Commercial real estate is vulnerable to high interest rates, particularly after an era of cheap money. And like countless others have said before, the one thing that’s separating the office sector, making it the most at risk, is demand.

Amid the continued prevalence of remote work, the office sector is currently undergoing an economic reckoning. Capital Economics, a London-based provider of independent macroeconomic analysis, forecasts, and consultancy, predicts a substantial decline in office values from peak to trough, amounting to 35%. Furthermore, this decline is unlikely to recover even by 2040, in the firm’s view.

To gain deeper insights into Capital Economics’ perspective on the office sector outlook, Fortune reached out to Kiran Raichura, the organization’s deputy chief property economist. Coincidentally, Raichura had just concluded a conference in Chicago, where he discussed the very same topic. According to Raichura, the office sector is experiencing a crash due to the two aforementioned reasons. However, he emphasized that the downturn in the office space is distinct from previous ones due to the presence of structural shifts.

Fortune held a conversation with Kiran Raichura, and the following transcript reflects key aspects of their exchange. Portions of the Q&A have been edited and condensed for the sake of clarity and brevity.

Fortune: I want to start by asking you about your methodology behind that 35% plunge in office values that you’re forecasting by the end of 2025, can you tell me more about that?

Kiran Raichura: The way we think about the outlook for real estate, generally, is that we look at both sides—meaning that we look at the occupancy market outlook and the pricing of the income stream from property. So the first thing that’s affecting all the real estate sectors across the board is the increase in interest rates that we’ve seen in the last couple of years. Generally speaking, we price commercial real estate against a risk free rate. For that, we can think about the 10-year Treasury Yield, from the end of 2021, that was 1.5%, it’s now nearly 4%. And so that increase in the risk free rate means we need to see an increase in commercial real estate capitalization rates. That’s affecting all the sectors across the board to a similar extent.

What is different for offices is that there’s been this very well documented reduction in the use of physical office space. The average physical occupancy in the U.S. is running at about 50% of its pre-pandemic levels, and that essentially has fed through to firms reducing the amount of office space they are choosing to have when their lease expires. So we’re generally seeing a reduction of space across the board, and that’s fed through at the market level to an increase in vacant space. The higher the vacancy rate, the weaker the rental growth.

Vacancy has already risen pretty sharply to around five percentage points, and we think there’s a lot more to come still, as other tenants leave their space in the next few years. And so that drives our rental forecasts to look pretty negative. Essentially, the 35% drop is a combination of the rise in rates and the increased risk premium for people’s concerns about the office sector itself—and the drop in net operating incomes, which is a combination of rising vacancy and falling rents because of remote and hybrid work.

How has the office forecast changed from your original outlook?

It definitely wasn’t always 35%. It’s been downgraded quite substantially in the last few months. On the back of the banking crisis in March, we took a big red pen to all of our commercial real estate forecasts in the U.S. because it had a really substantial impact on both the availability and the cost of debt finance for real estate investment. And it’s also had an impact in terms of investors’ confidence in the outlook for real estate. So I think arguably, maybe we should have already been expecting some of those price reductions that we are now forecasting.

Now, actually, there seems to be a lot more clarity in terms of what’s happening in the office sector with investors giving back assets to their banks, where they don’t make economic sense to put more money into or because they’re old and don’t have any tenants. So yes, essentially the realization that the market needs to reprice substantially has driven us to make a big downgrade in our forecast as well.

Why do you think office values are unlikely to recover, or regain their pre-pandemic peaks, even by 2040?

There’s a couple of reasons why we don’t think they’re going to recover. In terms of the demand for office space, we think it’s a structural change in that demand. In a normal recession, you get job losses, firms go bust, and they vacate office space. There’s an adjustment and then you get this growth back in the economy and things come back again. This time around there is a structural drop, so that firms, even if they maintain their same size, they’re reducing the amount of space they want for any given amount of employment. It’s going to have a downward impact, and it’s not going to bounce back in the same way it would normally

In terms of interest rates, there’s a fairly structural shift in our view. They reached such lows before and are now at nearly 4%. But our forecast is that they stay roughly between 3% and 3.5% over the next five years, so the level of interest rates will be much higher than it was in the last few years. And that means that the level of real estate cap rates need to be higher, which means prices will tend to be lower. We also don’t think they’ll bounce back with any support from a reduction in the interest rate environment either.

One way you could think about it is following the onset of the GFC [Great Financial Crisis], essentially from the end of 2007, you had again a really large drop in office values, and it wasn’t until the end of 2019 that they reached that peak again. That was a 12 year period, but it wasn’t a structural shift, it was very much a cyclical drop. So we think this time, it’s very easy to see it taking much longer to recover.

Is the office sector crashing? And if so, is there still more downturn to play out?

We definitely think the office sector is crashing. We’re seeing that evidence in the data, and also hearing it from our clients. It’s a very tough time to be an office landlord, and we know that banks are withdrawing debt from the sector as well. And the big drivers of this being the increase in interest rates that’s affecting all sectors, and the widespread adoption of remote and hybrid working that’s having a major impact on the demand for office space. And so that’s going to weigh on land landlords’ income streams from that sector. And it also creates a lot of uncertainty, which investors are trying to price in, and generally has a detrimental impact on prices.

We’re forecasting a 35% peak-to-trough fall. The valuation indices so far show around a 15% fall in values, so we’re not actually even halfway yet in terms of that being reflected in valuations. In terms of the deals that are happening, some of those bigger discounts are already taking place. In west coast cities, like San Francisco, which is the market that we think is going to be the worst in terms of values, assets that were valued pre-pandemic at a certain price have now been traded at a 60% to 70% discount, or even more to that previous value.

Do you think office values will ever fully recover?

I think they will. You can think about it as just the asset itself, and in some cases, it will be a case of changing its use. That wouldn’t necessarily reflect office numbers, but for that landlord or investor, they could certainly extract some decent value out of it and then see that go reasonably well in the future. On the downside, the risk is in some of the hardest hit cities where they might not come back to the same extent. Essentially the worst hit assets, some of those will become fairly obsolete and will be either knocked down or converted to other uses. But certainly the best quality office space will come back and will probably end up doing very well.

We’ve largely spoken about your national office outlook, what can you tell me about markets on a metro-level?

I spoke earlier about the two factors driving prices down: the impact of higher interest rates and the impact of the structural change on the office sector. With interest rates, again, the effect is across the board, and the structural change in demand is also across the board but to varying degrees. There’s a handful of different factors that are driving different levels of office utilization. So think about things like commute times and costs. Longer commutes mean that people are less inclined to work in the office. Housing costs also make a difference because they’re impacting where people actually end up wanting to live, which is obviously a negative for somewhere like San Francisco where it’s very expensive.

Those cities that we think are going to be the losers, essentially are those that are in the west coast or near the west coast. Then the relative winners are a lot of those places in the south and southeast in Texas and Florida, where there’s less of a downward impact from those specific factors. But also there’s some pretty strong cyclical growth in office-type employment in those cities. They’ve been attracting new workers, they’ve been attracting new firms in the last few years, partly because of migration patterns. And so that stands as a positive factor for some of the southern and southeastern markets, so that kind of drives a difference in our forecasts. Essentially, we think that the change in vacancy is going to be smaller in the southern markets, and therefore the impact on rents and net operating incomes much smaller.