Basel III and Risky Banking Behavior: Too Little, Too Lenient, Too Late?

Knowledge at Wharton

Ask about Knowledge at Wharton

Episode Description

KNOWLEDGE AT WHARTON ARCHIVES: As the world haltingly recovers from the recession, regulators are struggling to modify the financial system to prevent another crisis. The latest effort: stricter capital requirements to help prevent large banks from collapsing under the weight of unexpected losses. While the new proposals -- called Basel III -- are designed to reduce risk-taking by assuring that banks continue lending in a weak economy, Wharton Finance professors Richard Herring, Franklin Allen, and others are skeptical that the new proposals will accomplish this goal.

More on this interview: https://knowledge.wharton.upenn.edu/article/basel-iii-and-risky-banking-behavior-too-little-too-lenient-too-late/

#basel3 #banking #wallstreet #financialcrisis πŸ“‚ Finance & Accounting

Read the full transcriptHide the full transcript
  1. [Music] We're meeting today with Wharton Finance Professor Richard Harring to discuss the
  2. recently approved Basel 3 Accords and how they will affect international bank capital requirements. Thanks for joining
  3. us today. My pleasure. I guess we should start by saying that the Bosel Committee
  4. does not refer to it as Basel 3. They have studiously avoided the term. uh they'd rather think of it as just sort
  5. of a addition to Basel 2, kind of like a Windows program that is issued as 2.4 or
  6. something. Um but it is uh in many ways a marked change. Um it's tidying up some
  7. mistakes that they made in Basel 2, which isn't to say that they've actually corrected them. Um but it also is
  8. reflective of a new world governance system for uh the international financial system. Um before last year
  9. really uh the top level governing body was the group of 10 leading countries and those included only one Asian
  10. country Japan and they were all wealthy countries. Um, it was actually 11 because they put Switzerland in because
  11. it's it's a financial center, but apart from that, it was really a G10 creature and it was a G10 central banks and the
  12. Basel committee reported the G10 and um that made sense when it was formed in 1974 because those countries did control
  13. about 80% of uh world financial activity. But the whole balance of economic power and wealth has shifted to
  14. Asia in a dramatic way over the last few years. So that it no longer really made
  15. sense to make rules for the world financial system with a group that was that unrepresentative. Um so we now have the
  16. group of 20 that includes um all of the brick countries Brazil, Russia um uh India and China as well as Indonesia and
  17. other leading uh developing countries. So we have about the same proportion of world financial activity covered
  18. actually a whole lot more of the world's population and um a whole lot more of the world's economic output. Um but it
  19. does make for a much more heterogeneous group. It was surprising in a way that they
  20. were able to agree on what they wanted to do with the financial system because it's usually much more difficult to to
  21. reach consensus in a heterogeneous group than a small one that's homogeneous. But
  22. last year um they met in Pittsburgh. Um it was uh it's hard for them to meet because they're a target for everybody.
  23. And so Pittsburgh is kind of a nicely protected place because you can sort of close down the bridges and and uh have
  24. it uh uh safe. And I think Obama wanted to show off a US city that had kind of regenerated itself. But they actually
  25. came out with a very clear set of things they wanted to improve with the world financial system and they created a
  26. body. They transformed the financial services forum into the financial services board that was charged with
  27. really carrying out those things and dividing it among the existing structures we have. Well, the Basel
  28. committee which used to be basically the G10 plus Luxembourg in Switzerland has now has 27 members and so the uh
  29. decision process is in some ways much more reflective of a a broader range of interests. Well, the problem that they
  30. set out to fix, uh, one of the most obvious problems, and there were numerous ones, is that, um, they had
  31. made a really serious mistake in Basel 2 in ignoring the numerator in the minimum
  32. capital standard, the numerator and the middle and the the riskadjusted capital standard was supposed to um compare
  33. capital as the numerator to risk weighted assets. And the risk weights were supposed to
  34. indicate how risky the asset was. Well, there were huge problems in the risk weights. Uh we can get to that later.
  35. But the real problem with the capital definition is that they didn't even reconsider it. And the reason they
  36. didn't reconsider it is that when they originally negotiated Basel 1, about 85% of the time was taken up trying to agree
  37. on a definition of regulatory capital. Capital is one of those words that can have at least 10 or 12 different
  38. definitions. They started with capital to regulate banks because every single country had a capital ratio. But when
  39. you started looking at them in detail, they were all very different. In the United States, we had a huge proportion
  40. of loan loss reserves. Uh the Japanese counted unrealized gains on uh real estate and stock. Um the French that had
  41. just nationalized their whole banking system counted lots of preferred shares. And so that's why it took so long for
  42. them to come up with a right because and agreed. They were smart enough to know that if um you agree on a definition of
  43. capital that doesn't look like the capital we have, you're going to make it hard on our banks. Well, the Germans
  44. were sort of one end of the spectrum at that time, although they've now switched. And they thought that they we
  45. would ruin the rigor of the German regulation, in fact, banking regulation in general, if we went much beyond
  46. equity. and everybody else wanted something in addition to that. So they resolved it by having two kinds of
  47. regulatory capital. They ended up eventually with three kinds, but the third kind was never really used. Um but
  48. the first kind reflected the German view and it was originally shareholders equity um
  49. retained earnings and a very restricted kind of hybrid capital. It was a non-cumulative perpetual preferred
  50. security, which meant that if you missed a dividend, you didn't have to make it up. And perpetual meant that you never
  51. had to repay the principal. So like equity, it was with you to absorb loss without driving you out of business. The
  52. problem with debt contracts is that they can absorb loss, but only if you fail. And so all of this other stuff was
  53. really not very useful. Well, over time, banks, and this is a concern with the the uh very long leadin period to uh
  54. Basel 3, banks sort of whittleled away at that tier one definition because it was it was they considered the most
  55. burdensome kind of capital to have because issuing equity is is expensive. Um, and so what was originally a 4%
  56. requirement became generally sort of a 2% requirement with a lot of other stuff that was nothing like equity. Probably
  57. the most egregious example of that was that banks were able to count as cap as tier one capital deferred tax assets.
  58. Now, deferred tax assets are what you get when you make losses that you can't deduct because you don't have any
  59. income. Well, it's a kind of crazy thing to count as capital to protect you against going out of business because
  60. that's the time you're least likely to have the expectation of making income. So, that made no sense at all. The
  61. others were driven by trying to take advantage of tax laws because a lot of this is caused by um the asymmetry in
  62. the tax treatment of dividends and interest. Um, in general, governments will permit interest payments to be
  63. deducted for tax purposes, but dividends can't be. Uh, so there is an obvious incentive for people to try to finance
  64. themselves with debt instruments. and investment bankers spent a whole decade trying to invent instruments that were
  65. enough like equity that the regulators would accept them but enough uh like debt that the um tax authorities would
  66. permit them to be deducted. So you ended up with this very debased system. So they they whittleled that down to 2% and
  67. now uh bezel 3 is bringing it back up to 4%. Yeah, because one of the most embarrassing things that happened um
  68. during this crisis from a regulator's point of view is that in every case where a bank went under had to get a
  69. huge intervention. The regulators said well in the last reporting period they actually had way above the minimum
  70. required capital which simply means that our capital requirements were were kind
  71. of uh trivial and unimportant. Whittleled away too much. Yeah. Well, they were they were simply uh not uh sufficient.
  72. Uh and they were not what they were intended to be. Um as a result, the market paid no attention at all to
  73. regulatory capital. And that I guess is in the end the the thing that was uh most demeaning to the regulators because
  74. they could say all they wanted about regulatory capital and it made no difference in the market expectation.
  75. They were looking at tangible equity. Um so one of the first things to fix was the numerator and that meant more higher
  76. quality capital and in the eyes of regulators higher quality capital is equity because that's with you to absorb
  77. losses and let you continue doing businesses. And so they um not only uh have required more equity capital but
  78. they've tried to require three kinds of equity capital actually four. Um, the first one is pretty much like tier 1 was
  79. supposed to be, but rather than being 4%, it's now 4 1.5%. Um, there's kind of exaggerating by saying there's this huge
  80. increase in capital requirements from 2 to 4 and a half. Well, if you look at the original Basel agreement, they're
  81. really only going up half a percent for the minimum, which isn't very impressive. Um, then they're adding
  82. another 2.5% for what they call a buffer zone. And this is an attempt to um have the
  83. market accept the fact that capital can decline without making the bank more unsafe. Um what they may have done
  84. inadvertently however is create uh yet another problem because in principle when a bank starts eating into that
  85. buffer they're going to stop it from paying dividends and there may be sanctions on salaries and that sort of
  86. thing. Um and that could cause the market to lose a lot of confidence. I there should be some sanction, but
  87. ideally you'd like it to be graduated from something trivial to something more serious. Um, and it's not at all clear
  88. that that's the best way to handle the buffer. it may be a lot better to use something that is colloially called a
  89. Koka or a con mandatory convertible uh debt um which is a more complicated thing but it would allow tanks to get
  90. the tax benefit of having uh an interest thing but give you equity when you needed it. So, our banks, do you think
  91. going to voluntarily add yet another buffer on top of the buffer so that they don't Well, yeah, because um you know,
  92. living in the uncertain world, we do um nobody can be certain what you're going to earn in a year and you don't want to
  93. be in the position of having to not pay dividends. So, uh even as we saw with current capital requirements, banks
  94. voluntarily maintain a buffer above the minimum and so they'll probably maintain
  95. a buffer above that. But the third piece of it which is innovative and if it had
  96. been done well would have um been a real improvement is to deal with the fact that having risk adjusted capital
  97. requirements makes them much more proyclical than they otherwise would have been because it means that as the
  98. economy improves risk goes down. So the denominator goes down assuming you have the same capital. In fact, you'll be
  99. increasing it because retained earnings will increase. You have bigger lending capacity to sort of blow up the boom
  100. even larger. But on the way down, as risk increases, the denominator increases and as you get losses, capital
  101. declines. And so, you're going to contract lending even more. So, this is a you see this as a flaw in how they've
  102. designed this at this point? Yes, it's one of the the huge costs of having um riskadjusted capital requirements. Um,
  103. and they've tried to correct this in a very timid way. Um, and what they have done is to put it under pillar two,
  104. which means that nobody's going to know what they do because that's a secret agreement between the bank and and their
  105. supervisor, but they're going to add on a um an amount up to 2%. Could be anything from 0 to 2%. We won't know.
  106. when they think that demand is getting out of hand or the supply of credit is getting out of hand. So that will not be
  107. transparent. That will not be transparent. Um but it's intended to slow down the expansion so you won't
  108. have a pick a crash. The problem is that um inherently it's just very very difficult for a bank examiner making say
  109. 75 or $80,000 to go into a $10 million a year banker and say uh I know you're very profitable but I think that you're
  110. sitting on a lot of risk. Now, it's true they are because it's a standard saying
  111. that the worst loans are put on in the best of times. And we know that's true. But it's not very credible when uh the
  112. bank is in such good shape. Regulators have a lot more uh leverage when things get bad because they have uh some
  113. benefits to give out. So, um I think it's very unlikely that this is ever going to be imposed because it's always
  114. ambiguous about whether we are in a bubble or whether we're we have improving fundamentals because usually
  115. there's a little bit of each underneath. So, you've me you've mentioned a couple
  116. weaknesses here. Could you just kind of Well, then there's a fourth kind of capital before we get to the the rest of
  117. the weaknesses. Um and that is a leverage ratio. Um, and that's the one reason that that our banking system is
  118. actually in less trouble than Europe's because uh almost single-handedly Sheila Bear, the chairman of the FDIC, held on
  119. to the leverage ratio in the United States, which is um a a ratio of tier one capital to unweighted assets, but
  120. taking off balance sheet commitments on in their loan equivalent form. uh and so
  121. that put a limit on how much leverage you could have. Well, it turned out certain of the European banks, some
  122. leading banks like UBS and Deutsche Bank had leverage ratios as high as 50%. So assets to to tangible equity
  123. were so large that you had to be virtually perfect in risk management to be able to survive. Um it Switzerland
  124. became terrified when they figured it out and so they too have put in a leverage ratio because their banks are
  125. literally too big to save. They're both so large outside Switzerland that that it's large relative to their
  126. GDP. Um but the other European banks are not so sure they want to do it. So the compromise that was made is that they're
  127. going to aim for maybe a 3% leverage ratio, but it will be under tier 2. So, we're not going to really know. And they
  128. aim ultimately to make it tier one and an explicit requirement, but that may or may not happen because the Europeans are
  129. clearly dragging their feet. The Germans in particular, which is ironic given their earlier position, but they have a
  130. very weak stateowned sector and it's very hard for them to um raise equity. Uh moreover, the rule uh was made that you
  131. can't count loans from government as tier one and uh that that was a blow to Germany. And so I think the compromise
  132. was they got a pillar two leverage ratio. One of the other things about this that's uh striking is how long a
  133. delay there is before it's actually implemented. And of course the banks were afraid that this would um adversely
  134. affect them economically and so they they won at that buffer time. Um but um well they actually wanted lower
  135. standards. Uh but uh the compromise ended up being buffer time because uh most of the banking um so in fact you
  136. can read them on the BIS website in comments but uh all the the major bank lobbying groups and many of the major
  137. banks sent in arguments that look if you do this you're going to send the world economy into spiraling recession or
  138. depression because um if you raise required capital we're going to have to raise spreads on loans and um we're
  139. going to lend even less than we are now. And when you break down the argument into pieces, there's an element of truth
  140. to what banks are saying in the sense that um from the standard finance point of
  141. view, if they raise equity, they should be compensated in a lower cost of debt. But that's not going to happen for banks
  142. in general because we have guaranteed all of their debt. The one thing we did not do in the United States in any bank
  143. failure was let anybody who had a debt claim on a bank suffer except for layman brothers. And so they're not going to
  144. get the relief in a lower cost of debt that they could expect in a free capitalist system. So uh you know you
  145. can understand why they view that as a pure cost. But the reason that this long phase in period is thought to be helpful
  146. is that it's rather expensive to um do an IPO to raise debt. Um the big transactions costs and given the fact
  147. that banks are kind of in the doldrums at this point. Uh their PE ratios are not very attractive. So they'll have to
  148. pay a lot. Um so what they computed was how long it would take a bank to retain earnings to get to these ratios and you
  149. know you can make various assumptions about how that will happen but uh it it ends up to be something like 2018 or
  150. 2020. Um, and one of the subsidies we do not count, but we should in uh figuring out
  151. how much we're subsidizing the banking system is that we're really subsidizing those retained earnings by keeping
  152. interest rates so low that they have a natural build in spread that's just huge. And uh so there is some reason to
  153. believe they actually will retain the earnings to get there and you know once they're there I think most of the costs
  154. were really transitional costs. Um but there's a risk in that sort of policy. If we were to have a double dip
  155. recession, uh which I think is not impossible, uh we would be in even worse shape than
  156. uh we were before because uh banks would be um still unprepared to to deal with it.
  157. Um and uh the other problem is a political one. The longer you wait to phase in these things, the more
  158. effective bank lobbies are in weakening the fundamental regulations. So, what else is it important to know
  159. about Basil 3? Uh, well, that is it's a very complex package. Um, one thing they've already tried to do and it's
  160. clear they can't do it effectively is correct some of the risk weights. Um, one of the obvious mistakes they made
  161. was underestimating the riskiness of very lowrated debt like subprime u or uh CMOS or C collateralized mortgage back
  162. obligations. Um, and so they tried to fix that um I guess about a year and a half ago uh because it was such an
  163. obvious problem by doubling the risk weight on it. And it took about oh I don't know a month maybe five weeks for
  164. investment bankers to figure out a way around it. And because you can still tanch things, uh, all you have to do is
  165. create a new collateralized mortgage back security. And sometimes they would have it sort of guaranteed so that you
  166. would, uh, recolateralize it again and you could make enough of it AAA so that even with the greatly increased risk
  167. weights on the bottom piece, uh, you could actually lower your capital requirements. So, it's yet another
  168. indication of just how um I think misguided the whole notion is that they can estimate risk with enough precision
  169. to to have much of an impact. Um what what might have been a better approach? Uh I think in the end they're fooling
  170. themselves that they think they can do a lot better than a leverage ratio. uh you
  171. can do some simple things like uh taking government debt out or things that that
  172. are perfectly safe but uh in some ways I think Basel one had a closer it was it had distortions in it we all knew what
  173. they were uh but the distortions in Basel 2 are so subtle that I don't think that regulators will catch on to them
  174. and external analysts will have trouble tracking In in Europe, there are actually 168
  175. different implementation choices a bank can make. So, you'll have no idea how to compare them with another
  176. bank that may have made different choices. And with a lot of stuff coming under pillar 2 that we just don't know
  177. about, it's it's just very hard to know whether a bank that seems to have a lot of capital is very conservative or
  178. whether the regulators are scared to death and are causing them to hold more capital. But the other new thing about
  179. the Basel 3 requirements which isn't yet implemented has to do with um liquidity because one of the clear
  180. things uh actually it was uh overemphasized in my view was that liquidity was a major problem in the
  181. crisis. In fact, I think the regulators wasted a whole year treating it like a normal liquidity crisis when really as
  182. early as as August or September of of 2007, it was obvious that a number of banks were insolvent. But instead, the
  183. regulators chose to flood the market with liquidity. And um so they've decided that that banks need to be responsible
  184. for more of their own liquidity. And they're putting in two ratios for that. Um, one of them is pretty clear. The
  185. other one is sort of not ready for prime time, but they wanted to have a short-term ratio that would um enable a
  186. bank to survive a severe stress test for 30 days. Well, they did a trial run of it and um very few banks could survive
  187. it. So, they now have a less severe stress test. The banks are supposed to meet with 30 days on on their own
  188. liquidity. Um the other idea is to um limit the dependence of banks on wholesale funding. And so they have in
  189. mind a sort of long-term ratio that um they haven't really uh resolved yet. It's a hard thing to do. It may not be
  190. appropriate to do for every kind of financial institution. So it's not clear that's going to see the light of day.
  191. Well, thanks very much. We really appreciate you coming in today and uh look forward to following up if there's
  192. sounds like there may be a basil four in our future with all these things left undone. Well, it's not just that it's
  193. you have independent regulatory actions being undertaken by all the other major players, not least of which is the
  194. DoddFrank bill, which does not always correspond of what Basil is doing and um is turning out to be very very hard to
  195. implement.

Brief Summary

Wharton Professor Richard Harring discusses Basel 3 Accords and their impact on bank capital requirements and liquidity regulations.

Key points

  • Basel 3 Accords Discussion. Professor Richard Harring discusses the implications of the Basel 3 Accords on banking regulations.
  • Shift to G20 Governance. The shift from G10 to G20 reflects a new world governance system for finance.
  • Flaws in Basel 2. Exploring the serious mistakes made in Basel 2 regarding capital definitions.
  • Basel 3 Requirements. The Basel 3 requirements aim to hold banks accountable for their liquidity management.
  • Challenges of Implementation. Implementing Basel regulations is complicated by independent regulatory actions like the Dodd-Frank bill.

Tension

40/ 100Low tension

Related episodes