WEBVTT 00:00:00.000 --> 00:00:02.779 Welcome back to the Deep Dive. We are doing something 00:00:02.779 --> 00:00:04.919 a little different today. Usually we tackle a 00:00:04.919 --> 00:00:09.779 broad concept or a historical event, but today 00:00:09.779 --> 00:00:13.279 we are opening a specific file. It's a file that 00:00:13.279 --> 00:00:16.899 affects millions of homeowners. Yet, judging 00:00:16.899 --> 00:00:21.519 by this stack of research on my desk, maybe 1 00:00:21.519 --> 00:00:24.420 % of them actually understand the machinery inside 00:00:24.420 --> 00:00:26.219 it. Yeah, that sounds about right. We are talking 00:00:26.219 --> 00:00:28.000 about the adjustable rate mortgage or the... 00:00:28.140 --> 00:00:31.160 The ARM. The ARM. You know, it's funny. That 00:00:31.160 --> 00:00:33.859 acronym brings up such a visceral reaction depending 00:00:33.859 --> 00:00:35.880 on who you're talking to. Oh, absolutely. If 00:00:35.880 --> 00:00:38.539 you talk to a theoretical economist, they see 00:00:38.539 --> 00:00:41.079 a tool of, you know, incredible efficiency and 00:00:41.079 --> 00:00:43.439 flexibility. But if you talk to someone who lost 00:00:43.439 --> 00:00:46.259 their home in 2008 or someone who works in consumer 00:00:46.259 --> 00:00:48.520 advocacy. It's a different story. That acronym 00:00:48.520 --> 00:00:51.619 is basically. A weapon of mass destruction. That 00:00:51.619 --> 00:00:53.759 is the vibe I'm getting from these sources. It 00:00:53.759 --> 00:00:57.520 is so polarizing. And I think for most people, 00:00:57.619 --> 00:01:00.060 myself included, before I read all this, the 00:01:00.060 --> 00:01:02.380 knowledge just stops at the name. You see ARM. 00:01:02.520 --> 00:01:04.560 You think variable rate. You think, OK, my payment 00:01:04.560 --> 00:01:07.359 might change. And that's it. You look at the 00:01:07.359 --> 00:01:10.099 monthly payment, ask if you can afford it, and 00:01:10.099 --> 00:01:12.739 you sign the paper. And that's the danger. But 00:01:12.739 --> 00:01:15.319 looking through this research, the reality is 00:01:15.319 --> 00:01:17.909 so much more complicated. It's not just a loan, 00:01:18.049 --> 00:01:20.590 is it? It's a mechanism for transferring risk. 00:01:20.870 --> 00:01:23.370 That is the perfect way to frame this entire 00:01:23.370 --> 00:01:26.090 conversation. I mean, if you take nothing else 00:01:26.090 --> 00:01:28.250 away from this deep dive, I want you to understand 00:01:28.250 --> 00:01:32.409 that an ARM is a machine designed to shift the 00:01:32.409 --> 00:01:35.549 burden of uncertainty. A machine. Think about 00:01:35.549 --> 00:01:37.950 it this way. When you take out a mortgage, you 00:01:37.950 --> 00:01:40.430 are engaging in a massive financial bet on the 00:01:40.430 --> 00:01:43.030 future of the global economy. The question is, 00:01:43.069 --> 00:01:46.810 who is taking that bet? You. Or the bank. Risk 00:01:46.810 --> 00:01:48.969 transfer is such a polite phrase, though. It 00:01:48.969 --> 00:01:50.629 sounds like something an actuary does in a quiet 00:01:50.629 --> 00:01:53.670 office. But let's ground this. What does that 00:01:53.670 --> 00:01:55.930 actually mean for the person sitting at the kitchen 00:01:55.930 --> 00:01:58.750 table paying their bills? Okay, let's contrast 00:01:58.750 --> 00:02:02.049 it with the good old fixed rate mortgage. In 00:02:02.049 --> 00:02:04.530 a standard 30 -year fixed loan, you are essentially 00:02:04.530 --> 00:02:07.730 buying insurance. Insurance. You are paying a 00:02:07.730 --> 00:02:11.090 premium. It's baked in as a slightly higher interest 00:02:11.090 --> 00:02:14.289 rate. To make sure that no matter what happens 00:02:14.289 --> 00:02:16.469 in the world, no matter if inflation hits 10 00:02:16.469 --> 00:02:19.569 % or the Federal Reserve goes crazy. My bill 00:02:19.569 --> 00:02:22.409 is the same. Your bill is exactly $2 ,000 a month, 00:02:22.530 --> 00:02:25.469 period. The bank holds all that interest rate 00:02:25.469 --> 00:02:27.770 risk. If the cost of money goes up, the bank 00:02:27.770 --> 00:02:30.610 loses. And you, while you win. Right. I'm paying 00:02:30.610 --> 00:02:32.750 for peace of mind. I'm locking in my reality 00:02:32.750 --> 00:02:36.189 for the next 30 years. Exactly. But with an adjustable 00:02:36.189 --> 00:02:38.110 rate mortgage, the bank is coming to you and 00:02:38.110 --> 00:02:40.449 saying. We don't want to sell you that insurance 00:02:40.449 --> 00:02:43.050 anymore. We want you to be the insurance company. 00:02:43.250 --> 00:02:45.770 Wait, I'm the insurance company? Right. For the 00:02:45.770 --> 00:02:48.349 bank? You are insuring the bank's profit margin. 00:02:48.430 --> 00:02:51.110 If their costs go up, you pay them more. If their 00:02:51.110 --> 00:02:54.530 costs go down, you pay them less. You are tethering 00:02:54.530 --> 00:02:56.770 your family's budget directly to the volatility 00:02:56.770 --> 00:02:59.129 of Wall Street and the yield curve. Which, when 00:02:59.129 --> 00:03:01.090 you put it that way, sounds absolutely insane. 00:03:01.330 --> 00:03:04.469 Why would I ever agree to act as a shock absorber 00:03:04.469 --> 00:03:06.569 for a billion -dollar bank? I mean, they have 00:03:06.569 --> 00:03:09.199 the reserves, a checking account. The price tag. 00:03:09.400 --> 00:03:12.460 That's the hook. That is always the hook. Because 00:03:12.460 --> 00:03:14.979 you're agreeing to take on that, you know, that 00:03:14.979 --> 00:03:17.659 terrifying risk. The bank rewards you. They give 00:03:17.659 --> 00:03:19.979 you a lower starting interest rate compared to 00:03:19.979 --> 00:03:22.659 a fixed loan. The teaser. The teaser rate. Exactly. 00:03:22.900 --> 00:03:24.719 It's a tradeoff. You get affordability today 00:03:24.719 --> 00:03:27.719 in exchange for total uncertainty tomorrow. So 00:03:27.719 --> 00:03:30.120 it's a gamble. I'm betting rates stay low. Yeah. 00:03:30.199 --> 00:03:32.360 The bank is protecting itself in case they don't. 00:03:32.379 --> 00:03:34.599 It is a gamble. But here's where it gets really 00:03:34.599 --> 00:03:37.050 interesting and a little counterintuitive. We 00:03:37.050 --> 00:03:39.150 have some scholarly perspectives in the source 00:03:39.150 --> 00:03:41.969 stack that argue borrowers should generally prefer 00:03:41.969 --> 00:03:45.090 errands. Really? Even with the risk of my payment 00:03:45.090 --> 00:03:47.830 just exploding? The argument is based on long 00:03:47.830 --> 00:03:50.849 term averages. Historically, short -term interest 00:03:50.849 --> 00:03:53.370 rates are usually lower than long -term fixed 00:03:53.370 --> 00:03:55.750 rates. That's just how the yield curve normally 00:03:55.750 --> 00:03:59.069 works. So the theory goes, even if your rate 00:03:59.069 --> 00:04:01.870 fluctuates up and down, the average rate you 00:04:01.870 --> 00:04:04.629 pay over 10 or 15 years will likely be lower 00:04:04.629 --> 00:04:06.770 than that expensive insurance premium you pay 00:04:06.770 --> 00:04:09.490 for a 30 -year fixed rate. OK, I see the math 00:04:09.490 --> 00:04:12.509 there. I get it on a spreadsheet. But that assumes 00:04:12.509 --> 00:04:15.789 I can survive the bad years. That's the key. 00:04:16.300 --> 00:04:18.620 An average works great in a model, but in real 00:04:18.620 --> 00:04:21.839 life, one bad month where my payment doubles 00:04:21.839 --> 00:04:24.560 can ruin me. I can't eat my average savings if 00:04:24.560 --> 00:04:27.160 I've been foreclosed on. Precisely. And that 00:04:27.160 --> 00:04:29.779 is the huge disconnect between economic theory 00:04:29.779 --> 00:04:32.699 and household reality. The model assumes the 00:04:32.699 --> 00:04:35.019 borrower has financial resilience, which is a 00:04:35.019 --> 00:04:37.120 fancy way of saying you have enough cash sitting 00:04:37.120 --> 00:04:39.300 around to handle the shocks. But that's not why 00:04:39.300 --> 00:04:42.189 people take these loans. No. Most people taking 00:04:42.189 --> 00:04:44.329 ARMs are doing it because they are stretched 00:04:44.329 --> 00:04:46.569 thin and they need that lower initial payment 00:04:46.569 --> 00:04:48.589 just to get in the door of the house in the first 00:04:48.589 --> 00:04:50.810 place. So the people least able to handle the 00:04:50.810 --> 00:04:52.949 shock are the ones signing up for the shock absorber 00:04:52.949 --> 00:04:55.589 product. That is the irony at the heart of the 00:04:55.589 --> 00:04:58.089 ARM market. And that's why our mission today 00:04:58.089 --> 00:05:01.069 is to decode the jargon. We see these terms in 00:05:01.069 --> 00:05:04.290 the documents. Indices, margins, negative amortization, 00:05:04.629 --> 00:05:07.129 recast triggers. It's a whole other language. 00:05:07.560 --> 00:05:10.100 It is. And we need to unpack what those mean, 00:05:10.259 --> 00:05:12.639 because they aren't just fine print. They are 00:05:12.639 --> 00:05:16.120 the gears in that risk transfer machine. And 00:05:16.120 --> 00:05:18.560 as we will see later in this discussion, that 00:05:18.560 --> 00:05:22.199 machine has a history of breaking down. We're 00:05:22.199 --> 00:05:25.300 talking errors, scandals and billions of dollars 00:05:25.300 --> 00:05:28.240 in overcharges. That part of the research actually 00:05:28.240 --> 00:05:29.759 blew my mind. The error rates we're going to 00:05:29.759 --> 00:05:31.540 talk about later. I don't want to spoil it, but 00:05:31.540 --> 00:05:33.899 it's terrifying. It's stunning. But let's start 00:05:33.899 --> 00:05:37.300 at the foundation, the architecture. If I sign 00:05:37.300 --> 00:05:40.939 an ARM, how is my rate actually built? Because 00:05:40.939 --> 00:05:43.139 it's not just some random number the bank CEO 00:05:43.139 --> 00:05:44.980 picks out of a hat, right? No, no, not at all. 00:05:45.019 --> 00:05:47.920 It is a very strict mathematical formula. And 00:05:47.920 --> 00:05:49.560 it has two distinct components. You have the 00:05:49.560 --> 00:05:52.189 index and you have the margin. Index and margin. 00:05:52.329 --> 00:05:54.329 If you want a visual, think of the index as the 00:05:54.329 --> 00:05:56.389 moving part, like the tide in the ocean. It goes 00:05:56.389 --> 00:05:59.730 up and down, and the margin is a fixed pier built 00:05:59.730 --> 00:06:02.550 on top of that tide. Your rate is the top of 00:06:02.550 --> 00:06:04.870 that pier. Okay, let's unpack the index first. 00:06:04.990 --> 00:06:07.129 This is the variable part, the tide. Correct. 00:06:07.310 --> 00:06:09.850 The index is a published financial benchmark. 00:06:10.370 --> 00:06:13.110 It's meant to reflect the lender's cost of borrowing 00:06:13.110 --> 00:06:16.110 on the credit markets. Critically, it is objective, 00:06:16.410 --> 00:06:18.329 it is public, and it moves up and down with the 00:06:18.329 --> 00:06:20.810 economy. And the source lists a few of these, 00:06:20.930 --> 00:06:22.990 and some of them sound like alphabet soup. We 00:06:22.990 --> 00:06:25.069 have the CMT. The Constant Maturity Treasury. 00:06:25.370 --> 00:06:27.730 That's tied to U .S. Treasury securities. It's 00:06:27.730 --> 00:06:30.970 very common, very stable. It basically tracks 00:06:30.970 --> 00:06:32.949 what the U .S. government pays to borrow money. 00:06:33.149 --> 00:06:37.649 And there's one that caught my eye, COFI, the 00:06:37.649 --> 00:06:40.769 Cost of Funds Index. But it says specifically 00:06:40.769 --> 00:06:44.000 the 11th District. Why does the geography matter 00:06:44.000 --> 00:06:46.259 here? Ah, this is one of the most interesting 00:06:46.259 --> 00:06:49.279 nuances in the history of ARMS. The 11th District 00:06:49.279 --> 00:06:51.740 Cost of Funds Index tracks the weighted average 00:06:51.740 --> 00:06:54.220 cost of funds for savings institutions in Arizona, 00:06:54.480 --> 00:06:56.819 California, and Nevada. So it's a regional thing. 00:06:57.000 --> 00:06:59.899 But why would a borrower in, say, Florida, care 00:06:59.899 --> 00:07:02.259 if their loan is tied to Nevada banks versus 00:07:02.259 --> 00:07:04.360 the U .S. Treasury? What's the practical difference? 00:07:04.620 --> 00:07:06.459 The speed. It's all about the speed of change. 00:07:07.040 --> 00:07:09.519 KIOFI was historically known as a lagging index. 00:07:09.759 --> 00:07:12.470 It moved like a giant cruise ship. not a speedboat. 00:07:12.550 --> 00:07:14.810 A cruise ship. I like that. If the Fed raised 00:07:14.810 --> 00:07:17.009 rates today, other indices might jump tomorrow. 00:07:17.230 --> 00:07:20.389 But Kofi. It might take months to fully reflect 00:07:20.389 --> 00:07:22.149 that change because it's based on a weighted 00:07:22.149 --> 00:07:24.149 average of what banks are paying on old savings 00:07:24.149 --> 00:07:26.610 accounts and CDs over time. I see. So if I'm 00:07:26.610 --> 00:07:29.269 a borrower and I see a storm coming, you know, 00:07:29.269 --> 00:07:31.629 interest rates are hiking up, CoFi gives me a 00:07:31.629 --> 00:07:33.689 head start, a little breathing room before my 00:07:33.689 --> 00:07:35.829 payment goes up. Exactly. It gives you a buffer. 00:07:35.949 --> 00:07:38.709 It was a very borrower friendly index for that 00:07:38.709 --> 00:07:41.910 reason. Whereas LIBOR. LIBOR. The London Interbank 00:07:41.910 --> 00:07:44.410 offered rate. Now that sounds important. Is that 00:07:44.410 --> 00:07:46.990 the speedboat? LIBOR is the raw nerve of the 00:07:46.990 --> 00:07:49.610 global banking. system or it was it's largely 00:07:49.610 --> 00:07:52.889 been phased out now after some uh major scandals. 00:07:52.910 --> 00:07:54.750 But for decades, it was the king. Right. The 00:07:54.750 --> 00:07:57.370 rate fixing. The rate fixing. Yeah. But its function 00:07:57.370 --> 00:07:59.529 was to be the rate at which major global banks 00:07:59.529 --> 00:08:02.430 lend to one another overnight. It reacts instantly 00:08:02.430 --> 00:08:05.730 to global tension, credit crunches or policy 00:08:05.730 --> 00:08:08.709 changes. If you are tied to LIBOR, you are feeling 00:08:08.709 --> 00:08:11.569 the market's pulse in real time. No lag at all. 00:08:11.689 --> 00:08:14.829 And interestingly, the source mentions the CPI, 00:08:14.930 --> 00:08:18.250 the Consumer Price Index, can be used to tying 00:08:18.250 --> 00:08:20.829 my mortgage to the price of milk and gas. It 00:08:20.829 --> 00:08:23.680 can be, though it's less common. That ties your 00:08:23.680 --> 00:08:26.040 mortgage to inflation directly. But there is 00:08:26.040 --> 00:08:28.560 one other type of index mention that should raise 00:08:28.560 --> 00:08:31.899 a massive red flag for any listener. Some lenders 00:08:31.899 --> 00:08:34.799 use their own internal costs of funds as the 00:08:34.799 --> 00:08:37.559 index. That sounds incredibly suspicious. It's 00:08:37.559 --> 00:08:39.840 like I've investigated myself and found that 00:08:39.840 --> 00:08:41.740 my costs have gone up. So you need to pay me 00:08:41.740 --> 00:08:44.100 more. It certainly creates a massive conflict 00:08:44.100 --> 00:08:46.279 of interest. It allows for what we might call 00:08:46.279 --> 00:08:49.899 profit margin padding. If the index isn't a neutral 00:08:49.899 --> 00:08:52.799 third party number like treasuries. or LIBOR, 00:08:52.960 --> 00:08:55.820 the borrower has to trust the lender implicitly. 00:08:56.220 --> 00:08:59.019 And in finance, trust is good, but verification 00:08:59.019 --> 00:09:01.740 is always better. Which is a tall order for the 00:09:01.740 --> 00:09:04.100 average person. OK, so that's the index. It moves. 00:09:04.220 --> 00:09:06.259 That's the tide. Now talk to me about the margin. 00:09:06.360 --> 00:09:08.539 You said this is the fixed peer. Right. The margin 00:09:08.539 --> 00:09:11.019 is the lender's profit. It is a set percentage 00:09:11.019 --> 00:09:13.220 that gets added on top of the index. And this 00:09:13.220 --> 00:09:16.200 is key. It typically stays fixed for the entire 00:09:16.200 --> 00:09:18.799 30 -year life of the loan. So it never changes. 00:09:19.019 --> 00:09:23.220 Never. So if your margin is, say, 2 % and the 00:09:23.220 --> 00:09:26.840 index is at 4%, your rate is 6%. If the index 00:09:26.840 --> 00:09:30.179 jumps to 8%, your rate becomes 10%. The margin 00:09:30.179 --> 00:09:32.720 is the constant markup. And the source says this 00:09:32.720 --> 00:09:35.480 margin is usually between 2 % and 7%. That's 00:09:35.480 --> 00:09:37.320 a pretty wide range. It is a wide range, yes. 00:09:37.460 --> 00:09:40.000 It depends on the lender, the product, your credit 00:09:40.000 --> 00:09:43.220 score, all sorts of factors. And this leads us... 00:09:43.240 --> 00:09:46.659 to a really critical term, the fully indexed 00:09:46.659 --> 00:09:49.120 rate. Which is just the formula, index plus margin, 00:09:49.259 --> 00:09:51.860 the real price of the loan. Exactly. It's the 00:09:51.860 --> 00:09:54.679 true market rate for your loan on any given day. 00:09:55.100 --> 00:09:59.179 But here is the trap. Most ARMs don't start at 00:09:59.179 --> 00:10:02.179 the fully indexed rate. They start with a teaser. 00:10:02.500 --> 00:10:05.919 Ah, the teaser rate. The classic drug dealer 00:10:05.919 --> 00:10:07.659 model. The first one is cheap. Get them hooked. 00:10:08.029 --> 00:10:10.750 It is a promotional aid, absolutely. A lender 00:10:10.750 --> 00:10:13.049 might offer a start rate that is substantially 00:10:13.049 --> 00:10:16.049 below the fully indexed rate. Let's say the index 00:10:16.049 --> 00:10:18.590 plus margin should be 7%, but they offer you 00:10:18.590 --> 00:10:21.149 4 % for the first year. And the psychological 00:10:21.149 --> 00:10:24.029 trap here is just so obvious. You get used to 00:10:24.029 --> 00:10:27.789 paying that 4%. You budget for 4%. You buy a 00:10:27.789 --> 00:10:30.909 new car, assuming your mortgage is based on 4%. 00:10:30.909 --> 00:10:33.909 And then the adjustment period ends, the teaser 00:10:33.909 --> 00:10:37.490 expires, and suddenly you are hit with the reality 00:10:37.490 --> 00:10:39.679 of the market. This is what the industry calls 00:10:39.679 --> 00:10:41.919 payment shock. And it's a double whammy, right? 00:10:42.080 --> 00:10:45.059 It is. It isn't just that the underlying index 00:10:45.059 --> 00:10:47.320 might have gone up. It's that the artificial 00:10:47.320 --> 00:10:49.700 subsidy the bank was giving you was removed at 00:10:49.700 --> 00:10:52.320 the same time. So you get hit twice. So that 00:10:52.320 --> 00:10:54.759 adjustment timeline is crucial. How often does 00:10:54.759 --> 00:10:57.460 this actually happen? It varies a lot. You have 00:10:57.460 --> 00:10:59.899 the adjustment period, which dictates how often 00:10:59.899 --> 00:11:02.360 the rate resets. It could be every single month. 00:11:02.379 --> 00:11:04.340 It could be every year or every three years. 00:11:04.460 --> 00:11:07.269 Every month. That sounds exhausting. I'd be checking 00:11:07.269 --> 00:11:09.250 the financial news every morning in a cold sweat. 00:11:09.429 --> 00:11:12.309 It creates immense uncertainty. But there are 00:11:12.309 --> 00:11:13.929 guardrails, and this is where we need to get 00:11:13.929 --> 00:11:16.450 into the math, because understanding these guardrails 00:11:16.450 --> 00:11:18.509 or caps is the difference between an annoying 00:11:18.509 --> 00:11:21.570 increase and, well, bankruptcy. Okay, let's talk 00:11:21.570 --> 00:11:24.330 caps. The source breaks these down into a structure, 00:11:24.470 --> 00:11:27.889 usually three numbers, like X, Y, Z. I want to 00:11:27.889 --> 00:11:29.610 pause on the math here because I think this is 00:11:29.610 --> 00:11:31.870 where people get really hurt. It is. We see these 00:11:31.870 --> 00:11:35.059 numbers in our mortgage documents. 2 slash 2 00:11:35.059 --> 00:11:37.899 slash 6. It looks like a locker combination. 00:11:38.820 --> 00:11:41.919 So let's role play this. Let's say I borrow $500 00:11:41.919 --> 00:11:46.600 ,000 on an ARM. My start rate, the teaser, is 00:11:46.600 --> 00:11:50.580 a nice low 4%. Okay, so at 4 % on a 30 -year 00:11:50.580 --> 00:11:52.940 term, your principal and interest payment is 00:11:52.940 --> 00:11:57.740 roughly $2 ,387, manageable for a decent income. 00:11:57.980 --> 00:12:00.019 Right, I can do that. Now, the first year ends. 00:12:00.460 --> 00:12:02.659 The teaser is gone. And let's say it's a bad 00:12:02.659 --> 00:12:04.740 year. Inflation is rampant. The index has jumped 00:12:04.740 --> 00:12:07.779 up by 3%. My margin is 2%. So the formula says 00:12:07.779 --> 00:12:10.879 my rate should now be my index plus margin, which 00:12:10.879 --> 00:12:14.059 is 9%. That's a huge jump. It is. A jump from 00:12:14.059 --> 00:12:16.600 4 % to 9 % would be devastating. But your initial 00:12:16.600 --> 00:12:18.519 adjustment cap, that's the first number in that 00:12:18.519 --> 00:12:21.659 226 sequence it steps in. That first two points 00:12:21.659 --> 00:12:23.799 means your rate is physically prevented from 00:12:23.799 --> 00:12:27.289 jumping all the way to 9%. It can only increase 00:12:27.289 --> 00:12:30.190 by a maximum of two percentage points on that 00:12:30.190 --> 00:12:33.669 very first adjustment. So 4 % plus 2 % equals 00:12:33.669 --> 00:12:36.730 6%. Okay, so I'm saved. My rate only goes to 00:12:36.730 --> 00:12:40.809 6%, not 9%. For this year, yes. Your payment 00:12:40.809 --> 00:12:44.769 now jumps from roughly $2 ,400 to about $3 ,000. 00:12:45.009 --> 00:12:49.110 That's a $600 a month increase. That hurts. I 00:12:49.110 --> 00:12:51.090 might have to cancel a vacation, maybe eat out 00:12:51.090 --> 00:12:53.870 less, but I'm not homeless. But here is the danger. 00:12:54.350 --> 00:12:57.509 The market rate is still 9%. You are effectively 00:12:57.509 --> 00:13:00.669 being subsidized by the cap for this year. So 00:13:00.669 --> 00:13:03.769 if rates stay high next year, the periodic cap, 00:13:03.870 --> 00:13:06.350 that second number, the second two, allows another 00:13:06.350 --> 00:13:09.210 2 % jump. So I go from 6 % to 8%. Exactly. And 00:13:09.210 --> 00:13:11.129 your payment goes up again and then again the 00:13:11.129 --> 00:13:13.549 year after that until you finally hit the market 00:13:13.549 --> 00:13:15.710 rate or you hit the lifetime cap, the third number. 00:13:15.809 --> 00:13:19.009 The 16 -hour example. That 6 -4 means your rate 00:13:19.009 --> 00:13:21.629 can never go more than 6 percentage points above 00:13:21.629 --> 00:13:24.879 your original start rate. So your 4 % start rate 00:13:24.879 --> 00:13:29.000 plus the 6 % cap equals a 10 % maximum. That's 00:13:29.000 --> 00:13:31.919 your doomsday number. So at 10%, my payment would 00:13:31.919 --> 00:13:34.940 be nearly $4 ,400. That's almost double where 00:13:34.940 --> 00:13:38.019 I started. Exactly. Even with caps, which are 00:13:38.019 --> 00:13:40.580 supposed to protect you, your payment can double. 00:13:40.679 --> 00:13:43.039 That is the risk you accepted for that lower 00:13:43.039 --> 00:13:45.240 initial rate. That is terrifying, but at least 00:13:45.240 --> 00:13:47.320 I can calculate the absolute worst case scenario. 00:13:47.600 --> 00:13:50.879 Yeah. But, and here's where it gets really murky. 00:13:51.500 --> 00:13:53.220 There is a distinction mentioned in the source 00:13:53.220 --> 00:13:56.720 between a rate cap and a payment cap. And confusing 00:13:56.720 --> 00:13:59.019 the two seems like it could be disastrous. This 00:13:59.019 --> 00:14:01.399 is one of the most dangerous misunderstandings 00:14:01.399 --> 00:14:04.299 in the world of mortgages. A rate cap limits 00:14:04.299 --> 00:14:06.259 the interest percentage, like we just discussed. 00:14:06.460 --> 00:14:09.440 A payment cap limits the dollar amount your check 00:14:09.440 --> 00:14:11.539 increases. Okay, so give me an example. Let's 00:14:11.539 --> 00:14:14.820 say you have a 7 .5 % payment cap. Your payment 00:14:14.820 --> 00:14:18.000 is $1 ,000 a month. This tap means it can only 00:14:18.000 --> 00:14:21.559 go up to $1 ,075 next year, no matter what happens 00:14:21.559 --> 00:14:23.899 to interest rates. That sounds great. It protects 00:14:23.899 --> 00:14:25.799 my wallet. I know exactly what my budget needs 00:14:25.799 --> 00:14:28.360 to be. It feels safe. It protects your cash flow, 00:14:28.460 --> 00:14:31.159 yes. But it does not protect you from the interest. 00:14:31.980 --> 00:14:34.580 If the underlying interest rate spiked, and the 00:14:34.580 --> 00:14:36.659 interest the bank is actually owed for that month 00:14:36.659 --> 00:14:41.220 is, say, $1 ,500, but you only paid $1 ,075. 00:14:43.029 --> 00:14:45.909 What happens to that difference? I assume I still 00:14:45.909 --> 00:14:48.110 owe it. The bank's not just going to forget about 00:14:48.110 --> 00:14:52.730 it. You do. You are short $425. The bank doesn't 00:14:52.730 --> 00:14:55.850 just forgive that. They take that $425 of unpaid 00:14:55.850 --> 00:14:57.909 interest and they add it to your principal loan 00:14:57.909 --> 00:15:00.970 balance. So my mortgage gets bigger. I paid my 00:15:00.970 --> 00:15:03.870 bill on time, but my debt went up. Correct. This 00:15:03.870 --> 00:15:06.669 is called negative amortization. You are making 00:15:06.669 --> 00:15:09.559 payments. But you are going deeper into debt 00:15:09.559 --> 00:15:11.860 every single month. It's like running on a treadmill 00:15:11.860 --> 00:15:14.419 that's slowly tilting backwards. That is a nightmare 00:15:14.419 --> 00:15:16.759 scenario. You could pay for five years and owe 00:15:16.759 --> 00:15:18.419 more than you bought the house for in the first 00:15:18.419 --> 00:15:21.200 place. And that leads to massive equity erosion. 00:15:21.899 --> 00:15:24.879 If housing prices drop at the same time, which. 00:15:25.149 --> 00:15:27.330 By the way, often happens when interest rates 00:15:27.330 --> 00:15:29.549 are rising. You can find yourself underwater 00:15:29.549 --> 00:15:32.690 almost instantly. You can't sell. You can't refinance. 00:15:32.690 --> 00:15:34.629 You are completely trapped. We're going to get 00:15:34.629 --> 00:15:36.710 deeper into that, specifically with option ARMs 00:15:36.710 --> 00:15:38.509 in a minute. That feels like the true crime section 00:15:38.509 --> 00:15:40.830 of this deep dive. Yeah. But I want to look at 00:15:40.830 --> 00:15:43.950 the menu available to borrowers first. Because 00:15:43.950 --> 00:15:46.649 ARMs aren't just one thing. There are different 00:15:46.649 --> 00:15:49.399 flavors. Right. And the most popular flavor, 00:15:49.559 --> 00:15:51.659 especially in the last two decades, is the hybrid 00:15:51.659 --> 00:15:54.480 ARM. This is pitched as the best of both worlds, 00:15:54.600 --> 00:15:56.960 right? That's the pitch, in theory. A hybrid 00:15:56.960 --> 00:15:59.500 ARM is fixed for a set period, and then it starts 00:15:59.500 --> 00:16:02.539 to float. You see them listed as 31, 51, 71. 00:16:02.980 --> 00:16:05.100 The first number is the number of years the rate 00:16:05.100 --> 00:16:08.039 is fixed. A 51 is fixed for five years. The second 00:16:08.039 --> 00:16:10.679 number is how often it adjusts after that period 00:16:10.679 --> 00:16:14.220 ends. So a 51 is fixed for five years, then adjusts 00:16:14.220 --> 00:16:16.360 every one year thereafter. I was reading in the 00:16:16.360 --> 00:16:18.419 source that the popularity... these just exploded. 00:16:18.779 --> 00:16:21.779 In 1998, they were less than 2 % of all loans. 00:16:22.000 --> 00:16:26.440 By 2004, it was almost 30%. Why the sudden surge? 00:16:26.740 --> 00:16:28.860 It was a sweet spot for the psychology of the 00:16:28.860 --> 00:16:31.000 American homebuyer at the time. People looked 00:16:31.000 --> 00:16:32.740 at the data and said, I'm probably going to move 00:16:32.740 --> 00:16:35.080 in five or seven years anyway. The average American 00:16:35.080 --> 00:16:37.860 moves every seven years. Why am I paying the 00:16:37.860 --> 00:16:40.620 premium for a 30 -year fixed rate if I'm only 00:16:40.620 --> 00:16:42.840 going to be in this house for five? So they're 00:16:42.840 --> 00:16:45.659 gaming the system. They are. The hybrid ARM gave 00:16:45.659 --> 00:16:48.559 them a lower rate than the 30 -year fixed, guaranteeing 00:16:48.559 --> 00:16:51.120 that low payment for the entire time they plan 00:16:51.120 --> 00:16:53.320 to own the home. It seemed like a no -brainer. 00:16:53.419 --> 00:16:55.519 It's a bet on your own life plans. You're betting 00:16:55.519 --> 00:16:58.960 you'll sell the house or refinance before that 00:16:58.960 --> 00:17:01.600 crucial date, the reset date. The reset date 00:17:01.600 --> 00:17:04.240 is the cliff edge. That's the day the clock strikes 00:17:04.240 --> 00:17:06.900 midnight. If you are still in that house on that 00:17:06.900 --> 00:17:10.759 date, the loan transforms from a safe predictable, 00:17:10.880 --> 00:17:14.380 fixed animal into a volatile, variable beast. 00:17:14.740 --> 00:17:17.480 A lot of people lost that bet in 2008 and 2009. 00:17:17.779 --> 00:17:20.079 A lot of people. Now let's talk about the product 00:17:20.079 --> 00:17:22.660 that really defined the bubble era for so many 00:17:22.660 --> 00:17:26.720 people. The option ARM, also known as the pick 00:17:26.720 --> 00:17:29.900 a payment loan. This feels less like a loan and 00:17:29.900 --> 00:17:32.170 more like a carefully constructed trap. This 00:17:32.170 --> 00:17:35.549 is a complex financial instrument that was, believe 00:17:35.549 --> 00:17:37.849 it or not, originally designed for very sophisticated 00:17:37.849 --> 00:17:41.029 borrowers, people with extremely lumpy incomes 00:17:41.029 --> 00:17:43.849 like commission -based salespeople or seasonal 00:17:43.849 --> 00:17:47.210 workers. Imagine a wealthy entrepreneur who makes 00:17:47.210 --> 00:17:50.650 $500 ,000 in December, but basically zero in 00:17:50.650 --> 00:17:53.990 January. The idea was flexibility. But it went 00:17:53.990 --> 00:17:56.549 mainstream in a big way. It did. And here is 00:17:56.549 --> 00:17:59.089 how it worked. Every month, you get a bill with 00:17:59.089 --> 00:18:01.250 four choices. Walk us through the four options 00:18:01.250 --> 00:18:05.150 on the menu. Option one, a 30 -year fully amortizing 00:18:05.150 --> 00:18:07.210 payment. This is like a normal mortgage payment 00:18:07.210 --> 00:18:09.769 paying both principal and interest. Option two, 00:18:09.930 --> 00:18:12.349 a 15 -year amortizing payment if you want to 00:18:12.349 --> 00:18:14.690 pay it off faster. Option three, interest only. 00:18:14.789 --> 00:18:16.589 You're just paying the rent on the money you 00:18:16.589 --> 00:18:18.490 borrowed, but you're not paying down the actual 00:18:18.490 --> 00:18:21.089 debt. Okay, and then the fourth one, the dangerous 00:18:21.089 --> 00:18:24.450 one. And option four, the specified minimum payment. 00:18:24.630 --> 00:18:27.519 And human nature being what it is. Almost everyone 00:18:27.519 --> 00:18:31.740 chose number four. The minimum payment. But that 00:18:31.740 --> 00:18:35.420 minimum payment was based on that super low teaser 00:18:35.420 --> 00:18:39.119 rate, wasn't it? Yes. Often as low as 1%. So 00:18:39.119 --> 00:18:41.940 imagine opening your bill. You owe $500 ,000 00:18:41.940 --> 00:18:45.940 on your house. The full payment is $3 ,000. The 00:18:45.940 --> 00:18:49.440 minimum payment is $1 ,200. And you've got a 00:18:49.440 --> 00:18:51.819 car payment. Your kids are in braces. Groceries 00:18:51.819 --> 00:18:54.220 are expensive. I'm clicking the $1 ,200 button. 00:18:54.700 --> 00:18:57.359 Every single time, no question. Exactly. But 00:18:57.359 --> 00:18:59.980 the actual interest rate on the loan isn't 1%. 00:18:59.980 --> 00:19:03.660 It might be 6 % or 7%. So that difference, that 00:19:03.660 --> 00:19:07.220 $1 ,800 in interest that you didn't pay, it doesn't 00:19:07.220 --> 00:19:09.700 just vanish. The bank adds it to the loan. They 00:19:09.700 --> 00:19:12.079 add it right onto the principal. So next month, 00:19:12.119 --> 00:19:15.539 your mortgage isn't $500 ,000 anymore. It's $501 00:19:15.539 --> 00:19:19.619 ,800. And the month after that, it's $503 ,600. 00:19:19.740 --> 00:19:23.059 You are literally borrowing more money from the 00:19:23.059 --> 00:19:25.569 bank each month. to pay the interest on the money 00:19:25.569 --> 00:19:27.430 you already borrowed. And eventually that bucket 00:19:27.430 --> 00:19:30.769 has to overflow. The math is relentless. This 00:19:30.769 --> 00:19:33.730 is the recast. This is the moment the music stops. 00:19:34.049 --> 00:19:37.089 This is the moment. The contract says, OK, this 00:19:37.089 --> 00:19:39.329 has gone on long enough. And there are two triggers 00:19:39.329 --> 00:19:41.529 for it. One is a time trigger. Usually every 00:19:41.529 --> 00:19:44.470 five years, the loan will automatically recast. 00:19:45.150 --> 00:19:48.049 But the scary one is the balance trigger. The 00:19:48.049 --> 00:19:50.369 MIG -AM cap. The negative amortization cap. The 00:19:50.369 --> 00:19:52.789 contract says if your loan balance grows to, 00:19:52.910 --> 00:19:56.529 say, 110 or 125 percent of the original loan 00:19:56.529 --> 00:19:59.009 amount, the loan automatically recasts right 00:19:59.009 --> 00:20:01.250 then and there. And what does recast mean in 00:20:01.250 --> 00:20:03.640 plain English for the borrower? It means the 00:20:03.640 --> 00:20:07.319 party is over. The bank completely recalculates 00:20:07.319 --> 00:20:09.539 your payment to be whatever it needs to be to 00:20:09.539 --> 00:20:12.559 pay off the entire new larger balance over the 00:20:12.559 --> 00:20:15.000 remaining term at the full interest rate. No 00:20:15.000 --> 00:20:17.359 more minimums, no more choices. That must be 00:20:17.359 --> 00:20:20.299 a massive, massive jump. It's colossal. We're 00:20:20.299 --> 00:20:22.299 talking about people going from paying that comfortable 00:20:22.299 --> 00:20:25.940 $1 ,200 a month to $3 ,500 a month literally 00:20:25.940 --> 00:20:28.839 overnight. That is the ultimate payment shock. 00:20:29.079 --> 00:20:30.819 And because you've eaten away all your equity. 00:20:31.279 --> 00:20:33.940 Or worse, your loan is now bigger than the house 00:20:33.940 --> 00:20:36.539 is even worth. You can't refinance. You're completely 00:20:36.539 --> 00:20:39.779 trapped. This one mechanism is what caused hundreds 00:20:39.779 --> 00:20:41.980 of thousands of foreclosures in the financial 00:20:41.980 --> 00:20:44.700 crisis. It's incredible that this was ever a 00:20:44.700 --> 00:20:46.779 mainstream product. It seems almost designed 00:20:46.779 --> 00:20:50.180 to fail. It was sold on the dream of affordability. 00:20:50.559 --> 00:20:52.980 Get into the house of your dreams for only $1 00:20:52.980 --> 00:20:55.720 ,200 a month. But it was structured on a foundation 00:20:55.720 --> 00:20:58.490 of pure risk. There's also something mentioned 00:20:58.490 --> 00:21:01.150 called a cash flow ARM. Is that just another 00:21:01.150 --> 00:21:03.230 name for the same thing? It's very similar, yes. 00:21:03.410 --> 00:21:05.609 It focuses on that minimum payment flexibility 00:21:05.609 --> 00:21:08.230 for the first few years. The source notes that 00:21:08.230 --> 00:21:11.069 some fixed rate loans even had these cash flow 00:21:11.069 --> 00:21:13.529 options built in. But generally, when we hear 00:21:13.529 --> 00:21:15.750 cash flow ARM, we are talking about that same 00:21:15.750 --> 00:21:18.869 dangerous mechanism, prioritizing low monthly 00:21:18.869 --> 00:21:21.390 outflow at the cost of a rapidly rising debt 00:21:21.390 --> 00:21:24.369 balance. Okay, so we see why a borrower might 00:21:24.369 --> 00:21:26.849 fall for this. Lower payments, qualifying for 00:21:26.849 --> 00:21:28.670 a bigger house. It's the American dream on easy 00:21:28.670 --> 00:21:32.369 mode. But why do banks push these? I mean, surely 00:21:32.369 --> 00:21:34.509 they know that a huge percentage of people will 00:21:34.509 --> 00:21:37.369 default. You mentioned earlier the asset liability 00:21:37.369 --> 00:21:40.089 mismatch. Can we dig into the economics from 00:21:40.089 --> 00:21:42.730 the bank's perspective here? Why do they need 00:21:42.730 --> 00:21:46.250 ARMs to exist? This is fundamental to understanding 00:21:46.250 --> 00:21:48.869 banking, and it explains why the bank isn't necessarily 00:21:48.869 --> 00:21:52.309 being evil in creating these. They are, from 00:21:52.309 --> 00:21:54.950 their perspective, just trying to survive. Banks 00:21:54.950 --> 00:21:57.670 are essentially middlemen. They borrow money 00:21:57.670 --> 00:21:59.690 from depositors. That's you and me putting money 00:21:59.690 --> 00:22:02.049 in a checking or savings account. And they lend 00:22:02.049 --> 00:22:04.089 that money out long term to homebuyers. Right. 00:22:04.130 --> 00:22:07.970 Simple enough. But deposits are short term liabilities. 00:22:08.190 --> 00:22:10.210 You can walk into the bank and take your money 00:22:10.210 --> 00:22:12.970 out today. The bank has to pay you a short -term 00:22:12.970 --> 00:22:15.690 interest rate on that savings account. A mortgage, 00:22:15.809 --> 00:22:18.329 however, is a long -term asset. It sits on their 00:22:18.329 --> 00:22:21.049 books for 30 years. So the bank is borrowing 00:22:21.049 --> 00:22:24.529 short and lending long. Exactly. And that creates 00:22:24.529 --> 00:22:27.630 a dangerous mismatch. Imagine a bank in 1975 00:22:27.630 --> 00:22:30.710 lends out billions in mortgages at a fixed 4 00:22:30.710 --> 00:22:34.529 % for 30 years. But then, in the early 80s, inflation 00:22:34.529 --> 00:22:37.630 goes crazy. To keep depositors from pulling all 00:22:37.630 --> 00:22:39.490 their money out and moving it to a competitor 00:22:39.490 --> 00:22:41.890 offering higher rates, the bank has to start 00:22:41.890 --> 00:22:44.349 paying them 6, 7, 8 % interest on their savings 00:22:44.349 --> 00:22:47.029 accounts. So the bank is paying out 8 % to get 00:22:47.029 --> 00:22:49.730 the money. But they're only collecting 4 % from 00:22:49.730 --> 00:22:52.410 all those old mortgages. They are bleeding cash 00:22:52.410 --> 00:22:55.569 on every single loan. This is exactly what happened 00:22:55.569 --> 00:22:57.869 during the U .S. savings and loan crisis. Paul 00:22:57.869 --> 00:23:00.650 Volcker, who is the Fed chairman, jacked up interest 00:23:00.650 --> 00:23:03.849 rates to fight inflation. The cost of funds for 00:23:03.849 --> 00:23:07.049 banks skyrocketed, but all their income was locked 00:23:07.049 --> 00:23:09.829 in those low fixed rate mortgages from the 70s. 00:23:09.849 --> 00:23:12.329 And the banks just collapsed. Thousands of them 00:23:12.329 --> 00:23:14.849 failed. It was a complete catastrophe. So the 00:23:14.849 --> 00:23:17.490 ARM is the bank's immune system against that 00:23:17.490 --> 00:23:20.470 ever happening again. Ah, I get it now. It's 00:23:20.470 --> 00:23:23.500 a survival mechanism. By using ARMs, the bank 00:23:23.500 --> 00:23:27.079 matches its income to its expenses. Yes. If rates 00:23:27.079 --> 00:23:29.279 go up, they have to pay depositors more. True. 00:23:29.420 --> 00:23:31.319 But they also start collecting more from the 00:23:31.319 --> 00:23:33.950 ARM borrowers almost immediately. It stabilizes 00:23:33.950 --> 00:23:36.589 their profit margin. It saves the bank by passing 00:23:36.589 --> 00:23:39.250 that volatility directly to the homeowner. That 00:23:39.250 --> 00:23:41.690 really clarifies the why. It's not just predatory 00:23:41.690 --> 00:23:44.049 greed, although that's part of it. It's structural 00:23:44.049 --> 00:23:46.990 survival for the banking industry. They are terrified 00:23:46.990 --> 00:23:49.069 of being caught holding the bag again like they 00:23:49.069 --> 00:23:51.589 were in the 80s. Precisely. They are transferring 00:23:51.589 --> 00:23:53.849 what we could call the Volcker risk directly 00:23:53.849 --> 00:23:56.509 onto your shoulders. Before we get to the scandals, 00:23:56.529 --> 00:23:58.650 and I really want to get to the scandals, I want 00:23:58.650 --> 00:24:00.990 to zoom out for a second because reading this 00:24:00.990 --> 00:24:04.019 source. I realize that my view of mortgages is 00:24:04.019 --> 00:24:06.799 very, very American -centric. I just assumed 00:24:06.799 --> 00:24:10.619 everyone did the 30 -year facing. But that's 00:24:10.619 --> 00:24:13.079 not the case globally at all, is it? Not even 00:24:13.079 --> 00:24:15.279 close. In fact, the United States is the global 00:24:15.279 --> 00:24:18.609 outlier. We are the weird ones. Here, the adjustable 00:24:18.609 --> 00:24:20.849 rate mortgage is almost a dirty word. It implies 00:24:20.849 --> 00:24:22.690 a government regulated product that you only 00:24:22.690 --> 00:24:25.369 take if you have to. But you go to the UK. Variable 00:24:25.369 --> 00:24:28.109 rates are the norm. The norm. So everyone just 00:24:28.109 --> 00:24:30.130 accepts that their payment might change month 00:24:30.130 --> 00:24:33.410 to month. Yes, absolutely. In the UK, borrowers 00:24:33.410 --> 00:24:35.670 are very focused on the current monthly cost. 00:24:35.849 --> 00:24:38.450 They ride the wave. If the Bank of England raises 00:24:38.450 --> 00:24:40.950 rates, everyone in the country feels it in their 00:24:40.950 --> 00:24:43.130 mortgage payment almost instantly. And they tighten 00:24:43.130 --> 00:24:44.990 their belts. It actually makes monetary policy 00:24:44.990 --> 00:24:50.079 very effective. Wow. The same is true in Australia 00:24:50.079 --> 00:24:52.480 and New Zealand. Variable rates are the standard 00:24:52.480 --> 00:24:55.079 way to buy a home. What about Canada? They're 00:24:55.079 --> 00:24:56.819 our neighbors. Surely they're similar to us. 00:24:56.960 --> 00:24:59.900 They're sort of a middle ground. In Canada, you 00:24:59.900 --> 00:25:02.180 might get a fixed rate, but it's usually only 00:25:02.180 --> 00:25:04.500 fixed for five years, maybe 10 at the absolute 00:25:04.500 --> 00:25:07.700 maximum. The amortization of the loan is 25 years, 00:25:07.880 --> 00:25:10.319 but the rate isn't locked for that whole time. 00:25:10.420 --> 00:25:11.960 So they have to renew it. You have to renew the 00:25:11.960 --> 00:25:14.200 loan at the new market rate every five years. 00:25:14.319 --> 00:25:16.640 So they are effectively rolling over hybrid loans 00:25:16.640 --> 00:25:18.980 constantly. They don't have that 30 year security 00:25:18.980 --> 00:25:21.859 blanket that we cherish so much here. And the 00:25:21.859 --> 00:25:24.960 source mentions Germany and Austria have a completely 00:25:24.960 --> 00:25:28.319 different model. The Bausparkissen. What is that? 00:25:28.670 --> 00:25:31.130 This is a fascinating cultural difference regarding 00:25:31.130 --> 00:25:34.170 debt and saving. The Bausparkassen are building 00:25:34.170 --> 00:25:37.029 societies. The model there is you have to save 00:25:37.029 --> 00:25:40.069 first. You agree to put money into a special 00:25:40.069 --> 00:25:42.470 savings account at a low fixed interest rate 00:25:42.470 --> 00:25:45.190 for years. So you are lending to the bank first 00:25:45.190 --> 00:25:47.410 at a low rate. Essentially, yes. You're building 00:25:47.410 --> 00:25:49.690 up capital in their system. And once you have 00:25:49.690 --> 00:25:52.670 met your savings quota, you are guaranteed a 00:25:52.670 --> 00:25:54.890 low fixed rate loan for the rest of your life. 00:25:55.049 --> 00:25:57.480 It's a closed system. What's the advantage? The 00:25:57.480 --> 00:26:00.039 advantage is absolute certainty. Your payment 00:26:00.039 --> 00:26:03.279 never changes. The rate is locked forever. The 00:26:03.279 --> 00:26:05.480 disadvantage is you have to plan years and years 00:26:05.480 --> 00:26:08.880 in advance. It's aimed at the once -in -a -lifetime 00:26:08.880 --> 00:26:11.859 homebuyer who plans to stay put. It doesn't sound 00:26:11.859 --> 00:26:13.720 very flexible for the modern worker who might 00:26:13.720 --> 00:26:16.079 move every four years for a new job. Exactly. 00:26:16.359 --> 00:26:18.440 It struggles with modern mobility. It's a very 00:26:18.440 --> 00:26:21.700 different mindset. Let's look at Singapore. The 00:26:21.700 --> 00:26:23.599 source goes into some specific detail there. 00:26:23.680 --> 00:26:26.890 They call them floating rate packages. Yes. And 00:26:26.890 --> 00:26:28.990 they use very transparent indices usually. The 00:26:28.990 --> 00:26:31.569 CYBOR, that's the Singapore Interbank Offered 00:26:31.569 --> 00:26:34.769 Rate, and the SOR, the Swap Offer Rate. Very 00:26:34.769 --> 00:26:37.109 clear benchmarks. And the structure there seems 00:26:37.109 --> 00:26:39.819 to be a progressive spread or margin. Right. 00:26:39.940 --> 00:26:41.660 This is interesting. They often structure the 00:26:41.660 --> 00:26:44.240 margin to increase over time. So year one might 00:26:44.240 --> 00:26:48.019 be cyber plus 0 .75 percent. Year two, the same. 00:26:48.140 --> 00:26:50.400 But then in year four and onwards, it jumps to 00:26:50.400 --> 00:26:53.859 cyber plus 1 .25 percent. That seems almost like 00:26:53.859 --> 00:26:56.039 a penalty for keeping the loan. It pushes you 00:26:56.039 --> 00:26:58.700 to pay it off. It incentivizes you to pay it 00:26:58.700 --> 00:27:02.240 down or refinance. It's a subtle nudge from the 00:27:02.240 --> 00:27:04.420 bank saying, we don't want this small margin 00:27:04.420 --> 00:27:07.359 loan on our books forever. And there was a unique 00:27:07.359 --> 00:27:10.720 innovation mentioned there. ANZ Bank introduced 00:27:10.720 --> 00:27:14.039 a loan pegged to the average of Sybor and Savar. 00:27:14.180 --> 00:27:16.720 Why average them? What does that do? That's an 00:27:16.720 --> 00:27:19.720 attempt to smooth out volatility. By averaging 00:27:19.720 --> 00:27:21.940 two different indices, you dampen the spikes. 00:27:22.359 --> 00:27:25.279 If one jumps and the other stays flat, the borrower 00:27:25.279 --> 00:27:27.720 doesn't feel the full hit. It's a clever bit 00:27:27.720 --> 00:27:29.559 of financial engineering to make the product 00:27:29.559 --> 00:27:32.039 a little safer. It really shows that the way 00:27:32.039 --> 00:27:34.400 we pay for our homes is not a law of nature. 00:27:34.640 --> 00:27:37.220 It's a product of local banking laws and national 00:27:37.220 --> 00:27:40.039 culture. We think the 30 -year fixed is the only 00:27:40.039 --> 00:27:42.200 way, but most of the rest of the world thinks 00:27:42.200 --> 00:27:44.019 we're crazy for even offering it. Absolutely. 00:27:44.079 --> 00:27:46.200 The 30 -year fixed is essentially a government 00:27:46.200 --> 00:27:48.680 -subsidized anomaly kept alive by Fannie Mae 00:27:48.680 --> 00:27:51.420 and Freddie Mac. Okay. We have delayed this long 00:27:51.420 --> 00:27:53.079 enough. We have to talk about the dark side. 00:27:53.450 --> 00:27:55.349 Because when you build a machine this complex 00:27:55.349 --> 00:27:57.890 with indices, margins, look back periods, caps, 00:27:58.069 --> 00:28:00.490 floors, triggers, things can break. And they 00:28:00.490 --> 00:28:03.470 break. They break often. The source material 00:28:03.470 --> 00:28:06.390 here contains some truly shocking statistics 00:28:06.390 --> 00:28:10.309 about just simple calculation errors. This is 00:28:10.309 --> 00:28:12.630 the back office disaster. It is. I mean, think 00:28:12.630 --> 00:28:14.630 about it. In the pre -digital age or even in 00:28:14.630 --> 00:28:17.410 the early digital age, a loan servicer had to 00:28:17.410 --> 00:28:19.789 manually track the index date. They'd have a 00:28:19.789 --> 00:28:22.029 note saying, OK, this loan adjusts on November 00:28:22.029 --> 00:28:25.019 1st. Looking back 45 days. That sounds incredibly 00:28:25.019 --> 00:28:27.660 specific and easy to mess up. It is. They have 00:28:27.660 --> 00:28:30.680 to find the LIBOR rate on exactly September 15th, 00:28:30.680 --> 00:28:32.759 not the 14th, not the 16th. Then they have to 00:28:32.759 --> 00:28:35.059 apply the correct margin, then check the initial 00:28:35.059 --> 00:28:37.440 cap, then check the periodic cap, check the lifetime 00:28:37.440 --> 00:28:39.599 cap, check for a payment cap. And they are doing 00:28:39.599 --> 00:28:41.859 this for thousands and thousands of loans, all 00:28:41.859 --> 00:28:44.559 with different terms. Millions of loans. And 00:28:44.559 --> 00:28:47.400 people make mistakes. The source cites a 1991 00:28:47.400 --> 00:28:50.160 GAO study. That's the Government Accountability 00:28:50.160 --> 00:28:53.119 Office. They audited a sample of eight arms and 00:28:53.119 --> 00:28:55.940 found that 20 to 25 percent of them had interest 00:28:55.940 --> 00:28:58.279 rates. errors. One in four. One in four was calculated 00:28:58.279 --> 00:29:00.680 incorrectly. And the financial impact was $10 00:29:00.680 --> 00:29:03.759 billion in net overcharges to homeowners. That 00:29:03.759 --> 00:29:06.740 is massive. That's $10 billion basically stolen 00:29:06.740 --> 00:29:09.559 from homeowners, mostly by accident. Mostly by 00:29:09.559 --> 00:29:12.019 sheer incompetence and complexity. But it gets 00:29:12.019 --> 00:29:15.240 worse. A 1995 government study concluded that 00:29:15.240 --> 00:29:18.720 50 to 60 percent of all U .S. arms contain calculation 00:29:18.720 --> 00:29:21.200 errors. Wait, stop. You need to repeat that. 00:29:21.259 --> 00:29:23.960 Half. You're telling me that it's possible that 00:29:23.960 --> 00:29:26.019 half of the adjustable mortgages in the country 00:29:26.019 --> 00:29:28.200 were calculated wrong? According to that government 00:29:28.200 --> 00:29:31.359 study, yes, a coin flip. Total overcharges were 00:29:31.359 --> 00:29:34.579 estimated in excess of $8 billion. And the causes 00:29:34.579 --> 00:29:37.079 were just mundane clerical errors. The wrong 00:29:37.079 --> 00:29:39.279 index date was selected. An incorrect margin 00:29:39.279 --> 00:29:41.799 was applied. The caps were ignored. Just simple 00:29:41.799 --> 00:29:43.859 data entry errors that cost people thousands 00:29:43.859 --> 00:29:46.180 and thousands of dollars. That is terrifying 00:29:46.180 --> 00:29:48.660 because most people just pay the bill. You assume 00:29:48.660 --> 00:29:51.359 the bank's computer knows math. If I get a statement 00:29:51.359 --> 00:29:53.579 from a major bank, I don't get out my calculator 00:29:53.579 --> 00:29:56.019 and look up the historical LIBOR rate from 45 00:29:56.019 --> 00:29:59.720 days ago. Who does that? Nobody does that. We 00:29:59.720 --> 00:30:02.440 assume the computer is right. But the computer 00:30:02.440 --> 00:30:05.200 is only as good as the data and the programming 00:30:05.200 --> 00:30:07.880 that was entered into it. And with ARMs, the 00:30:07.880 --> 00:30:11.440 complexity just breeds mistakes. If a clerk hits 00:30:11.440 --> 00:30:14.819 the wrong key and your rate is off by just... 00:30:15.200 --> 00:30:18.140 0 .125%. You think, oh, that's just pennies. 00:30:18.180 --> 00:30:20.640 It's pennies this month. But thanks to the magic 00:30:20.640 --> 00:30:23.059 of compound interest and the fact that this error 00:30:23.059 --> 00:30:25.819 sets the new baseline for every future adjustment 00:30:25.819 --> 00:30:29.940 over 5 or 10 years, that single keystroke could 00:30:29.940 --> 00:30:33.220 cost you $10 ,000 or $15 ,000. And the bank never 00:30:33.220 --> 00:30:36.170 catches it. Or never admits it. Well, the errors 00:30:36.170 --> 00:30:37.829 are usually in their favor. There's not much 00:30:37.829 --> 00:30:39.829 incentive to go looking for them. And it wasn't 00:30:39.829 --> 00:30:41.609 just errors. There was, of course, predatory 00:30:41.609 --> 00:30:43.990 lending. We touched on this with the option ARM, 00:30:44.029 --> 00:30:46.349 but the scale of it, it's breathtaking. This 00:30:46.349 --> 00:30:48.470 is where the intent turns from incompetent to 00:30:48.470 --> 00:30:50.690 malicious, especially in the years leading up 00:30:50.690 --> 00:30:53.549 to 2008. The source notes that in 2006, over 00:30:53.549 --> 00:30:57.250 90 % of all subprime mortgages were ARMs. 90%. 00:30:57.250 --> 00:30:59.630 So the most vulnerable borrowers were given the 00:30:59.630 --> 00:31:02.930 most volatile product by design. These loans 00:31:02.930 --> 00:31:05.410 were sold to people who the banks and brokers 00:31:05.410 --> 00:31:09.329 knew for a fact could not afford the fully indexed 00:31:09.329 --> 00:31:11.990 rate. They could only afford the teaser. The 00:31:11.990 --> 00:31:14.109 business plan was never for them to pay off the 00:31:14.109 --> 00:31:16.630 loan. So what was the plan? The plan was for 00:31:16.630 --> 00:31:19.549 them to refinance or sell the house at a profit 00:31:19.549 --> 00:31:22.849 before the reset date hit. It was a giant game 00:31:22.849 --> 00:31:25.329 of musical chairs fueled by ever rising home 00:31:25.329 --> 00:31:28.009 prices. And when the music stopped and house 00:31:28.009 --> 00:31:30.259 prices started to fall. They couldn't refinance. 00:31:30.259 --> 00:31:32.900 They couldn't sell. The reset hit. The payment 00:31:32.900 --> 00:31:35.339 doubled or tripled. And the foreclosures started 00:31:35.339 --> 00:31:37.660 like an avalanche. There's also a specific scandal 00:31:37.660 --> 00:31:40.380 mentioned in Ireland, the tracker mortgage scandal. 00:31:40.799 --> 00:31:43.480 This one really, really boils my blood because 00:31:43.480 --> 00:31:46.519 it's so brazen. This is a classic case of banks 00:31:46.519 --> 00:31:48.240 trying to change the rules of the game when they 00:31:48.240 --> 00:31:51.000 start losing. In Ireland, tracker mortgages were 00:31:51.000 --> 00:31:53.599 very popular. They were tied directly to the 00:31:53.599 --> 00:31:57.059 European Central Bank, or ECB, rate. When the 00:31:57.059 --> 00:31:59.900 ECB slashed rates to basically zero after the 00:31:59.900 --> 00:32:02.559 crisis to stimulate the economy, these mortgages 00:32:02.559 --> 00:32:05.220 became extremely unprofitable for the Irish banks. 00:32:05.519 --> 00:32:07.480 So the banks were losing money on them because 00:32:07.480 --> 00:32:09.940 the rate was too low. They made a bad bet. Yes. 00:32:10.460 --> 00:32:13.180 they were stuck receiving tiny interest payments 00:32:13.180 --> 00:32:15.900 and instead of sucking it up i mean remember 00:32:15.900 --> 00:32:18.150 that The contract is supposed to hold risk for 00:32:18.150 --> 00:32:21.549 both sides. The banks deliberately and systematically 00:32:21.549 --> 00:32:25.190 moved customers off these products in what the 00:32:25.190 --> 00:32:27.430 Central Bank of Ireland later called an adverse 00:32:27.430 --> 00:32:29.869 manner. How did they do that? They would call 00:32:29.869 --> 00:32:32.529 people and offer them a better deal to switch 00:32:32.529 --> 00:32:34.890 to a new fixed rate, which was actually higher. 00:32:35.069 --> 00:32:37.529 Or they would deny them their contractual right 00:32:37.529 --> 00:32:39.609 to return to the tracker rate after a period 00:32:39.609 --> 00:32:42.170 of fixing was over. They essentially just lied 00:32:42.170 --> 00:32:44.450 and forced people onto higher variable rates 00:32:44.450 --> 00:32:46.789 to boost their own profits. So they cheated. 00:32:46.910 --> 00:32:49.190 They realized the bet went against them, so they 00:32:49.190 --> 00:32:51.750 just flipped the table over. It resulted in a 00:32:51.750 --> 00:32:55.490 massive years -long investigation. It's a stark 00:32:55.490 --> 00:32:57.730 reminder that in the borrower -lender relationship, 00:32:57.950 --> 00:33:00.509 the power dynamic is just so heavily skewed. 00:33:00.650 --> 00:33:03.349 The bank has the lawyers, the data, and the control. 00:33:03.710 --> 00:33:05.849 It really emphasizes that you have to watch your 00:33:05.849 --> 00:33:08.829 lender like a hawk. You cannot be a passive participant 00:33:08.829 --> 00:33:11.569 in this. You absolutely cannot. You have to be 00:33:11.569 --> 00:33:13.410 your own advocate. Let's touch on a couple of 00:33:13.410 --> 00:33:15.269 technical nuances before we wrap this all up. 00:33:15.660 --> 00:33:17.960 Pre -payment. This seems like a bright spot. 00:33:18.119 --> 00:33:20.740 Right. Can I pay off an ARM early without getting 00:33:20.740 --> 00:33:23.920 hit with a penalty? Generally, yes. Most ARMs 00:33:23.920 --> 00:33:26.460 allow you to pay extra principal without a penalty, 00:33:26.579 --> 00:33:28.819 though you always have to check the fine print 00:33:28.819 --> 00:33:30.720 for specific penalties in the first three to 00:33:30.720 --> 00:33:33.559 five years. But there is a nuance to how it helps 00:33:33.559 --> 00:33:36.049 you that is different from a fixed loan. How 00:33:36.049 --> 00:33:38.329 so? What's the difference? If you prepay principal 00:33:38.329 --> 00:33:41.029 on a standard fixed loan, you shorten the term. 00:33:41.130 --> 00:33:43.430 You might knock years off the end of the loan, 00:33:43.549 --> 00:33:46.309 but your required monthly payment stays exactly 00:33:46.309 --> 00:33:49.509 the same. Right. On an ARM, because the payment 00:33:49.509 --> 00:33:53.109 is recalculated at every adjustment period, prepaying 00:33:53.109 --> 00:33:55.509 principal often lowers your required payment 00:33:55.509 --> 00:33:57.049 at the next adjustment. Oh, that's interesting 00:33:57.049 --> 00:33:59.750 because the calculation is based on the new lower 00:33:59.750 --> 00:34:03.480 remaining balance. Exactly. The loan essentially 00:34:03.480 --> 00:34:06.619 re -amortizes itself over the remaining term 00:34:06.619 --> 00:34:09.179 at the lower balance every time it adjusts. So 00:34:09.179 --> 00:34:11.119 if you come into some money and pay down $20 00:34:11.119 --> 00:34:13.860 ,000, your required monthly payment will actually 00:34:13.860 --> 00:34:16.900 drop at the next reset. It improves your cash 00:34:16.900 --> 00:34:19.079 flow immediately rather than just shortening 00:34:19.079 --> 00:34:21.820 the far -off loan end date. That's a useful feature 00:34:21.820 --> 00:34:23.519 if you're trying to manage your monthly budget 00:34:23.519 --> 00:34:26.280 flexibility. It is. It gives you a knob to turn, 00:34:26.420 --> 00:34:29.360 a little bit more control. And finally, pricing. 00:34:30.510 --> 00:34:32.650 How do banks even figure out if these complex 00:34:32.650 --> 00:34:36.119 loans will make them money? It just sounds like 00:34:36.119 --> 00:34:38.880 pure chaos with all these moving parts and caps 00:34:38.880 --> 00:34:41.159 and indices. They use brute force mathematics. 00:34:41.380 --> 00:34:43.780 The source mentions they use Monte Carlo simulations. 00:34:43.940 --> 00:34:46.800 Like the casino in Monaco? Similar concept. They 00:34:46.800 --> 00:34:50.159 run 10 ,000, maybe 100 ,000 computer simulations 00:34:50.159 --> 00:34:53.059 of possible future interest rate paths. What 00:34:53.059 --> 00:34:55.760 if rates go up 1 % a year for five years? What 00:34:55.760 --> 00:34:57.880 if they drop? What if they stay flat? What if 00:34:57.880 --> 00:35:00.239 they spike violently and then drop? So they are 00:35:00.239 --> 00:35:03.280 playing out every possible multiverse scenario 00:35:03.280 --> 00:35:06.449 for the economy. They calculate the bank's total 00:35:06.449 --> 00:35:09.329 cash flow and profit for every single one of 00:35:09.329 --> 00:35:12.510 those scenarios to find the fair value and the 00:35:12.510 --> 00:35:14.969 effective interest rate of the loan. So when 00:35:14.969 --> 00:35:17.289 they offer you a rate, they have mathematically 00:35:17.289 --> 00:35:19.989 proven to themselves that on average, over all 00:35:19.989 --> 00:35:22.489 those tens of thousands of futures, they will 00:35:22.489 --> 00:35:24.849 win. So they have played out the next 30 years 00:35:24.849 --> 00:35:27.710 10 ,000 times before they even offer you the 00:35:27.710 --> 00:35:30.829 loan. Meanwhile, you are just guessing based 00:35:30.829 --> 00:35:32.389 on what your brother -in -law told you about 00:35:32.389 --> 00:35:34.670 the economy. Exactly. It is the definition of 00:35:34.670 --> 00:35:37.269 asymmetry of information. They have calculated 00:35:37.269 --> 00:35:39.889 the odds down to the hundredth of a decimal point. 00:35:40.070 --> 00:35:42.989 You are walking into the casino blind. That is 00:35:42.989 --> 00:35:46.329 humbling and a little terrifying. It is. It's 00:35:46.329 --> 00:35:48.030 why understanding the rules of the game is so 00:35:48.030 --> 00:35:49.909 important. Okay, let's try to land the plane 00:35:49.909 --> 00:35:51.710 here. We've covered the mechanics, the risks, 00:35:51.869 --> 00:35:54.389 the history, the global context, and the scandals. 00:35:54.670 --> 00:35:56.809 What does this all mean for you, the listener? 00:35:57.190 --> 00:35:59.150 It means that an adjustable rate mortgage is 00:35:59.150 --> 00:36:02.489 a powerful tool, but it is a very, very sharp 00:36:02.489 --> 00:36:04.909 tool. I like to think of it like a chainsaw. 00:36:05.030 --> 00:36:07.730 It can cut down a tree very efficiently or it 00:36:07.730 --> 00:36:10.170 can cut off your leg. It is not a set it and 00:36:10.170 --> 00:36:12.510 forget it product like a fixed mortgage. You 00:36:12.510 --> 00:36:14.610 are entering into an active partnership with 00:36:14.610 --> 00:36:16.650 the financial markets. You are the shock absorber. 00:36:16.730 --> 00:36:20.230 You are. And the tradeoff is crystal clear. You 00:36:20.230 --> 00:36:22.429 get a discount on your payment today in exchange 00:36:22.429 --> 00:36:24.429 for taking on the risk of inflation tomorrow. 00:36:24.889 --> 00:36:27.449 If you understand that risk and you have the 00:36:27.449 --> 00:36:29.889 cash reserves to handle a payment shock, it can 00:36:29.889 --> 00:36:32.710 be a savvy financial move. The scholarly view 00:36:32.710 --> 00:36:35.570 supports that if you have the financial resilience 00:36:35.570 --> 00:36:38.449 to weather the storm, the ARM often wins in the 00:36:38.449 --> 00:36:40.849 long run. But if you are stretching every dollar 00:36:40.849 --> 00:36:43.849 just to afford that low teaser rate, then you 00:36:43.849 --> 00:36:46.099 are walking into a trap. Whether it's a simple 00:36:46.099 --> 00:36:49.280 rate reset, a cap breach, or that dreaded negative 00:36:49.280 --> 00:36:52.000 amortization recast, the math will eventually 00:36:52.000 --> 00:36:54.019 catch up with you. And you can't even trust that 00:36:54.019 --> 00:36:56.639 the math on your statement is correct. No. And 00:36:56.639 --> 00:36:58.420 that, for me, is the biggest takeaway from all 00:36:58.420 --> 00:37:00.739 of this. The complexity of these products is 00:37:00.739 --> 00:37:03.260 so high that even the banks, with all their resources, 00:37:03.639 --> 00:37:05.699 struggle to administer them correctly. Which 00:37:05.699 --> 00:37:07.880 brings us to our final provocative thought. It 00:37:07.880 --> 00:37:10.340 does. We live in a world of automated finance. 00:37:10.659 --> 00:37:13.639 We click accept, we set up auto pay, and we assume 00:37:13.639 --> 00:37:15.719 the institution has it right. We put our faith 00:37:15.719 --> 00:37:18.599 in the algorithm. But history specifically, that 00:37:18.599 --> 00:37:23.000 1995 study showing a 50 to 60 % error rate, suggests 00:37:23.000 --> 00:37:25.900 that faith might be misplaced. So here is the 00:37:25.900 --> 00:37:28.079 question for you listening right now. If you 00:37:28.079 --> 00:37:31.079 have an ARM, or if you had one in the past, have 00:37:31.079 --> 00:37:34.800 you ever actually audited the bank's math? Did 00:37:34.800 --> 00:37:36.739 you ever go back and check if they used the right 00:37:36.739 --> 00:37:39.460 index on the right date? Did you check if the 00:37:39.460 --> 00:37:42.280 margin was added correctly? Did you verify every 00:37:42.280 --> 00:37:45.280 single cap? Because in a game where the house 00:37:45.280 --> 00:37:47.900 has already calculated how to win, the one time 00:37:47.900 --> 00:37:50.380 they might be variably wrong, either in your 00:37:50.380 --> 00:37:53.159 favor or more likely against it, is in the fine 00:37:53.159 --> 00:37:55.480 print of that monthly calculation. Maybe it's 00:37:55.480 --> 00:37:57.559 time to get out the calculator and that old contract. 00:37:57.940 --> 00:38:00.340 Thanks for listening to The Deep Drive. Stay 00:38:00.340 --> 00:38:00.800 curious.