices are not completely gone but the accounting rules have greatly toughened up the circumstances in which something a bank can put off a balance sheet into a separate investment vehicle. max? >> mr. chairman, you mentioned several large firms that came under pressure in 2008 and alsoo big to fail. my question was, where do you draw the line between bailing out a bank and allowing it to fail? is it arbitrary or is there some sort of methodology that the bank goes by? >> that's a great question. first of all, i want to resist

that work doctrine a little bit. these firms proved to be too big to fail in the context of a global financial crisis. that was a judgment we made at the time based on their size, their complexity, interconnectedness and so on. it was not something we ever thought was a good thing, and one of the main goals of the financial reform is to get rid of it because it's bad for the system. it's bad for the firms. it's unfair in many ways, and it would be a great accomplishment to get rid of too big to fail. it's not something we advocate or support in nep way. we were forced into a situation where we were having to choose the least bad of a number of different options. now, it's a good question. i think in the case of the -- of the during the crisis, you know, we had basically to make

judgments on a case by case basis, and we were trying to be as conservative as possible. i think in the case certainly of aig, there was really not much doubt in our minds that this was a case where action was necessary if at all possible. lehman brothers was in itself probably too big to fai sense that its failure had enormous negative impacts on the global financial system, but there we were helpless because it was essentially an insolvent firm. it didn't have enough collateral to borrow from the fed. we can't put capital into a firm that's insolvent. this was before the t.a.r.p. or anything else had provided capital that the treasury could use. so we really just had no legal way to do it. if we could have avoided that, we would have done so. so it was somewhat ad hoc i think although the two cases where we intervened, bear stearns and in aig i think the

case was pretty clear, given not only the firms themselves but also the context, the environment that was -- that was going on at the same time. now, interestingly, we've had to get much more into this issue since the crisis because there are a number of different rules and regulations which actually require the fed and other regulatory agencies to make some determination about how systematically critical a firm is. for example, the new bazel three capital requirements require the largest most systemically critical firms to have a capital surcharge. they have to hold more capital than firms which respect ashat international bank regulators have workeding to to try to set up a set of criteria relating to size, complexity, interconnectedness, derivatives, a whole bunch of criteria help determine how much extra capital they want have to hold.

likewise, the fed now when it approves a merger of two banks, it has to evaluate whether the merger creates a systemically more dangerous situation. so, we have worked hard and we have put out criteria that were describes some of the variety of criteria including some numerical thresholds that we look at to try to figure out if a merger creates a systematically critical firm which if it does, we're not supposed to allow that merger to happen. so the science of doing this is progressing. it's still very in its infancy. but again, in the crisis, our actual interventions were limited to well, principal interventions were bear stearns and aig along with other agencies. we also provided assistance to a couple of other other institutions but nothing to the extent that the aig situation

involved. but we are looking very seriously at this, and indeed, now that the fed has become much more focused on financial stability, we are -- we have a whole division of people working on various metrics, various indicators, both to try to identify of risks to the system and also to try to identify firms that need to be you know particularly carefully supervised and maybe hold extra capital because of their -- the potential risks that they bring to the system. david? >> thank you, mr. chairman. my name's david pomeroy. one vulnerability you mentioned was credit agencies were assigning aaa ratings to securities that carried much more risk than perhaps might warrant. it seems like the incentives would be aligned for the buyers to seek out ratings that were more accurate because they would be taking on more risk. was there a systemic problem as

far as how ip sentives were aligned within the credit ratings system that allowed these faultyatings to prove mull gate? >> you identified one of thu wo that somehow instead of thecutye one who hires and pays the credit rater, you would think that it would be in the interest of the buyers who after all, are the ones bearing the risk to band together somehow and pay the credit rater to give them the best opinion they can about what the credit quality is of the security. unfortunately, that model doesn't seem to work. there are very few examples if any that i know of where it works. the problem is a free rider problem. basically, if -- if five investors get together and pay standard & poor's to rate a particular issuance, unless they can keep that many completely secret, anybody else can find out what the rating was and then

they can basically take advantage of that without the having to pay, having to be part of the consortium that paid. so there have been a lot of ideas out there how you can reyou can have tour the payment system to create better incentives for credit raters. but it is a challenging problem because again, just obvious solution of having the investors pay only works if the investors collectively can share the cost and somehow keep that information from being spread among other investors. okay. 2:00. i'll see you on thursday to talk about the aftermath of the crisis. thank you. [ applause ] >> stud,do our usual two questions to the discussion boards so we can pick up and have discussion on these issues next week, that would be