2018 Economic & Housing Predictions


Overview of 2017

We started the year with a new President and a lot of questions and uncertainty, but 2017, in terms of economics, turned out to be about as dull as you can get. The sharp dollar oil crash in 2015 and 2016 provided a fertile environment for oil earnings and manufacturing orders to rise. World economic growth hit a decade high as globalism prospered. In the U.S., the housing market continued on it’s slow and steady pace higher, while low inventory had little impact on sales as total home sales topped  6,000,0000. Home prices in 2017 increased with inventory staying below 6 months.  This maintained the 21-year trend of inventory remaining below 6 months except during periods when economic distress led to forced selling.  For the 36th year, the scare of stronger inflation and much higher rates never came to fruition. Even with 2-plus horrible hurricanes, crazy wildfires and “rocket-man Kimmy” lobbing missiles, the VIX stayed calm, gold didn’t go parabolic, and the longest job expansion in history continued.

For those interested in how well I did predict economic events for 2017 – find my 2017 Prediction article here:


Those are the dead we bury, and now it’s time to look to what 2018 brings us.

Predictions for 2018

1.  Inflation, 10-year Yield and Mortgage Rates

For the Federal Reserve, the question for 2017 was if it was going to be two or three rate hikes.  In February of 2017, I wrote that the Fed needed more inflation to warrant a 3rd rate hike because their 2% target rate for PCE inflation would not be sustained through 2017.


One of the main reasons why inflation looked weak in 2017 was that rent inflation slowed and rent is the most significant component of CPI inflation. Medical cost inflation also slowed dramatically, but this doesn’t make up as much of the CPI.

From BLS:


Medical inflaiton 2018

Due to the factors mentioned above, core inflation (both CPI and PCE) fell in 2017 while headline inflation held up because oil prices, both Brent and Light Crude were up over double digits.

From Doug Short.


Inflation won’t go parabolic unless your country’s currency is getting clobbered in the global market. Since the U.S. dollar is still elevated after its first big move before the first-rate hike, I don’t expect inflation to be a pressing issue in 2018 because core inflation  (PCE and CPI) won’t be over 2% for a significant period if at all.


US dollar

Even if core inflation heads higher in 2018, since we had such low inflation in 2017, I don’t see a small rise in core inflation is a problem. Wage growth has been slow as well, so the Fed won’t need to take aggressive action with rate hikes.

What does this mean for the bond market and mortgage rates?

Each year for the last two decades, we have heard the same old story that this year, mortgage rates would increase. Unless you were listening to me, that is.  If you have followed my work, you know that for the last few years I predicted that the 10-year yield would stay on the channel of 1.60% – 3%.  For the most part, this has occurred with the range staying between 1.56% -2.62%.  The 10-year yield in that range means mortgage rates would be between 3.625% – 4.375% — and this has been mortgage rate range for this cycle. The only times that rates broke out of this range was during short-term economic events, such as the 2012 European bond market scare and the 2016 Brexit, when the 10-year broke under 1.60% and during the tail-end of the taper sell-off in bonds when yields broke over 2.62%. So for 2018, I maintain that the 10-year will stay in the Mohtashami channel of 1.60% to 3%, which means we could see high-end risk mortgage rates between 4.5% – 4.625%. If mortgages rates do go above 4.75%, I believe this would be very short-lived. The 10-year yield did fall in 2017, but it didn’t break below 2%. The 10-year yield came close at 2.02% but held the line. I am also looking for yields to invert in 2018. It really wouldn’t take much for that to happen.  A minor stock market sell-off and oil prices falling would send money into the bond market leading the 10-year yield to break under 2% and slapping high fives with the 2-year yield. The best thesis I can make for higher rates is if the oil has another strong year with higher prices, core inflation jumps significantly,  wage inflation really picks up, and global manufacturing data has another spectacular year.

(A must follow on twitter)

US Yield Curve ends the year at its flattest level of the expansion: 51 bps spread b/t 10-yr and 2-yr yields.

2018 Inversion

2. Home Prices and Inventory 

Home prices have legs to go higher in 2018 because inventory for existing homes can’t get over 6 months. I genuinely believe the topic of inventory doesn’t get enough attention due to its complexity.  In simplistic terms, higher home prices limit monthly inventory. Post-1996, homes prices have deviated from historical norms making over 6-months supply an unusual event outside the housing bust years.

From Doug Short:

JAN 1996

We can see the impact of home prices on monthly supply for both existing and new homes, post-1996.

Existing monthly supply

From Bloomberg :
Monthly supply

New home supply

From Fred:

Decemeber New home supply

The only times we have had more than 6 months plus inventory was during the housing bust years. This period could be characterized by over-investment in housing before the bust,  worsening demographics for home sales after 2007, widespread speculation demand during the housing bubble years (2003-2006), opportunities to take on debt beyond the ability to pay and mega cash out the boom. This was the perfect storm for future rising supply.  Today, we don’t have enough home sales from mortgage buyers to create a supply of homes that succumb to distress selling,  like we saw in 2006-2011. The loan profile in this cycle looks fantastic, and we have no more exotic debt in the system after 2012 either.

With the housing bust, a massive supply of homes was added into the marketplace due to forced selling into a weaker demographics patch. On another note, the supply never surpassed over 5 months from 1999-2005, and yet we had more sales from mortgage buyers then. (Still, think low inventory prevents transactions?)

From the NAR:


In 2017, the tenure for housing (how long homeowners stay in their homes) hit an all-time high. This is likely due to a lack of selling equity by homeowners, smaller family size,  or simply not wanting to move up or down. We have been building bigger and bigger homes since 1975, and the need to move up has become less of a need for families.

From the NAR:

In 2017 we heard some analysts use terms like: “Inventory Crisis,” ” No Homes To Buy”   and “Record-Breaking Demand” to describe the housing market, but while these terms make cute headlines, they are not based in fact.

If we had a massive inventory crisis and record-breaking demand in 2017, then we would expect certain things to also be present. For example, we would expect the YoY real home price inflation to be much stronger.

From Doug Short:

Jan YoY Real

We would also expect purchase application data to be much stronger — not stuck at 1998 levels with mild  YoY growth. In 2016 we had 25% plus an increase in the heat months, while in 2017 we had negative 1% – 10% growth during the heat months which suggests that demand was just slightly up YoY from mortgages buyers. Demand from cash buyers was better than expected and grew in some of the reports year over year.

From Calculated Risk:
MBADec202017 2018

Annual months of housing inventory got to 4.3 months in 2004.   We are testing these levels now for existing homes.  For 2018, I don’t expect to see too much movement up or down in inventory. I don’t wait to see any forced selling in this cycle. This will not happen until we see a job loss recession, which won’t happen in 2018.  Investors who bought homes to rent are still making good returns in this environment of high rents and low-interest rates.

For this reason, don’t expect these investors to dump many properties into the market in 2018.  The notion that all investors would simultaneously drop properties, causing an increase in supply and lower home prices makes no sense, especially when one considers that most of these properties are tied up in rental contracts.  Likewise, baby-boomer homeowners won’t be dropping their homes on the market in any scale either.  Some are using their homes for income, using reverse mortgage financial vehicles, and others lack the selling equity or desire to move.  Mobility among Americans has dropped since 2000.

A must-follow on twitter @Graykimbrough

If we do see some YoY increases in inventory in 2018 and no increase in sales I wouldn’t worry too much about that. It is rare to see a monthly supply rise and existing home sales rise post-1996.  Total units available might be a better data point to track than monthly supply.

3.  Housing Starts & New Home Sales 

For the last many years, I have stated that new home sales and housing starts will have a slow but steady growth.  Both metrics are still shallow in historical terms. Early in this cycle, there was much confusion among housing analysts, some of which made outlandish forecasts like housing starts would hit the 50-year moving average while new home sales were going to below. The builder’s stocks were struggling in 2013-2015.  Since the end of 2015, after a severe market correction in these stocks, the housing community took a more realistic outlook on the sales potential of these companies. In 2017 the builders stock performed well even though total new home sales won’t be high on a historical basis.

From Doug Short:


For 2018, I expect housing starts will continue a path of slow growth.  This growth will come from single-family residence construction since the multifamily boom we saw early in this cycle has cooled significantly. We didn’t see much increase in total housing starts in 2017 primarily due to the cooldown in multifamily construction, but single-family construction looked solid and just right to me.

From Calculated Risk:

StartsShortNov2017 (1)

In 2018 the bulk of the total growth now needed to come from single-family construction.  Some have suggested we don’t have the necessary labor to build  these homes, but before you buy into that thesis, make sure they read this article by Freddie Mac economist, Len Kieffer:


As for new home sales, I will continue to contend that to have more unit sales in this economy, we will need to have smaller (i.e., more affordable) homes available.  This has been a theme of mine since 2014.  It appears that the builders now finally are making some strides to halt this multi-decade run of building bigger and bigger homes. I believe the major miss in sales estimates in 2014 shook up the industry for the better. However, as we can see below, new homes are still more significant and more expensive than existing homes.

New home size continues to trend lower (slowly) as more entry-level homes are built.

This was what the trend looked like before the builders began to build smaller homes.
New Home Size 1

In 2017 I predicted a 4%-7% growth in sales with a possible upside if the median sales price didn’t increase much. I stand by that prediction for 2018. For this new year, I expect to see 2%-5% growth in new home sales that could go higher if the median sales price remains stable, and the trend of building smaller homes continues. Housing starts and new home sales are still very low in historical context, so there is room for growth —  especially when one considers that we now have over 154,000,000 people working, mortgage rates below 5% since early 2011, and we are in the midst of the longest job expansion ever and possibly the longest economic expansion ever.

From Doug Short:


4. Existing-Home Sales 

For 2017 I predicted no growth in existing home sales at all. I expected sales to be in the range of 5.15 – 5.45 million because we didn’t have the demographic demand, and the would-be move-up buyers lacked selling equity. I was not a believer in the theory that the lack of homes on the market kept sales down in 2017, although this theory was widely repeated in 2017. In fact, the three highest existing home sales prints of this cycle occurred at the same time as the lowest monthly inventory prints. The last existing home sales print for November was 5,810,000.  This number may be revised a tad smaller but, in general, all sales were good early and late in 2017.

For 2018, I anticipate existing home sales to be in the range of  5.27 – 5.53 million units.  If we end the year showing negative growth, with rising inventory once again, don’t worry, be happy. This would be “normal” especially when purchase applications are still trying to party like it is 1999.   In years 2020-2024 we will have a much better demographic profile than we had from 2008-2019 for mortgage buyers.

I will admit that in my 2017 Prediction article, I got one thing badly wrong. I thought cash buyers would fall again as a percentage of sales.  Cash sales instead were flat to higher in 2017. There was only one month that cash buyers made up 16%-19% of sales, which was my target range. For the other 11 months, the prints for this metric came in higher than I anticipated, in at the 20%-22% range.

Despite my previous blunder, for 2018, I am going to stick with the prediction that cash buyers will fall as a percentage of total sales.  I anticipate cash buyers to make up 16%-19% of sales. If this thesis fails again, then this means we have some added demand for existing home sales. This could also lead to another year of beating sales estimates. Historically cash buyers have made up approximately 10% of the market place.  It seems unlikely that we will continue to maintain 20% plus cash buyers in the coming years unless we see a job loss recession which would add some more distress supply. However, the key thing to remember as cash buyers fall is that mortgage buyers need to grow to replace them for total demand to be higher. This is why it is essential to see in years 2020-2024 that the purchase applications finally break out of 1990 levels and join the 21st century.

From Doug Short:


Our best existing home sales in 2017 occurred when inventory was at the lowest,  in January and in November. When stock rose in the spring,  the seasonal demand months, sales didn’t rise much. It would be interesting to see if this same scenario replayed itself in 2018 or if we see year over year increases in inventory without many years over year growth.


 Unsold inventory is at a 3.4-month supply at the current sales pace, which is down from 4.0 months a year ago. 

DRf6A5iX0AAAdf7 (1)

5. Jobs and the Economy 


Regarding the job market,  since 2014 ended,  I predicted that monthly job creation numbers would decline and job openings would increase. Since 2014 ended,  we have experienced a decline in the rate of growth of total jobs created, but we are still well above the number needed for population growth. In 2017 I predicted that 170K- 140K jobs would be created monthly, and we achieved just a bit over that high-end number with one more report left to be released.  For some analysts,  this was hard to understand since we were told that 96,000,000 Americans are looking for work.  One would think that with over 6,000,000 job openings and the longest job expansion on record, with over 154,000,000 Americans working, that those 96 million who were sitting on the sidelines would find jobs and join the employed.  In reality, no 96 million Americans are looking for work. The unemployed in this country is mostly made up of retired people, stay at home moms, teenagers & young adults who are in school and those with addiction issues. For 2018 I am predicting monthly job creation numbers in the 157,000 – 129,000 range, which is still well above population growth, but lower than last year. With over 6,000,000 job openings in play, that number should be attainable.

Business investment: 

Growth in business investment started to come back in 2016 after a few quarters of decline. This wasn’t just about energy as we can see in the chart below. With the new tax plan in motion,  we “might see” more companies boosting their investments in 2018. I do not believe that domestic investment will result in 4% total GDP growth, but for 2018, this sector has room to grow, especially combined with the energy sector.

US CAPEX not entirely driven by energy sector Overall business fixed investment ex-energy continued to expand at about a 4% pace in Q3


Note: Oil prices and new manufacturing orders run hand in hand. Another must follow on twitter

This may come as a surprise to you, but US manufacturing activity is usually positively (!) correlated with oil prices

Oil prices and Manu Dec 2017


In 2017, we had some critical PMI data all over the world, but especially here in the U.S. Even with strong exports in oil, our trade deficit remains. For 2018, I don’t see much change in the U.S. trade deficit. The dollar is too strong to create a boom in exports, and we continue to enjoy the cheaper priced goods from other countries.

From Doug Short:


Consumer spending:

What a Christmas so far! Look for some eggnog hangover from the strong Christmas sales.  I wish I could say that because so many people bought bitcoin in 2017, we could expect a boom in retail sales in 2018, but that would be a lie. Retail sales should be excellent in 2018 but don’t expect too much of a growth. Car sales trended lower in 2017 until the hurricanes, which led to a build-up in inventory and increased demand. Let’s hope for 2018 we don’t get a repeat of those events.  I’m curious to see what people do with the extra money they get with the tax cut.

From Calculated Risk:



One note on wages. We have a massive older workforce in this century of people ages 55 and up. One thing about wage growth is that when you break down the Fed wage tracker, it looks better for ages 25-54 (prime-age labor force age group) than ages 55 and above. This has always been the case and is one factor why total median wage growth is below the prime-age labor force wage growth.

Wage Growth per age group:

2.0%   Ages 55 +
3.6%   Ages 25-54
7.1%   Ages 16-24
3.3%   Overall

Fed Wage Tracker:

Jan Fed wage tracker


The rate of growth for population and productivity has been falling for decades, even while being at cumulative highs.  This is why we have been seeing lower GDP growth.  This is also why I have predicted that GDP would have a 2 handle in the past few years. The notion that the U.S. could have 4%-6% sustained growth is unrealistic. Because Q1 always comes in soft, it makes it even more difficult to have achieved 3% total growth for a year.

A must follow is @Ernietedeschi . This is probably my favorite chart of all time.


For 2017, GDP will end up being in the range of 2.5% -2.8%.  Q4 is tracking at 2.8% per the last Atlanta fed data point. Remember that GDP data is consistently revised up and down. Some are saying that the previous three-quarters had the highest growth of the cycle, but this isn’t true.  We had a 3.9% GDP growth in Q2, Q3, and Q4 combined in 2014, but Q1 in 2014 was very weak.  That Q1 seasonality impacts total growth, so if we don’t see 3%-4% growth in 2018, don’t worry about it. 2018 looks fine because domestic investment, productivity, inventory rebuild, and consumption look fine. I am predicting GDP to be in the 2.3% -2.6% range. I am not sure how quickly we can get that infrastructure plan up and go, which would be a substantial domestic investment which could increase GDP.

From Doug Short: 
Q3 GDP revision Jan

Recession Watch? 

I have taken much joy smashing the Great American Collapse crowd. The PMI data looked spectacular in 2017. Even with substantial growth in manufacturing, the stock market at all-time highs, all 4 major confidence indexes at cycle highs and retail sales at all-time-highs, GDP growth wasn’t 3%-4%. Think about that for a second.

From Doug Short:

Jan Manufacting Boom!
Nevertheless, if you want to know what to look for to predict an upcoming recession – here are your clues:

1. Yield inversion, when the long end and short end invert, that means we are heading into a recession sometime soon. I believe we will see an inversion in 2018, but it won’t mean a recession soon. It would be the longest time before the first inversion to a recession ever in history if we inverted in 2018. Not likely a 2018 recession thesis.

2. Unemployment claims have to spike higher. Even with two major hurricanes in 2017, unemployment claims have remained under 300K for a long time– even when manufacturing went into recession. You need to see claims rise aggressively before the “R-word” is uttered.  That means you, Warren Mosler.

From Doug Short:


3. Leading economic indicators would have to confirm the rise in unemployment claims.  We have never seen a recession without leading economic indicators falling 4-6 months in a calendar year.

From Doug Short:


4. The Fed would have to start fighting an over-heated economy. In 2017 I anticipated that there would be two rate hikes, but we got three. For 2018, I am awaiting two more rate hikes unless core PCE get over 2%. If that happens, then we can expect a third and maybe even a fourth hike. The main point is that the fed isn’t fighting inflation because we don’t see it in the core PCE or CPI data and asset inflation isn’t part of the Fed’s dual mandate yet.

From Doug Short:


Things that make you go hummmmmm:

I am curious to see if we experience a down-trend in the rate of growth of consumer spending following the strong Christmas sales.

I am curious to see if the car sales trend lower this year as they appeared to start to do before the 2017 hurricanes.

I am curious to see if an infrastructure bill will be passed in a bipartisan fashion.

I am curious to see if  Paul Ryan will cut entitlements. Republicans tend to talk a lot about reforming entitlements, but this never results in any meaningful change. I find it hard to believe that with the specter of the mid-term elections on the horizon, that anything of significance will be done to reform entitlements, especially considering that they couldn’t even repeal and replace ACA.

I am curious to see if the bond vigilantes, (wait.. I can’t stop laughing… ok), crawl out of their holes where they have slept for the last 36 years. I doubt it but its fun to speculate.

I am curious to see if Harry Dent and Peter Schiff will stop going on info-wars. (Snicker) 

Finally, I know we have a great divide in this country. I was a voting Republican up until 2017 when I became an independent. Despite our divisions, we shouldn’t be divided on our opinion that this country isn’t a failure. We can’t have the biggest economy in the world with GDP per-capita numbers like we have, with over 154,000,000 people working,  with over 88% of the prime-age labor force working full-time jobs, with over 6,000,000 jobs openings available, if we are a failed economy and a failed country.  To some on social media, I am known as the American bear killer. But the truth is I realize that economic cycles come and go as do recessions and recoveries. I also recognize that America has a real poverty problem and an affordable housing problem for low-income Americans. However, make no mistake I refuse to show mercy or pity for those who consistently bash this country with false economic narratives and death calls its collapse. This will always be a personal war against the American bears till the end do us part!

I leave you with my photo of the year from CNBC. I call it the Dr. Grumpy Doom and Gloom. Do yourself a favor and don’t be this guy!

Happy New Year Everyone!


Logan Mohtashami is a financial writer and blogger covering the U.S. economy with a specialization in the housing marketLogan Mohtashami is a senior loan officer at AMC Lending Group,  which has been providing mortgage services for California residents since 1987. Logan also tracks all economic data  daily on his own facebook page https://www.facebook.com/Logan.Mohtashami

12 thoughts

  1. based on the chart “Home Prices and Consumer Price Index” from dshort, U.S. housing prices tracked the consumer price index from 1960 to 2000 very closely.

    during the 2000-2006 housing bubble, prices rose significantly above that long-term trend line and then reverted back to it during the 2006-2012 housing crash.

    in 2017, we can observe that housing prices have once again risen significantly above the consumer price index

    during the next housing bust, do you expect the same kind of reversion to the mean to occur?

    I look forward to your thoughts

    thank you

    Robert Campbell San Diego, CA

    1. CPI inflation home price index has gone off line from historical trends. However, real home prices on a national basis aren’t back to 2006 levels and it will be hard to get their any time soon. Real home prices and real affordability are actually better now than then because real home prices are back more to 2004 levels and mortgage rates are 2% lower than trend. If mortgage rates go above 6% then the real affordability metric gets hit greatly. If real home prices also go back to 2006 levels with 6% mortgage rates then that adds another layer of damage to affordability. Post 1981, ever single housing cycle has had 2% plus lower interest rates to help demand. However, for that to continue in the next cycle that means we would need to see 1.25% – 2.25% 30 year fixed rates to happen which means the 10 year yield would need to go negative. So, even if mortgage rates stay the same, in the next housing cycle it would be a net increase just due to the fact that mortgage rates didn’t decline 2% plus. To answer the reversion to mean question, my fear is actually that real home prices can get stronger when mortgage demand gets back to the 21st century level making affordability worse because we won’t have the mass crash in home prices like we did from 2006-2011.

  2. Thank you for your response however the chart I was referring to showed the long-term correlation between the consumer price index and U.S. housing prices in dollar terms – and not in term of U.S. dollars that have been adjusted for inflation.

    On an inflation-adjusted basis, U.S. housing prices (as per Robert Shiller’s 126 year chart of “Real U.S. Housing Prices” that I’m sure you are familiar with) have been consistently reverting back to a slightly upward sloping trend line during the past 5 real estate downturns since the 1960s.

    So asked differently, do you think real U.S. housing prices will revert back to the mean like they have repeatly done during the past five boom-bust cycles – or will they stay elevated this time around?

    I once again look forward to your thoughts.

    Robert Campbell
    San Diego, CA

    1. No on both nominal and inflation adjusted prices to 1996 trend terms over the next decade. The demographics for housing is going to get better in the next decade which will leave us with replacement buyers at worst. The combination of factors that would create homes prices to go back to that trend would be a job loss recession with a rising rate enviornment back to 6%-8% mortgage rates. I just can’t see inflation picking up to warrant those rates without wages going a long with them. The next economic recession will have a lot less distressed sales than the housing bust years so the downward pressure in home prices once inventory gets over 6 months will be limited to a degree as we won’t have the factor of prime age labor force peaking like we did in 2007. The next decade will have better household formation data just because of the new supply of people ages 30-39.

  3. I appreciate your comments – and it will interesting to see how your forecast plays out.

    In the short-run (i.e. 4-8 years) and based on my 50+ years of experience in the real estate business, it has been my experience that the movement in real estate valuations are far more dependent on sentiment and psychological factors than they are economic fundamentals.

    I know that’s been the case in California – which is where I live.

    Robert Campbell
    San Diego, CA

    1. CA is very unique because it has a price inflation model that kicks out 82% of the working population from buying homes once you take out cash buyers and those making 3X median income. So we bring a lot heat in housing inflation here more than any other states because the people living near water that own homes do make a lot money. Nationally speaking we had about 6 states that created a lot distress inventory in the housing bust and now those states are clean from exotic loans post 2012. 20%-30% of the homes in this cycle were bought with cash with existing home sales running below 5.7 million this entire cycle. We don’t really have the demand curve is scale terms yet from mortgage buyers to create a mass supply of distress homes once the recession hits. It will fall on late cycle lending to those with low down payments that will be the biggest at risk owners for default, FHA loans.

    1. If you model out median income to median prices, some would say that 70% of California residents are priced out housing. I have that number at 82%, I brought that up at the BNY Mellon Stock conference in 2014 when doing an interview with Bloomberg Financial. I take a different approach because a lot affordability indexes and based on a 20% down payment model with not much debt payments on the owners side. So they have an unrealistic debt to income ratio starting point. Adjust it to taking off those making 3X median income owner and have a more realistic take that many first time home buyers don’t have 20% down to buy a home, then you can bring that number to 82% of the working population is priced out of housing. Obviously if you head east home prices get cheaper so that factor has to be taken into consideration. Part of the issue with CA housing is that it does have a solid educated dual income household that typically buys homes. However, that household is some what limited because it applies only to certain jobs. Hopefully this gives you an idea of what it is.

  4. Hello Logan. I have inherited a home in Santa Monica in the “North of Montana Avenue” area, in which a starter home or tear-down sells for $3M. Although I cant predict the market, in terms of selling at the high end, I presume from your sentiments (on the market) that there will PROBABLY be another year or two before housing prices will falter?

    1. I grew up on in Santa Monica 5th street and Montana. For prices to fall inventory needs to rise, so you can track that data to see when pricing shifts on a monthly basis. This won’t happen over night so you will have a lot time to see when the inventory channels will change to negative. That is the best advice I can give you in terms of that question rather than a open blanket statement that has many variables in it over time. Track monthly supply once you start seeing rising inventory and lower demand that is when pricing will change.

Comments are closed.