Timing your retirement right can sometimes make quite a
difference.
When it is only a matter of perhaps some months the
overall impact is normally relatively small but issues
such as the beginning of a new tax year (with new
personal allowances), beginning and ends of Pension
Input Periods (PIPs) and other timerelated matters
nevertheless deserve to be taken into account.
When delaying your final retirement date by a longer
period of time (e.g. years) it may have a greater
impact.
On the face of things it would perhaps seem quite
trivial to calculate the effect of delaying retirement
by a few years.
However, in practise, these calculations can be quite
complicated and may include many factors and
interactions between them.
Just to give a few examples: maintaining your earned
income for longer not only gives you extra cash in hand
here and now but also means that you will not need to
touch our pension pot which will allow it to grow or
longer.
Delaying retirement will also in all probability mean
that your other pension entitlements will increase – and
obviously the time spent in full retirement will become
shorter (which means more money on an annual basis). By
delaying claiming your State Pension you will also
increase your entitlement for the rest of your life.
In order to account for all these variables and their
respective interactions NFA use socalled computer
simulation models and optimisation engines to identify
the most advantageous retirement date given your
personal circumstances,
It is not the intention to give you a detailed lecture
here on how such simulations are carried out but to put
it simply: all the different relevant parameters are
expressed as equations and the combined effect of all
these equations are calculated for various time periods.
It is important to note that there always are some
uncertainty when such models are used as assumption have
to be made regarding e.g. inflation rates, possible
escalation of the State Pension, other pension payments
etc.
Remember, no model can be better than the data used to
construct it!
The figure below shows a very simplified output for a
hypothetical client containing only a few variables. In
real life such models will often contain 1020 or more
variables. 

Example of a Simplified retirement simulation
Using simulations (or Cash Flow modelling as it can also be
called) in this way can give a reasonable overall idea of the
extra funds need to fill a potential gap.
However, there is a problem when such a gap is to be filled
either in full, or in part, by e.g. investments as they are far
less predictable than e.g. the escalation of the State Pension
etc,
Just making the simple assumption (as many ‘forecasters’ indeed
do!) that investments will follow a historical norm and grow on
average by say, 45% per year, over the next 2030 years, can
lead to very serious miscalculations.
Whilst the assumption of an average growth at this rate may hold
true over longer periods of time it does not help you much if
you incur serious losses already in the first 35 years of your
retirement period. In this case there will be an awful lot of
‘catching up’ to do, and you may not have the necessary time!
In order to make some rational sense out of such uncertainties,
NFA employs what is technically called: Monte Carlo
Simulations.
The mere name may make you apprehensive but you can rest assured
that it has nothing to do with gambling your pension pot on the
roulette!
Monte Carlo Simulations (MCS) are basically used to answer
the question of 'what might happen' if a lot of random events
all turn against me time and time again  or, what is the worst
case scenario.
Technically speaking, a MCS will use all the relevant parameters
related to your situation and simply calculate the end result
for a given date in the future (e.g. in 30 years time) assuming
that everything goes as planned.
Once one calculation has been made, the process will be
repeated, but this time one parameter will be changed at random
(positive, negative or unchanged) and the result computed. Then,
this result will be recalculated on the basis of a random change
in another of the parameters  and so on...
Once 10,000 or more scenarios have been calculated a pattern
starts to emerge.
This is often presented as a histogram and a typical example is
given below:
Result from Monte Carlo simulation
The key feature that is of most interest to NFA is the size of
the column to the very left in the illustration  as this is an
indicator of the risk there is to run out of money.
As it can be seen from the example given above, in this case
there is less than a 5% risk that the client’s pension pot will
fall below £26,000 after 30 years even if everything that can go
wrong does go wrong!
It is of course not much fun if the worst does happen  but
given the rather low risk of this and the fact that there are 30
years in which action can be taken to mitigate this event the
illustrated plan can only be described as quite robust!
It is the hope that the information on this page has shown you
that by using a number of modern computersimulation techniques,
NFA can help you to obtain a quite accurate picture, not only
about your immediate shortterm financial situation but also
about your likely future situation,
No computer model can predict the future but at least they can
provide a foundation upon which sensible and informed decisions
can be made.