[link|http://tycho.usno.navy.mil/leapsec.html|http://tycho.usno.navy.mil/leapsec.html]
The Earth is constantly undergoing a deceleration caused by the braking action of the tides. Through the use of ancient observations of eclipses, it is possible to determine the average deceleration of the Earth to be roughly 1.4 milliseconds per day per century. This deceleration causes the Earth's rotational time to slow with respect to the atomic clock time. Thus, the definition of the ephemeris second embodied in Newcomb's motion of the Sun was implicitly equal to the average mean solar second over the eighteenth and nineteenth centuries. Modern studies have indicated that the epoch at which the mean solar day was exactly 86,400 SI seconds was approximately 1820. This is also the approximate mean epoch of the observations analyzed by Newcomb, ranging in date from 1750 to 1892, that resulted in the definition of the mean solar day on the scale of Ephemeris Time. Before then, the mean solar day was shorter than 86,400 seconds and since then it has been longer than 86,400 seconds.We wouldn't have leap seconds if the earth weren't slowing. The mean solar day is increasing in length, so to have a uniform clock we have to periodically account for the accumulated difference. I didn't say "the earth is slowing a second every year". But the increased length of the day is definitely where leap seconds come from.
The length of the mean solar day has increased by roughly 2 milliseconds since it was exactly 86,400 seconds of atomic time about 1.79 centuries ago (i.e. the 179 year difference between 1999 and 1820). That is, the length of the mean solar day is at present about 86,400.002 seconds instead of exactly 86,400 seconds. Over the course of one year, the difference accumulates to almost one second, which is compensated by the insertion of a leap second into the scale of UTC with a current regularity of a little less than once per year. Other factors also affect the Earth, some in unpredictable ways, so that it is necessary to monitor the Earth's rotation continuously.
In order to keep the cumulative difference in UT1-UTC less than 0.9 seconds, a leap second is added to the atomic time to decrease the difference between the two. This leap second can be either positive or negative depending on the Earth's rotation. Since the first leap second in 1972, all leap seconds have been positive and there were 23 leap seconds in the 34 years to January, 2006. This pattern reflects the general slowing trend of the Earth due to tidal braking.
Confusion sometimes arises over the misconception that the regular insertion of leap seconds every few years indicates that the Earth should stop rotating within a few millennia. The confusion arises because some mistake leap seconds for a measure of the rate at which the Earth is slowing. The 1 second increments are, however, indications of the accumulated difference in time between the two systems. (Also, it is important to note that the current difference in the length of day from 86,400 seconds is the accumulation over nearly two centuries, not just the previous year.) As an example, the situation is similar to what would happen if a person owned a watch that lost 2 seconds per day. If it were set to a perfect clock today, the watch would be found to be slow by 2 seconds tomorrow. At the end of a month, the watch will be roughly a minute in error (30 days of 2 second error accumulated each day). The person would then find it convenient to reset the watch by one minute to have the correct time again.
The direction of the leap second comes from the variation in spin. Over the long run, though, they tend positive, and the current 2 second difference is entirely due to the slowing of the earth's spin.