In the technologist’s dream world, buying a home would be easy. You would have all the information you need at your fingertips without having to get off the couch. Once your mind was made up, it would require little more than pushing a button to make an offer. The only thing that might test your patience would be waiting for the seller to respond. Preparing a home for sale would be more difficult to do from the couch, but it too would be much easier than today.

Reality has yet to catch up with the technologist’s dream, but that day is coming. By putting listings and a wealth of supporting information into people’s hands, home-buying portals such as Zillow have made it possible to browse homes in great detail from anywhere with a web connection. At the same time, other companies such as Ellie Mae have helped make the industry more efficient by streamlining mortgage applications, appraisals, documentation and so on. Thanks to such innovation, the time it takes to search for a home and buy or sell happens faster than ever.

Yet every rose has its thorn. Although making the process quicker and more efficient is good and welcome, it has helped to shrink the amount of housing inventory for sale. This is an unwelcome side effect at a time when buyers are engaged in fierce competition over a relatively small number of homes for sale by historical standards. Prices are pushed up as a result.

Consider how such an accelerated process can constrain inventory. To keep things simple, suppose that exactly one new home is listed every day and that the process to final sale takes exactly 100 days. If on each of 100 days exactly one home is sold, then there is a stable inventory of 100 homes on the market.

Now suppose the process is improved so that it only takes 99 days for a sale. On one of those days, two homes are sold — one that was on the market 100 days and one that was on it for 99. As a result, the inventory shrinks to 99 homes. A few days later, the process is improved again and shortened to 98 days and, once again, inventory shrinks by one. It is easy to see that as long as the improvements keep coming, inventory will continue to shrink.

Now let’s turn to actual data from the housing market. The inventory of homes for sale has indeed fallen. According to data from Zillow, it decreased nationally from a peak of more than 2.4 million homes in July 2011 to fewer than 1.4 million in July 2017.

Yet at the same time, the rate at which homes have been listed and sold has actually risen! Over the same period, the number of homes sold each month has increased from less than 300,000 to almost 500,000. How could inventory have fallen when the number of homes being listed and sold was growing? Because homes spent less time on the market. Even though homes flowed into the market at an increasing rate, the pace at which they were sold picked up even more, so the inventory of homes for sale at any given moment declined.

Of course, the time it typically takes to buy or sell a home depends on many factors besides innovation. For instance, when a buyers’ market becomes a sellers’ market, one would expect the time it takes to buy a home to rise and the time it takes to sell one to grow shorter. But that’s not what happened. Data from Redfin show that during the past several years, the time required to buy and sell a home have both decreased, consistent with innovation making the process more efficient. Does that mean innovation is the only factor at play? No. But it is at play nonetheless, reducing the time homes spend on the market and therefore, as I’ve illustrated, also reducing the inventory of homes for sale at any given moment.

Replenishing the inventory of homes for sale is the key to rebalancing the market. A crucial part of the solution is building more new homes. In addition to creating new inventory directly, new homes also spark a sequence of reactions known as vacancy chains, in which every buyer but the last also vacates their old home and sells it the next buyer. Because they enable several additional homes to be sold, new homes have an outsized impact on the volume of homes sold and on inventory.

There are also gains to be had from undoing perverse financial incentives that help keep many homes off the market even though their residents would rather move, especially in the most expensive areas where lack of affordability and low inventory are most acute. The $500,000 cap per couple on the exemption from capital gains tax when selling an owner-occupied property, for example, has many older residents in pricey coastal markets aging in place rather than subjecting their children’s future inheritance to a tax hit. Were it less onerous for the elderly to move, their exit from homeownership would spark vacancy chains, too. By capping homeowners’ property tax growth well below the rate of housing price appreciation, California’s infamous Proposition 13 creates a similar disincentive for moving. So does the grandfathering of higher mortgage-interest deduction caps in the recent federal tax reform bill.

Last but not least, it’s important to realize that the swing of the pendulum from a low-inventory sellers’ market to a high-inventory buyers’ one has elements of a self-fulfilling prophecy. About two-thirds of buyers are selling one home and buying another in a short space of time. Today, these potential repeat buyers are reluctant to sell because they fear the risks of buying in a sellers’ market. However, if they were to catch wind of the notion that the sellers’ market is reversing, that might prompt them to put their homes on the market. That would, in turn, help make the notion a reality. If a buyers’ market were to follow, an innovation-induced reduction in inventory would all of a sudden be welcome.