Challenge for the bears - bring it on!!

Union Square? Christ. A farmers market, filenes basement and what else? Oh yeah, homeless drug addicts. And the housing stock sucks.

"Since the forward curve is usually _not_ inverted, according to your logic, at most times in history you would have to assume that mortgage rates are going up and therefore MUST factor in a capital loss on the back-end."

That's a fair point. Although perhaps the relative steepness of the curve might give you pause. There is non-inverted, and then there is non-inverted.

Essentially, you're saying that as best you can tell 5% mortgage rates & 3% 10-year yields will remain 10 years from now because you believe the curve will remain fully static. My 2% increase argument was that it will shift fully forward. Reality is that it tends to go somewhere in between, which was what I was getting at with my 1.5% argument. Maybe 1% would have been better, but I wouldn't bet on it staying fixed fully and not shifting forward at all.

Here's the real question, though. If you see a fixed yield curve going out 10+ years with 5% 30-yr mortgages and 3.25% 5/1 ARMS, why did you give poor 300_mercer crap about doing a 5/1? The guy's just taking your interest rate view and applying the next logical step. Why be a sucker and pay 5% if you can do 3.25% dropping to 2.50% in 5 years once it starts floating? ;)

"Now, has whoever bought (and sold) the bond overpaid for it? Have they secured low-rate funding? Is it a netting transaction?"

I was suggesting that when the market is willing to pay you 5% (say) for money lent between years 5-20 today, then saying "I'll use cash I have if interest rates are higher than 3%" is a form of undercharging cost-of-capital.

Inonada: I stand corrected.

Maklo: Now I'm even more mystified by why you are willing to spend $400k and accept personal liability on an additional 1.6m for an asset that, if I'm reading your numbers right this time, you think generates 96k implicit gross income (i.e., is equivalent to 8k/mo rental) of which 53k goes to expense before before interest. That looks like an unlevered return of about 2%. When you lever it up at 5%, the interest on the loan is 80k, 17.5 of which you expect the US to pick up -- so you appear to expect a loss of 19.5k per year.

You then project that you'll make this up when you sell and even something on your investment. But your projected sale requires that you sell to someone who, like you, is willing to accept a negative expected return on the cash flow and has very large tax subsidies. Leave aside Nada's quite reasonable concerns about whether your interest rate assumptions are consistent. You project selling at $2.8m with a rental value of 12k/mo. That means that for your buyer, interest plus maintenance -- just the hard costs -- will be more than the equivalent rent. Is that plausible?

What if you sell into a market in which homeowners expect to earn more than the bank, on cash flow alone, and don't count future appreciation because they don't expect their buyers to do so either?

Another question:

Is it plausible to assume that prices will continue to include the full projected value of the tax subsidy?

I think not. Buy-to-let investors don't get that subsidy. So if prices include it, those investors can make money by selling to homeowners. If prices stay like this, won't the entire middle-class-and-up rental market disappear? (indeed, in many US cities, there isn't much of an upper middle class rental market)

So it seems more likely that either (a) investors set the price, and lucky owner occupant buyers get to keep their tax subsidy without giving it to sellers, or (b) investors sell out and rentals become rare. If the former, prices need to drop enough to just about wipe out your entire downpayment. If the latter, there are a lot of rental units that are going to be up for sale. is it plausible to assume that the NYC condo/coop market supply could increase by 5x or so with no impact on prices?

Nada, I can find 16x in Carnegie Hill. You can buy a $6000 2 bed rental for $1.1mm. I look at sales in 10 E 85th Street for comps to my rental.

I also look at 40 E 88th where the maintenances are low.

Nada - I think both 5/1 and 30 are fair. Not sure how I was giving 300_Mercer "sh*t" but I just wanted to remove interest rate risk and prognostication from the analysis - one less variable to think about in a complex equation. I am actually considering financing my 30yr with a 5/1 ARM in a couple of months, am almost ambivalent at this point.

FinanceGuy - I haven't responded to you because you sound like someone who just read Brealey and Myers and I just have no idea what you are talking about. Maybe you are just over my head. But I would think about the difference between nominal and real interest rates and realize that we live in a real world.

inonada:
"when the market is willing to pay you 5% (say) for money lent between years 5-20 today"

poster has stated, though, that these funds will otherwise be kept "in cash." So, wouldn't in this particular situation the relevant opportunity cost be the yield on a money market account (for the deposit, too)?

"poster has stated, though, that these funds will otherwise be kept "in cash." So, wouldn't in this particular situation the relevant opportunity cost be the yield on a money market account (for the deposit, too)?"

If you want to live in make-believe land, sure.

If you have the option of:

1) Having your cash freely available over years X.
2) Loaning out your cash to others (buying a bond) over years X.
3) Promising to use that cash to maintain an investment over years X.

Then 2 & 3 are equivalent, not 1 & 3. If the bank is paying 0% on cash, and I come up to him and say I'll pay him 1% to lend me for 10 years, and my argument is "but you're only earning 0% anyways", do you see the problem with that?

"I think both 5/1 and 30 are fair."

Got it, that explains it. You think the interest rate curve will remain exactly fixed from here on out. In 10 years we'll still have 0% short-term rates, 1.5% 5-year rates, 3% 10-year rates, and 4% 20-year rates. Your expected case is that absolutely none of the forward curve will come into play. Inflation will remain at 2% but the the Fed will continue with 0% interest rates for over a decade.

Under this expected case, you'll tread water with a 5% return on equity annually on a 20% downpayment. If you are wrong by 0.5% and we have 3.5% 10-year rates and 5.5% mortgages, your ROE will be 0%. The problem with this view is that one can do higher-reward, lower-risk spreads that give superior outcomes. Suppose I plan on borrowing a 10-year now and a 10-year in 2021 to finance a 20-year long. If the 10-year rate in 2021 ends up being 3.5%, I will still make money. My ROE doesn't hit 0% until 5%. At 5%, your ROE hits -15% annually.

So even if your expected case is right (which I don't think it is), your assumptions vastly underprice risk.

Rhino, I'm with you on being able to purchase a bread-and-butter Carnegie Hill coop at $1000 a sq ft or so. Now I don't know the details of the rental market there and what you have exactly, but $5.50 a sq ft seems awfully rich to what one could actually achieve. The median ask in the neighborhood for a 2BR/2BA is $5, and if Carnegie Hill is anything like other neighborhoods you can usually do appreciably better than the median ask. If apt23 is reading this & sipping coffee on her terrace directly overlooking Central Park (how awesome is that?) while paying $5-ish a sq ft, I would think one could do better than $5.50 in bread-and-butter apts in Carnegie Hill.

> If apt23 is reading this & sipping coffee on her terrace directly overlooking Central Park

inonada:

Agreed that an argument along the lines of "if my rate rises, I just use my cash to pay off the mortgage, therfore I am not exposed to rising rates" is unsound.

But if we take the ARM out of the equation as maklo suggests and assume a 30yr or 15yr fixed mortgage and look at the two real-world scenarios poster is trying to compare:

1) buying apartment and paying off mortgage
2) paying rent and keeping any available funds in cash

It seems for a 20yr hold and using purchase price and rent equivalent as given by original poster (and for a wide range of assumptions on reno, transaction costs etc.) poster will do better under 1) unless property value collapses drastically.
Obviously, this is not a general result. If you invest your funds, depending on the return you are able to achieve, 2) may be better.

"You think the interest rate curve will remain exactly fixed from here on out."

No that is not what I meant. I clarified later that what I meant was that I am ambivalent on choosing between the 5/1 and 30yr today.

I am sure mortgage rates will fluctuate over the next 10 years. I would even lean to say that they have modestly more upward bias than downward. However, I just don't agree that that alone translates into a direct price decrease as you factored in prices are determined by a far more complex set of factors.

Maybe it is easier if we just think about everything in real terms with the 2% inflation assumption. In that case, someone who buys today with a fixed mortgage will see a major chunk of their monthlies declining in real terms, as the fixed nominal interest charges decrease at 2% per year in real terms. Meanwhile, assume real rents stay the same in real terms. In this case you come out even better in the later years.

A 30-year jumbo mortgage at 5%, factoring in the tax deduction and adjust to real terms, is really at 1.5%.

There is no guarantee that interest rates will be at 2%, even though that is the expectation. Deflation is the major risk factor for a debtholder, as the real value of his nominal liability goes up over time. Now that is a whole other debate, but my view is that the chances are unlikely for major deflation.

"Rhino, I'm with you on being able to purchase a bread-and-butter Carnegie Hill coop at $1000 a sq ft or so. Now I don't know the details of the rental market there and what you have exactly, but $5.50 a sq ft seems awfully rich to what one could actually achieve. The median ask in the neighborhood for a 2BR/2BA is $5, and if Carnegie Hill is anything like other neighborhoods you can usually do appreciably better than the median ask."

Nada - apologies as I may not know the data here (is $1,000/sf median in Carnegie Hill?) but comparing the median of buys with a better-than-median "bargain" on the rents, is a skewed approach and that will lead you to draw a biased conclusion.

carnegie hill = 86 to 96 west of lex--1000$ psf median?? in shithole buildings??

another morning spunked away nada. congrats!

Polison/Maklo:

The issue is why you are considering "opportunity costs" in the first place. If your goal is to compare the cost **to you** of buying or renting, then it might seem logical to use your own personal alternative investments to measure opportunity costs.

But the largest unknown in the cost of buying is future resale value. And both Maklo and Inonada's conclusions are completely dependent on their guesses about that.

So I think the actual exercise is to run this analysis twenty years down the line, not now: not what you'd be willing to pay, but what your buyer will be willing to pay, because your maximum price is entirely dependent on your prediction of your buyer's. (That's more Keynes than B&M).

If you are trying to predict future prices, your personal opportunity cost is irrelevant. What is important is your buyer's opportunity cost.

Opportunity cost to the MARKET (i.e., future buyer) should reflect the risk of THIS investment, not a different one, because usually the market tends to price similar risks similarly (that is the Brearly & Myers point).

It's hard to know how a future market at an unknown time will perceive or price risk. But I think we can safely assume two things.

1. The equity interest in an investment is always riskier than a senior debt interest in the same investment. So **usually** the market will price the opportunity cost of the downpayment higher than the mortgage rate. (It is not doing so now-that's key evidence of a bubble).

If you use the same rate for downpayment and mortgage, as Maklo does, or a lower rate, as Polison suggests, to predict **future** prices, you are asserting that your buyer -- the market -- is going to be willing to take on risk without compensation. Or you are asserting that the banks have their number wrong now, so that current mortgage rates are a bad predictor of future mortgage rates.

Markets are wrong all the time, but I'm interested in understanding the story you are telling that leads you to predict a future error in a particular direction.

Do you think the banks are over-charging for risk, so the mortgage rate is higher now than it is likely to be in the future? Inonada thinks the reverse, that the Fed is paying banks to ignore various risks and probable inflation, so the current mortgage rate is actually lower than it is likely to be later. If you think he's wrong, why?

Or do you think that homeowners will continue to be willing to assume risk for free, proving B&M wrong? Conventional finance theory is wrong on many issues; most obviously, it has trouble with bubbles. Still, it's backed by strong arguments and it is worth exploring why you think it is wrong here, if you do.

2. Basic sophomore micro-economic pricing theory--Adam Smith via Samuelson and Mankiw--says that when producers can make extraordinary profits, they usually do. Increased supply then tends to compete away the excess profit opportunity.

Maklo, you are willing to give the entire value of your tax subsidy to your seller. Fine, it's a free country and, more importantly, that's current market pricing. But you also predict your buyer will return the favor.

Unless I'm missing something, this pricing means that investors who purchased to rent can make more money by selling to you than by continuing to hold to rent. That's an unstable situation. If this market works the way Adam Smith suggests, investors will move apartments from rental to owner-occupied until prices adjust so that this profit opportunity disappears. You may be willing to give away the tax subsidy, but the market is unlikely to reciprocate when you sell.

Sophomore economics is often over-simplified. I'm interested in your story of why you think the simple prediction -- that prices will drop until investor-landlords are indifferent between holding/renting and selling to people who get the tax subsidy -- is wrong.

Finally, Maklo, when I try to run your analysis, using your assumption of 2% inflation and your theory of what a buyer like you would pay, 20 years from now, it doesn't generate the number you predict for your resale value. Even using your view, not mine, on the first two points. So I'm trying to understand what you are doing differently.

FG, I really have no clue what you are talking about.

All I can glean from is that you have some theory that leads you to believe that asset prices are high today (a "Bubble") and thus should decline over time because of "basic sophomore micro-economic pricing theory" laws on supply and demand. Therefore you assume prices going down from $2M to $860K in ten years. Real? Nominal? Didn't specify. And then you say you must factor that capital loss into the buy vs. rent equation and BIG SURPRISE, you come out favoring the rent side.

So effectively you are saying Buying is too Expensive because ... Buying is too Expensive. Why are Prices too Expensive? Because we are in a Bubble. Why are we in a Bubble? Because Prices are too Expensive.

Such circular logic is just too much for my brain to handle.

I'm just trying to understand how you concluded that paying $2m plus 53k/year for a unit that rents for $96k / year is appropriate.

I already said I was wrong about the $860k number -- that's what a simple model suggests the cash flows would be worth at a 15% return, which I mistakenly thought you were projecting. But you aren't. In fact you are predicting a negative return on an ongoing basis, which you expect to make up on sale. So I'm trying to understand your sale price prediction.

I'm asking you three pretty simple questions about your model to understand why you expect to make money when you sell.

1. You use the same rate for the "opportunity cost" on your downpayment as the current bank mortgage rate. Your calculations seem to assume that in 20 years, you'll be able to sell to someone who makes a similar calculation (am I misunderstanding here?). Usually investors demand more expected return for a downpayment than for a loan. Why do you think the usual rules won't apply when you sell?

2. Your calculation gives your seller the benefit of the buyer's tax break. I think you are also assuming that when you sell, your buyer will do the same for you. That isn't what usually happens, for reasons I tried to explain. Why do you think the usual rules won't apply when you sell?

3. You project that (a) rents and expenses will increase by 2%/year for 20 years, while interest rates, taxes and risk premiums remain the same, and also that (b) the price of your unit will increase by the 2% per year. But when I plug in these inflated numbers from (a) to **your** model of how much the unit is worth, it doesn't generate (b). Why do you think (b) is right anyway?

As for my own views, I think that current prices are too high for a number of reasons, but one that is quite striking in your case is that you seem to be projecting a 2% return on a cash purchase which you propose to finance at 5%. Borrowing at 5% to earn 2% makes sense only in limited circumstances.

My initial hypothesis was that you were making the classic bubble mistake -- bubbles happen when enough people decide that it's ok to pay a price disconnected from fundamentals (such as rental value) because they expect to be able to sell for even more to someone else who reasons in the same way.

Your later comments suggest that instead you might have some view of the fundamentals that I'm not seeing in the price model. My goal here is to figure out if I made a mistake understanding your model, or you made a mistake constructing it.

does anyone on this thread have a job?

PS: I should not have called bubble thinking a "mistake." Anyone who did a bubble analysis from 2000-2006 was definitely right.

Bubbles happen because if enough people decide that fundamentals are irrelevant, they are right -- fundamentals become irrelevant. The only mistake is thinking that bubbles will continue forever or they will never occur.

Lucille -- my job does include trying to use this stuff, so I can sort of justify working through it here as on-the-job remedial training.

>The only mistake is thinking that bubbles will continue forever or they will never occur.

What about the lazy theory of "reversion to the mean"?

Ok FG. Let try this another way. A few questions for you.

You think the general market is overpriced today. By what percentage, roughly speaking?
What is your view on the % rise/fall in household real income in Manhattan? Particularly for the demographic that can afford a $2M apartment?
What is your view on inflation/deflation over the next ten years?

Don't need to be precise, ranges are fine.

LB - I quit my job a long time ago.

oh are you a housewife like me?

polisson, the two choices you pose are to either stay in cash for 30 years, or to buy an inflated risky asset and hold for 30 years. Given those choices, the latter will pretty much always win. Even buying S&P at the top of the 2000 bubble will likely win on those terms.

However, the world is not limited to just those two choices. There are other better-priced risky assets out there. There are opportunities to be had. Those that had cash in 2009 had opportunities in the stock market that were simply not available to those that had their cash tied up.

What is most out-of-touch with "tie up money in house for 30 years" vs. "stay in cash for 30 years" set of choices is that the choice of "tie up money in a bond for 30 years" is not given.

Lucille, I think these were the guys who sat around a campfire when they were young and tried to outdo each other by telling the scariest stories . Of course they did't realize that everybody except the last two had fallen asleep . ( Unless, of course they were all involved in a circle jerk, which has what's become of this thread) .

wait...you forgot to add how rich you are.

columbiacounty, you forgot

inonada,

I am well aware of the fact that the world is not limited to those two choices and had tried to indicate this at the end of my post.

I was merely saying that for the original poster, given that they have decided to limit themselves to the two choices of buying the, as you say, inflated, asset or staying in cash, it seems very likely that they will do better buying.

And there, right on cue is CC, the crazy old coot, troll of SE. Pathetic , really .

why?

It's sort of sad how dependent CC is on AR, his remote Internet love. Pathetic .

Yes, but it's w67thstreet he would jump off the cliff for.

No, Apt23 would jump off the cliff for west67th, CC ONLY for AR.

Yes, we all know that apt23 wants w67thstreet to go through her backdoor,

but if you search back, you'll find that columbiacounty's explicitly stated opinions don't fall much short of that.

that's actually true, ph41. it got pretty bizarre, i remember when they used to talk like that and cc had the light of life in him. that light belonged to his boy w67th. only in the aftermath did he completely latch on to ar. he used to be like a real person and stuff. crazy.

lucille, what do you think changed columbiacounty? Who was looking through that window in his shower?

i think columbiacounty looked out that wet window into the cold dark night and what he saw was so gruesome and offensive that is permanently damaged him. what he saw out there, hb, was a reflection of his own soul.

You really think that his reflection was visible?

having fun?!

i think that was the one and only time he was able to see his reflection. because once he saw it, it consumed him entirely and he became his own reflection. the real columbiacounty is lost in the realm of wet dirty naked men who scrub and scrub but just cannot get clean. they can never get clean, hb. it's a fate worse than hell, hb. because it's cold and they're wet and shivering. at least hell sounds nice and toasty.

No Lucille, you know that thoughts, dreams, fantasies of AR keep CC warm. That's why he's so pathetic- his Internet love affair is his only comfort.

"I was merely saying that for the original poster, given that they have decided to limit themselves to the two choices of buying the, as you say, inflated, asset or staying in cash, it seems very likely that they will do better buying."

Yes. Although one has to question the choice of "I will put $1M in cash for the next 30 years and not touch it no matter what opportunities may come, even though I do not know what they are yet".

I guess if he had given us the choices of "I'm going to burn all my money in a big bonfire" and "I'm going to flush 90% of my money down the toilet" we'd be talking about how the latter is the right choice given the options.

but they know eachother in real life. i distinctly remember ar saying to him i'll see for dinner..up in cc i guess, she has a house up there too, remember?

I think AR has written CC off by this point. She's got her own problems and doesn't need all of CC's problems and zero solutions. It was fun 2 years ago when they were all in the same camp, but now with AR buying a place and going on vacation without CC, plus AR, for all her problems and mental issues, at least married a productive husband whereas CC is toward the bottom of his long decline stage...

Thanks every one for the fun discussion. If I may summarize the crux of the arguments against buying:
1. Nada can find a great apartment which would trade for >1100 per sq ft at under $4 per sq ft in prime areas despite his not able to post such current listings and average rent for such listing being close to $5 per sq feet.
2. There is no place for large amount of cash in a portfolio. Also, people keep cash because of a lack of investment oppourtunities not because they need ample liquidity as a renter but less if they have a paid for apartment.
3. Rates will go up but not the rents - highly unlikely in my view. Hence buying an apartment is not an inflation protection - economic history tells us otherwise, the prices of real assets go up with inflation regardless of the level of rates as you can always refinance.
4. Govt is throwing away free money in form of low rates. No need to take it.
5. 5/1 rates are riskly as the rates will go up in five years time but rents are not going to go up.
6. Assume real estate is overvalued despite maklo/mine examples where is you use average rents, it is cheaper to buy with inflation protection thrown in for free.
7. Upkeep of the apartment cost $250 per sq feet every 20 years in NPV despite the fact that $250 per sq ft in NPV consitutes a gut renovation no an update to kitchen and bathrooms.
8. People can make 8% per year before taxes despite most pension funds being underfunded due to this assumption. Actual average returns are close to 5% and come at a cost of peace of mind.

NOTE: I was a bear on real estate till 2009 who would have given all pro-renting arguments and acutally rent instead of buying (not a permabear). But low rates, 15-20% lower prices, realizing the quality differnce between rentals and propeties for sale, and good liquidity changed it for me.

For the record, I am not the sensitive type and take criticism.

and please excuse my typos

>For the record, I am not the sensitive type and take criticism.

ok, you can take criticism, but how well do you deal with a 5 page essay by financeguy?

i did not read it

Excellent strategy.

Now that I read his comments, he seems like a wannabe finance guy or aspiring MBA student. MAKLO summarizes the flaw.

I thought I summarized the flaw. It's too long. People want to read what gets to the point. People want to hear arguments that are concise.

alanhart, I used the wrong it's.

you are hfscomm1.

that's short and to the point.

If I may comment on the transaction cost. Buying cost is -1% after paying mansion tax (coop not a condo, hence no other major expenses) yes negative as I get refund from the broker. Call it even after making minor upgrades/fixes to the apartment. Selling cost is 5% and if you want to live there for a long time, you can amortize over 20 years.

>you can amortize over 20 years.

Except why would you amortize anything? You have a cost that is 20 years out. Pay for it in 20 years. Not ratably over 20 years.

The 100,000 cost to sell you apartment in 20 years is worth about $50,000 today if you assume merely 3% returns and growth over that period.

I am assuming the increase in price will be offset by the NPV effect. If you are not assuming any increase in price like NADA, the cost as you say is 2.5%. If I may ask NADA, how are high end rental like related (caledonia, 1 Union Square South), 2 cooper, 55 Thompson are able to collect rent north of $6-7 per sq ft. Do they not count in the market? Are people renting them stupid?

Nada, if the rates go up, your tax savings go up.

Maklo --
-RE prices are still high enough that I haven't tried to quantify how high. Timing is probably more important anyway, and there I'm at sea. As a guess, perhaps 1998 plus inflation, plus or minus an error rate of 20% or so. Assuming no double dip recession or financial crises, error is more likely on the upside, since there are quality improvements the statistics don't pick up. To be more precise, I'd want data about construction and conversion costs and rents that I don't have. And I still wouldn't know how to predict timing -- in Japan it's taken decades, so long that you can't wait it out.
-I haven't seen any numbers on changes in income levels -- or asset levels, which is probably at least as important -- in the $2m market. Why do you ask?
-As to inflation, I have no particular expertise or reason to disagree with the market's guess -- about 2%. I think the main risk is on the downside. Politics matters here; without a change in personnel at the Fed, the demise of the Republican party or the rise of a new party that believes in full employment, it is hard to see how inflation gets materially higher, but the program-slashers could easily push us into deeper unemployment and another recession. I expect that eventually the dollar will have to decline significantly since that's the only way to reduce the trade deficit, but I'm not sure how a lower dollar affects NYC RE prices, and in any event, this could take years.
-- If I were in a pessimistic mood, I'd predict a revived bubble in high end RE, fed by an explosively profitable finance sector, hollowing the city out of all normal life, before a financial crisis leaving us a dark imitation of Dallas. But a beneficent Providence will, one prays, preserve us from that prospect.

Mercer: There is only one argument against imagining you are likely to make money by buying. Prices are highly likely to drop, because they are currently so high that investors can make money shifting rentals to sales, renovating obsolete units, or developing new ones. All the rest is commentary.

Also, undergraduate economics would be more helpful than an MBA to understand this. It's just Adam Smith. High prices make producing new supply profitable. Investors respond by producing. Increased supply tends to reduce prices. When price drops to the cost of production, profit disappears and investors stop producing.

The only reason rents, opportunity costs, interest rates, risk premia and all that MBA stuff are even relevant is to try to predict the price when the investors will stop creating new supply.

You can get almost the same answer by just asking yourself: if an institutional investor owned this unit as a condo, would it make more money renting it out for its entire useful life, or selling it now. If the answer is renting, then it's a good price to buy. If not, not.

"Selling cost is 5% and if you want to live there for a long time, you can amortize over 20 years."

You are forgetting the 2% in transfer taxes. And probably something upwards 1% in other misc crap.
http://www.corcoran.com/guides/index.aspx?page=ClosingCosts

Do you honestly plan on living there for the next 20 years?

"Nada, if the rates go up, your tax savings go up."

Yes, I already included the effects of the much-beloved tax savings in my numbers.

"Rates will go up but not the rents - highly unlikely in my view."

I'm not sure anyone here has said that rents won't go up. Certainly not me.

"economic history tells us otherwise, the prices of real assets go up with inflation regardless of the level of rates"

OK, I think I understand now: you were asleep for aughts. Search online for "housing bubble", read about how real asset prices went up at a much faster than inflation.

"If I may ask NADA, how are high end rental like related (caledonia, 1 Union Square South), 2 cooper, 55 Thompson are able to collect rent north of $6-7 per sq ft."

I'm not sure, I've never been in them: for me, they have looked like a complete waste of time compared to other nicer options. I suspect you've seen some of them and agree. Based on the 6+ month marketing period for 55 Thompson which includes no fewer than 119 listings for at most 39 units under a dizzying array of unit numbers used to hide their trail, it appears that the rest of the market agrees as well.

I would guess that:

1) There's some level of discounting that you & I are not privy to.
2) The tenants are more transitory renters rather than long-term renters.
3) The vacancy rates are high.

So are they really achieving the $6-7 on their inventory? I doubt it. They're just pricing themselves aggressively on the price/occupancy curve.

Now you might say "So what? If I want to rent that place long-term I'm gonna have to pay $6-7, and I'm not going to get introductory discounting for later years, and I'm not going to get any discounting for keeping the place fully-occupied." True. But life is full of options, not the ones you seem to like to box yourself into. How is it that apt23 is overlooking Central Park at $5-ish a sq ft with a multi-year lease at 3% increases? Rather than you asking me 55 Thompson, shouldn't you be asking apt23 about how she was able to achieve that?

I don't know how to live life boxed by an arbitrary set of rules. "I must either use my money for an apt or stay in cash for the next 20 years no matter what." WTF??? If 2009 comes, and S&P is trading at 700-900 for months on end at once-in-a-generation bargain valuations, and Buffett who opens his mouth to the general public but once a decade says "BUY" with no personal financial interest other than making a very public proclamation of how badass he will be proven to be yet again, I gotta stay in cash???

Wall Street Wielding the Ax
The trading slump on Wall Street has battered profits and is about to cost some people their jobs. "Banks are chopping a lot of wood, both deadwood and live wood," said one expert.

http://online.wsj.com/article/SB10001424052702303627104576414202416820980.html?mod=WSJ_hp_LEFTWhatsNewsCollection

1) How comparable is the US to Japan? In terms of extremes of the housing bubble, speed of policy response, demographics, direct industrial competition from South Korea, Taiwan and China. The answers here matter much in determining whether recovery happens in 30 years or 10.

"1998 plus inflation, plus or minus an error rate of 20% or so"

Case-Shiller NY is around 165 for NY right now. I only have data back to 2000, to which it is indexed at 100. Let's say 1998 was around 90. So +/- 20%, you are saying the appropriate price is somewhere between 72 and 108, or a further 35% to 56% drop from today. Price/rent ratios are around what ... around 20? So you are only comfortable if they fall to b/w 8 and 13?

- First, as real incomes in Manhattan are up materially since 1998, are you implying that people are also less willing to spend money on housing than before?
- Second, 8 price/rent? That is absurdly low.
- Third, that is a pretty wide range. And on top of that you have this timing uncertainty element. Frankly, not very meaningful for most people.

"Hey store clerk dude, how much is that pencil?" "Hmmm let me think ... somewhere between 10 cents and 10 dollars. Oh by the way, the price may go down though in the next 6 months. Wanna buy it?"

If you hold these views, there's really little else to discuss ... because I am not sure what other factors can overcome such extreme views.

2) Income level matters because that is what drives affordability. As real income goes up, real rents go up. As nominal incomes go up, nominal rents go up. There is much historical data to support this. And it makes a lot of common sense.

3) Inflation at 2% with some policy-driven tail risk. Fine, in agreement here.

>How is it that apt23 is overlooking Central Park at $5-ish a sq ft with a multi-year lease at 3% increases?

This is a complete lie.
Why don't you discuss apt23's place in Miami that she bought and is justifying that she continues to go deeper underwater?
Why don't you talk about the time that she falsely reported to the police that her husband was commiting a felony gun crime?

"Case-Shiller NY is around 165 for NY right now. I only have data back to 2000, to which it is indexed at 100. Let's say 1998 was around 90. So +/- 20%, you are saying the appropriate price is somewhere between 72 and 108, or a further 35% to 56% drop from today. Price/rent ratios are around what ... around 20? So you are only comfortable if they fall to b/w 8 and 13?"

Here are links to Case-Shiller, SE's index, and CPI with data going back as far as they have it:

http://www.standardandpoors.com/indices/sp-case-shiller-home-price-indices/en/us/?indexId=spusa-cashpidff--p-us----
http://streeteasy.com/nyc/market/condo_index
ftp://ftp.bls.gov/pub/special.requests/cpi/cpiai.txt

Jan 1998 was at 83. I think he means to inflation-adjust that number. Inflation has been 1.34x, so 111.6. Which works out to a 30% drop.

Arguably better is the SE index, which was at 807 and is now at 1835. Doing the same there works out to a 40% drop.

So if price/rent is 20-25x today, he's talking about a 12x-17.5x depending on which set of numbers you'd like to use.

"First, as real incomes in Manhattan are up materially since 1998, are you implying that people are also less willing to spend money on housing than before?"

There's a difference between spending more and getting more. Let me remind you of how the avg ppsf is up 3.6x while the avg constant-quality condo is up 2.6x over the past 15 years. If you look at the change in avg sale price, people you'll see a similar trend.

"As real income goes up, real rents go up."

I don't think there is any evidence to support this. Real rents in NYC have certainly not gone up for the 2 decades that Miller Samuel has data. If you think real incomes in NYC have gone up, then the data doesn't support your statement at all.

On a national level, you can look up the rent component of CPI. You'll see that it tracks overall CPI relatively flatly over decades. So real rents have been flat for decades. Yet real incomes have shot up lots.

As real incomes have gone up, people have spent more on real rent for superior apts in real terms. But constant-quality rents have stayed flat in real terms.

I completely agree with Inonada's last couple of posts.

I would only add that 8x isn't unreasonable at all -- it was the rule of thumb among NYC RE investors for most of the twentieth century (although actual prices were considerably lower during the years of heaviest suburban subsidy) and it is the standard in much of the real estate market in much of the country even today. For investors that expect to make money by holding to rent rather than by selling to speculators, it is hard to make a reasonable return if prices are much higher than 8-10x gross rents, unless real long term interest rates are extremely low.

Maklo, Sorry if you don't see anything to discuss. I thought you might be interested in explaining or understanding the bases of your prediction that RE prices will go up 2%/year, since that prediction is the core of your claim that current prices are reasonable.

After all, the standard mark of bubble thinking is believing that it's ok to pay more than the fundamentals warrant because you'll make it up when you sell. Bubble thinking has made many people rich and some of them poor again while causing huge misallocations of capital and tremendous collateral damage in the last couple of decades.

If your analysis boils down to predicting a continuing bubble, we (including you) might want to be clear about that -- whether or not we decide you are right. For one thing, it is another piece of evidence that this bubble still has a ways to float.

And if you aren't unwittingly relying on bubble logic, that's better still -- it means you have something to teach and I have something to learn, which is (for me at least) far more interesting than the prediction itself.

"There's a difference between spending more and getting more. Let me remind you of how the avg ppsf is up 3.6x while the avg constant-quality condo is up 2.6x over the past 15 years. If you look at the change in avg sale price, people you'll see a similar trend."

And let me remind you that 1996 was the exact low point (inflation adjusted) in the last 25 years of Case-Shiller data. So why is 1996 is a representative year? Using this as a baseline won't skew your conclusion?

We already factor depreciation into the buy vs rent formula. Now having just completed a gut renovation, I would argue at your suggested $250/sf every 20 years you end up with an improved property. I would argue even at $150/sf that is the case. But remember that monthly charges also factor in improvements that are made to the building itself over time, as (for co-ops) they are often funded by a mortgage, and the CC includes the interest going forward.

Nada, your ultimate conclusion is that improved incomes, better neighborhoods, lower crime and other quality of life factors have no impact on real prices and real demand as you imply. It just doesn't make any sense to me and I suspect you come these conclusions based on using skewed data.

"And let me remind you that 1996 was the exact low point (inflation adjusted) in the last 25 years of Case-Shiller data. So why is 1996 is a representative year? Using this as a baseline won't skew your conclusion?"

The low point was more like 1993, and I don't think low point vs. high point matters here. The 3.6x avg ppsf vs. 2.6x constant-quality difference since 1996 is simply pointing at a 2% improved quality annually in the avg sq ft sold.

If you'd like I can use 2006. Using the average of all of 2006, Miller Samuel avg ppsf for condos is up 6.5%. Using the average of 2006 from the SE index, we're down 6.3%. Overall, there's a 13.6% discrepancy. This amounts to a 2.6% annual improvement in the avg sq ft being sold.

Pretty much the same either way.

Here are some 25 year charts to look at.
It is multifamily apartment building specific but there is excellent data and how far away are multifamily performances to single units anyway.

http://www.masseyknakal.com/chairman/634043355622750061.pdf

"We already factor depreciation into the buy vs rent formula."

The SE and Case-Shiller indices are constant-quality indices. They are not deteriorated-quality indices, nor are they improved-quality indices. If you'd like, go read some papers or whitepapers to understand how they are constructed. It is these constant-quality indices that show only inflationary growth in home prices over the past century. Some nut did this going back 400 years on one block in Amsterdam. As with Shiller and US home prices, the guy found decades-long peaks and valleys to match the whims of irrationality-dujour. But inflation only at the end.

Deteriorated homes do not track inflation over the long run. There is economic reasoning for this, and there is ample data backing it up.

"Nada, your ultimate conclusion is that improved incomes, better neighborhoods, lower crime and other quality of life factors have no impact on real prices and real demand as you imply. It just doesn't make any sense to me and I suspect you come these conclusions based on using skewed data."

What part of the data don't you believe? That rents have tracked inflation nationally, or that they have done so in NYC?

Thanks, truthskr10. There's an independent set of data saying the same thing for you, maklo. I think you agree that real prices are up nearly 2x since 1998. You're in fantasy-land if you think real rents are up. The data says they're flat. Truthskr's personal datapoint 2000-2011 matches the data. My personal datapoints match the data.

The renters enjoyed all the same quality-of-life improvements you tout. Yet they are not paying more in real terms while owners are pay nearly 2x more.

Is it that you don't think there was ever a bubble in NYC, or that it's over?

"Is it that you don't think there was ever a bubble in NYC"

No and I think I have been pretty clear on this.

"or that it's over?"

I think we are getting pretty close to fair market prices. Market does not seem cheap, nor does it seem overly expensive to me.

What that says to me is, in the continuum of sale opportunities out there, you have a decent shot at finding one that fits you and maybe even turns out to be a good investment - if you are a better-than-average investor. From 2005-2007, I could not find anything that came close, and did not think it was worth the time and effort because the likelihood of uncovering one was low.

I think I have been pretty clear on this point. Is it over? If by this you mean the prices will flatline next year? I have no clue. Can it overshoot? Certainly. However, I don't feel confident in my ability to predict when the bubble ends, whether it overshoots, whether it flatlines in the next 3 years. I do feel more confident in predicting NYC's prospects, the prospects of my particular neighborhood and people's desire for housing over the next 10 years though.

Edit: I meant to say "Yes and I think I have been pretty clear on this."

"The low point was more like 1993, and I don't think low point vs. high point matters here. The 3.6x avg ppsf vs. 2.6x constant-quality difference since 1996 is simply pointing at a 2% improved quality annually in the avg sq ft sold."

Indexed to 2000, 1993 was 89. 1996 was 85.5.

The other flaw with your analysis is that you are comparing asset value to cost and pretending they are the same. An apples to apples comparison would factor in long-term interest rates as a proxy for equivalent housing cost. The 30-year secular decline in long-term interest rates and inflation certainly played a role here. I think about housing like a capitalized purchase of future housing costs and I think you do as well, so I don't know why you wouldn't apply that principle in this case.

Now you appear to believe that secular decline in interest rates is coming to an end. And we can focus on that as a separate topic. But to draw the conclusion that income, housing prices and rents don't move together in the long run from the data you presented flawed.

"Maklo, Sorry if you don't see anything to discuss. I thought you might be interested in explaining or understanding the bases of your prediction that RE prices will go up 2%/year, since that prediction is the core of your claim that current prices are reasonable."

FG - it's not that I don't see anything to discuss. I think we are discussing the wrong things.

I am assuming market prices are generally fair, or close to it today. And I am assuming real estate prices go up at inflation. I think that's slightly conservative for NYC, but that is a view based on more important underlying factors.

You (and Nada) think prices are expensive. Thus you factor in real price declines in 10 years.

I think we are more or less agreed on the framework, but the interesting thing at play here is the circularity. We are debating whether house prices are fairly or over-valued today, and using a framework where one of the inputs is the capital gain/loss in the future, which depends on what you think about prices today.

So the buy vs. rent framework is just one tool. But we cannot rely solely on it, because it is a circular equation.

Thus, we need to talk about real factors like views on real wages, future trends in quality of life factors, inflation, cost of capital ... and even the applicability and impact of these trends on real estate prices. You brought up Japan vs. US. I presented my view on that. Inonada talked about constant quality indices. I presented my view here. Ultimately there will probably be differing views on these and because much of it is future prognostication, no one really knows for sure. But at least we are talking about what really matters.

"Jan 1998 was at 83. I think he means to inflation-adjust that number. Inflation has been 1.34x, so 111.6. Which works out to a 30% drop.

Arguably better is the SE index, which was at 807 and is now at 1835. Doing the same there works out to a 40% drop."

Nada, so am I to suppose this all means you think prices are inflated by 30% to 40% in real terms from today? After all, you factored in a $300,000 nominal price drop in the buy vs. rent formula many posts ago. That's a 15% nominal price drop and if you factor in the 2% inflation rate (which we both agreed is reasonable) that's another 22% decrease as you deflate 2021 prices back to today's real prices.

Do you really think going back to 1996 real prices - i.e. the lowest of the last 25 years - is a reasonable "most likely" outcome in 2021?

You guys are approaching 300 posts.
Im pretty sure if you torture numbers long enough they'll tell you anything you want. :)

think about the lost productivity of finance's finest....

Maklo, Actually, the usual way to compare rent to buy is to compare a rental held for the economic life of the asset to current price. That eliminates the circularity that your method adds.

It also makes more sense if you are seeking to compare fundamental value (i.e., income producing capacity, or rent deferred) to current price. The asset should have the same fundamental value in the hands of one owner who holds it forever as in the hands of many owners who trade it back and forth -- particular owners may do better on trades, but only if some other owner does worse.

Once you have a handle on fundamental value -- the value coming from the fact that you can live in the unit and save rent on a comparable one -- then you will be better able to decide whether you are likely to be one of the traders who wins or one who loses.

Paying well over fundamental value doesn't mean that you won't make money on your trade. But it doesn't improve the odds either.

Maklo, rents reflect quality of life and income changes much more directly than sale prices, since they don't include as much guessing about future changes. That's another key reason to use conventional rent/buy analysis, rather than your improvised version.

Personally, I'm not sure I buy the quality of life argument in the slightest. I saw a great improvement in NYC quality of life from, say the peak of the last boom in 1988, to the bottom in 1996, by which time the crack epidemic was definitely in decline, but I'm hard-pressed to see the 3x improvement since 2000 -- was it the chain stores? the douchization of the yuppified neighborhoods? the last gasp of the middle class expiring on the UWS? the death of publishing? NYU's giving new meaning to public drunkenness on the Bowery? Meanwhile, RE prices increased even more in Las Vegas than in Manhattan, and I'm not sure that it's even grammatically permissible to use Las Vegas and quality of life in the same sentence.

QOL should be fully reflected in rents. If you think that renters have systematically different views about QOL in NY, or different income expectations, than owners, that's something worth talking about. Or if you think that NYC is likely to suddenly become a much better place to live than it is, and rents don't yet reflect that -- but we need a story about how that could happen without RE prices first dropping enough to allow interesting people to live here.

FG - I'm not going to rehash the buy vs rent methodology again. Go back and refer to past posts if you don't think I understand the concept of capitalizated rental equivalent costs. The difference is that you are assuming a 30-40% real capital loss and I am assuming the status quo. I would think the burden of proof is on you convince others that a further 30-40% real capital loss is reasonable.

To those of you who believe cap rates are too low, could they get in line with your desired historical levels if the rent go up 30% rather than prices coming down 30%. For Nada, the rental listings at the places I mentioned at $7-8 per sq feet per month. I am just using $6 including all the negotiations and discounts. $6 for high-end rental is low looking at the listings. Related does not discount much, check out there website. Very few vacancies.

Maklo, Sorry, I'm giving up. You correctly state that your reasoning is circular; you believe you are paying a fair price now based on the price you think you can get later, but the only reason you think you will get that price later is that you believe you are paying a fair price now. That's not a standard rent/buy analysis, and you clearly understand the reason why not.

I don't know why you think that the standard method makes the same mistake -- it doesn't. Neither Inonada nor I assumed a loss; we just plugged your numbers in to the standard method, and it implies that a loss is likely. We've each tried to see if you have some additional information or perspective that might make the standard analysis too simple. Apparently not. It's just bubble talk.

Mercer:

Absolutely. Indeed, historically, most RE bubbles deflate by nominal sale prices staying stable or declining very slowly until rents and costs catch up with prices. See the post-1987 corrections in NYC and CA for an example -- almost a third down in real terms, barely noticeable in nominal.

However, rents usually track inflation pretty closely over longer terms, because builders usually can build more if demand increases. So to expect higher rents soon, you pretty much have to believe that NYC is about to generate a lot of jobs or higher pay, faster than builders can build. To believe that rents will stay higher than inflation over 20 years, you need to believe that builders can't profitably build at current rents. Do you?

On the other hand, if you think that the market's estimate of 2% inflation is wildly off -- some people who comment here seem to think that they are living in the Weimar Republic -- then you might be quite justified in paying what otherwise would look like an unreasonable multiple of current rents in order to own before the great inflation. Assuming that, for some reason, you think that renters would continue to believe that NYC is a fine place to live in the middle of economic and governmental collapse -- perhaps you expect to be Will Smith's landlord of choice?

Rents can go up due to a lack of supply. All new rental buildings are well north of $5 even in non-prime areas like 42nd and 10th. No one is stupid enough to build to rent at $4 per sq feet. Archstone despite being owned by a bankrupt company average $5 per sq ft per month in NYC.